Radiomics strategies for risk assessment of tumour failure in head-and-neck cancer

Artificial Intelligence-Driven Radiomics in Head and Neck Cancer: Current Status and Future Prospects

BACKGROUND

Radiomics is a rapidly growing field used to leverage medical radiological images by extracting quantitative features. These are supposed to characterize a patient's phenotype, and when combined with artificial intelligence techniques, to improve the accuracy of diagnostic models and clinical outcome prediction.

OBJECTIVES

This review aims at examining the application areas of artificial intelligence-based radiomics (AI-based radiomics) for the management of head and neck cancer (HNC). It further explores the workflow of AI-based radiomics for personalized and precision oncology in HNC. Finally, it examines the current challenges of AI-based radiomics in daily clinical oncology and offers possible solutions to these challenges.

METHODS

Comprehensive electronic databases (PubMed, Medline via Ovid, Scopus, Web of Science, CINAHL, and Cochrane Library) were searched following the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines. The quality of included studies and their risk of biases were evaluated using the Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD)and Prediction Model Risk of Bias Assessment Tool (PROBAST).

RESULTS

Out of the 659 search hits retrieved, 45 fulfilled the inclusion criteria. Our review revealed that the application of AI-based radiomics model as an ancillary tool for improved decision-making in HNC management includes radiomics-based cancer diagnosis and radiomics-based cancer prognosis. The radiomics-based cancer diagnosis includes tumor staging, tumor grading, and classification of malignant and benign tumors. Similarly, radiomics-based cancer prognosis includes prediction for treatment response, recurrence, metastasis, and survival. In addition, the challenges in the implementation of these models for clinical evaluations include data imbalance, feature engineering (extraction and selection), model generalizability, multi-modal fusion, and model interpretability.

CONCLUSION

Considering the highly subjective and interobserver variability that is peculiar to the interpretation of medical images by expert clinicians, AI-based radiomics seeks to offer potentially useful quantitative information, which is not visible to the human eye or unintentionally often remain ignored during clinical imaging practice. By enabling the extraction of this type of information, AI-based radiomics has the potential to revolutionize HNC oncology, providing a platform for more personalized, higher quality, and cost-effective care for HNC patients.

Radiomics for clinical decision support in radiation oncology

Radiomics is a promising tool for the development of quantitative biomarkers to support clinical decision-making. It has been shown to improve the prediction of response to treatment and outcome in different settings, particularly in the field of radiation oncology by optimising the dose delivery solutions and reducing the rate of radiation-induced side effects, leading to a fully personalised approach. Despite the promising results offered by radiomics at each of these stages, standardised methodologies, reproducibility and interpretability of results are still lacking, limiting the potential clinical impact of these tools. In this review, we briefly describe the principles of radiomics and the most relevant applications of radiomics at each stage of cancer management in the framework of radiation oncology. Furthermore, the integration of radiomics into clinical decision support systems is analysed, defining the challenges and offering possible solutions for translating radiomics into a clinically applicable tool.

Assessing the stability and discriminative ability of radiomics features in the tumor microenvironment: Leveraging peri-tumoral regions in vestibular schwannoma

PURPOSE

The tumor microenvironment (TME) plays a crucial role in tumor progression and treatment response. Radiomics offers a non-invasive approach to studying the TME by extracting quantitative features from medical images. In this study, we present a novel approach to assess the stability and discriminative ability of radiomics features in the TME of vestibular schwannoma (VS).

METHODS

Magnetic Resonance Imaging (MRI) data from 242 VS patients were analyzed, including contrast-enhanced T1-weighted (ceT1) and high-resolution T2-weighted (hrT2) sequences. Radiomics features were extracted from concentric peri-tumoral regions of varying sizes. The intraclass correlation coefficient (ICC) was used to assess feature stability and discriminative ability, establishing quantile thresholds for ICCmin and ICCmax.

RESULTS

The identified thresholds for ICCmin and ICCmax were 0.45 and 0.72, respectively. Features were classified into four categories: stable and discriminative (S-D), stable and non-discriminative (S-ND), unstable and discriminative (US-D), and unstable and non-discriminative (US-ND). Different feature groups exhibited varying proportions of S-D features across ceT1 and hrT2 sequences. The similarity of S-D features between ceT1 and hrT2 sequences was evaluated using Jaccard's index, with a value of 0.78 for all feature groups which is ranging from 0.68 (intensity features) to 1.00 (Neighbouring Gray Tone Difference Matrix (NGTDM) features).

CONCLUSIONS

This study provides a framework for identifying stable and discriminative radiomics features in the TME, which could serve as potential biomarkers or predictors of patient outcomes, ultimately improving the management of VS patients.

Suppress the cell growth of cancer stem-like cells (NTERA-2) using Sox2-Oct4 decoy oligodeoxynucleotide-encapsulated niosomes-zinc hybrid nanocarriers under X-irradiation

Sox2 and Oct4 dysregulations could significantly increase in the cancer stem cell (CSC) population in some cancer cells and resistance to common treatments. In this study, the synergistic effects of Sox2-Oct4 decoy oligodeoxynucleotides-encapsulated Niosomes-zinc hybrid nanocarriers along with X-irradiation conditions as a combinational therapy tool were investigated in the treatment of cancer-like stem cells (NTERA-2). The NTERA-2 cell line known as a cancer-like stem cell line was used in this investigation. Sox2-Oct4 decoy oligodeoxynucleotides were designed based on the sequence of the Sox2 promoter and synthesized. Physicochemical characteristics of ODNs-encapsulated niosomes-zinc hybrid nanocarriers (NISM@BSA-DEC-Zn) investigated with FT-IR, DLS, FESEM, and ODNs release kinetic estimation assays. Further investigations such as hemolysis, uptake, cell viability, apoptosis, cell cycle, and scratch repair tests were performed. All the above assays were completed with and without X-ray exposure conditions (fractionated 2Gy). Physicochemical characteristics results showed that the Niosomes-Zn nanocarriers were successfully synthesized. NISM@BSA-DEC-Zn was efficiently taken up by NTERA-2 cells and significantly inhibited cell growth, increased apoptosis, and reduced cell migration in both conditions (with and without X-ray exposure). Furthermore, NISM@BSA-DEC-Zn treatment resulted in G1 and G2/M cell cycle arrest without and with X-irradiation, respectively. The prepared nanocarrier system can be a promising tool for drug delivery in cancer treatment. Decoy ODN strategy along with zinc nanoparticles could increase the sensitivity of cancer cells toward irradiation, which has the potential for combinational cancer therapies.

Segmentation-Free Outcome Prediction from Head and Neck Cancer PET/CT Images: Deep Learning-Based Feature Extraction from Multi-Angle Maximum Intensity Projections (MA-MIPs)

We introduce an innovative, simple, effective segmentation-free approach for survival analysis of head and neck cancer (HNC) patients from PET/CT images. By harnessing deep learning-based feature extraction techniques and multi-angle maximum intensity projections (MA-MIPs) applied to Fluorodeoxyglucose Positron Emission Tomography (FDG-PET) images, our proposed method eliminates the need for manual segmentations of regions-of-interest (ROIs) such as primary tumors and involved lymph nodes. Instead, a state-of-the-art object detection model is trained utilizing the CT images to perform automatic cropping of the head and neck anatomical area, instead of only the lesions or involved lymph nodes on the PET volumes. A pre-trained deep convolutional neural network backbone is then utilized to extract deep features from MA-MIPs obtained from 72 multi-angel axial rotations of the cropped PET volumes. These deep features extracted from multiple projection views of the PET volumes are then aggregated and fused, and employed to perform recurrence-free survival analysis on a cohort of 489 HNC patients. The proposed approach outperforms the best performing method on the target dataset for the task of recurrence-free survival analysis. By circumventing the manual delineation of the malignancies on the FDG PET-CT images, our approach eliminates the dependency on subjective interpretations and highly enhances the reproducibility of the proposed survival analysis method. The code for this work is publicly released.

Translational risk-adapted approaches to de-escalated radiation for human papillomavirus-positive oropharyngeal cancer: Past, present, and future

Interest in the use of de-escalated radiation to treat patients with newly diagnosed human papillomavirus (HPV)-positive oropharyngeal cancer has grown dramatically with the publication of prospective trials demonstrating the efficacy of such an approach. While the rationale for de-escalation--- namely to decrease treatment-related toxicity while maintaining the excellent rates of disease control historically observed in patients with this disease-is inherently obvious, uncertainty exists regarding how to best select patients for de-escalation. Consequently, risk-adapted strategies using a variety of translational and clinical platforms have been increasingly popularized to better refine treatment. These have integrated contemporary methods of mid-treatment response assessment using advanced technologies and molecular assays to customize the radiation dose. By monitoring the response as patients actively proceed through treatment, risk-adapted protocols have the potential to provide insight into the biological behavior of tumors and make individualized therapy possible. The purpose of this review is to summarize the evidence to date on risk-adapted approaches to de-escalated radiation-- highlighting the clinical, radiological, and biological data which may ultimately help usher the principles of precision medicine into practice for patients with HPV-positive oropharyngeal cancer.

Deep Learning-Guided Dosimetry for Mitigating Local Failure of Patients With Non-Small Cell Lung Cancer Receiving Stereotactic Body Radiation Therapy

PURPOSE

Non-small cell lung cancer (NSCLC) stereotactic body radiation therapy with 50 Gy/5 fractions is sometimes considered controversial, as the nominal biologically effective dose (BED) of 100 Gy is felt by some to be insufficient for long-term local control of some lesions. In this study, we analyzed such patients using explainable deep learning techniques and consequently proposed appropriate treatment planning criteria. These novel criteria could help planners achieve optimized treatment plans for maximal local control.

METHODS AND MATERIALS

A total of 535 patients treated with 50 Gy/5 fractions were used to develop a novel deep learning local response model. A multimodality approach, incorporating computed tomography images, 3-dimensional dose distribution, and patient demographics, combined with a discrete-time survival model, was applied to predict time to failure and the probability of local control. Subsequently, an integrated gradient-weighted class activation mapping method was used to identify the most significant dose-volume metrics predictive of local failure and their optimal cut-points.

RESULTS

The model was cross-validated, showing an acceptable performance (c-index: 0.72, 95% CI, 0.68-0.75); the testing c-index was 0.69. The model's spatial attention was concentrated mostly in the tumors' periphery (planning target volume [PTV] - internal gross target volume [IGTV]) region. Statistically significant dose-volume metrics in improved local control were BED Dnear-min ≥ 103.8 Gy in IGTV (hazard ratio [HR], 0.31; 95% CI, 015-0.63), V104 ≥ 98% in IGTV (HR, 0.30; 95% CI, 0.15-0.60), gEUD ≥ 103.8 Gy in PTV-IGTV (HR, 0.25; 95% CI, 0.12-0.50), and Dmean ≥ 104.5 Gy in PTV-IGTV (HR, 0.25; 95% CI, 0.12-0.51).

CONCLUSIONS

Deep learning-identified dose-volume metrics have shown significant prognostic power (log-rank, P = .003) and could be used as additional actionable criteria for treatment planning in NSCLC stereotactic body radiation therapy patients receiving 50 Gy in 5 fractions. Although our data do not confirm or refute that a significantly higher BED for the prescription dose is necessary for tumor control in NSCLC, it might be clinically effective to escalate the nominal prescribed dose from BED 100 to 105 Gy.

Multicenter PET image harmonization using generative adversarial networks

PURPOSE

To improve reproducibility and predictive performance of PET radiomic features in multicentric studies by cycle-consistent generative adversarial network (GAN) harmonization approaches.

METHODS

GAN-harmonization was developed to harmonize whole-body PET scans to perform image style and texture translation between different centers and scanners. GAN-harmonization was evaluated by application to two retrospectively collected open datasets and different tasks. First, GAN-harmonization was performed on a dual-center lung cancer cohort (127 female, 138 male) where the reproducibility of radiomic features in healthy liver tissue was evaluated. Second, GAN-harmonization was applied to a head and neck cancer cohort (43 female, 154 male) acquired from three centers. Here, the clinical impact of GAN-harmonization was analyzed by predicting the development of distant metastases using a logistic regression model incorporating first-order statistics and texture features from baseline 18F-FDG PET before and after harmonization.

RESULTS

Image quality remained high (structural similarity: left kidney ≥ 0.800, right kidney ≥ 0.806, liver ≥ 0.780, lung ≥ 0.838, spleen ≥ 0.793, whole-body ≥ 0.832) after image harmonization across all utilized datasets. Using GAN-harmonization, inter-site reproducibility of radiomic features in healthy liver tissue increased at least by ≥ 5 ± 14% (first-order), ≥ 16 ± 7% (GLCM), ≥ 19 ± 5% (GLRLM), ≥ 16 ± 8% (GLSZM), ≥ 17 ± 6% (GLDM), and ≥ 23 ± 14% (NGTDM). In the head and neck cancer cohort, the outcome prediction improved from AUC 0.68 (95% CI 0.66-0.71) to AUC 0.73 (0.71-0.75) by application of GAN-harmonization.

CONCLUSIONS

GANs are capable of performing image harmonization and increase reproducibility and predictive performance of radiomic features derived from different centers and scanners.

Artificial Intelligence in Head and Neck Surgery

This article explores artificial intelligence's (AI's) role in otolaryngology for head and neck cancer diagnosis and management. It highlights AI's potential in pattern recognition for early cancer detection, prognostication, and treatment planning, primarily through image analysis using clinical, endoscopic, and histopathologic images. Radiomics is also discussed at length, as well as the many ways that radiologic image analysis can be utilized, including for diagnosis, lymph node metastasis prediction, and evaluation of treatment response. The study highlights AI's promise and limitations, underlining the need for clinician-data scientist collaboration to enhance head and neck cancer care.

Development and Validation of Prognostic Models Using Radiomic Features from Pre-Treatment Positron Emission Tomography (PET) Images in Head and Neck Squamous Cell Carcinoma (HNSCC) Patients

High-dimensional radiomics features derived from pre-treatment positron emission tomography (PET) images offer prognostic insights for patients with head and neck squamous cell carcinoma (HNSCC). Using 124 PET radiomics features and clinical variables (age, sex, stage of cancer, site of cancer) from a cohort of 232 patients, we evaluated four survival models-penalized Cox model, random forest, gradient boosted model and support vector machine-to predict all-cause mortality (ACM), locoregional recurrence/residual disease (LR) and distant metastasis (DM) probability during 36, 24 and 24 months of follow-up, respectively. We developed models with five-fold cross-validation, selected the best-performing model for each outcome based on the concordance index (C-statistic) and the integrated Brier score (IBS) and validated them in an independent cohort of 102 patients. The penalized Cox model demonstrated better performance for ACM (C-statistic = 0.70, IBS = 0.12) and DM (C-statistic = 0.70, IBS = 0.08) while the random forest model displayed better performance for LR (C-statistic = 0.76, IBS = 0.07). We conclude that the ML-based prognostic model can aid clinicians in quantifying prognosis and determining effective treatment strategies, thereby improving favorable outcomes in HNSCC patients.

Multi-organ segmentation: a progressive exploration of learning paradigms under scarce annotation

Precise delineation of multiple organs or abnormal regions in the human body from medical images plays an essential role in computer-aided diagnosis, surgical simulation, image-guided interventions, and especially in radiotherapy treatment planning. Thus, it is of great significance to explore automatic segmentation approaches, among which deep learning-based approaches have evolved rapidly and witnessed remarkable progress in multi-organ segmentation. However, obtaining an appropriately sized and fine-grained annotated dataset of multiple organs is extremely hard and expensive. Such scarce annotation limits the development of high-performance multi-organ segmentation models but promotes many annotation-efficient learning paradigms. Among these, studies on transfer learning leveraging external datasets, semi-supervised learning including unannotated datasets and partially-supervised learning integrating partially-labeled datasets have led the dominant way to break such dilemmas in multi-organ segmentation. We first review the fully supervised method, then present a comprehensive and systematic elaboration of the 3 abovementioned learning paradigms in the context of multi-organ segmentation from both technical and methodological perspectives, and finally summarize their challenges and future trends.

Prognostic Value of a Combined Nomogram Model Integrating 3-Dimensional Deep Learning and Radiomics for Head and Neck Cancer

OBJECTIVE

The preoperative prediction of the overall survival (OS) status of patients with head and neck cancer (HNC) is significant value for their individualized treatment and prognosis. This study aims to evaluate the impact of adding 3D deep learning features to radiomics models for predicting 5-year OS status.

METHODS

Two hundred twenty cases from The Cancer Imaging Archive public dataset were included in this study; 2212 radiomics features and 304 deep features were extracted from each case. The features were selected by univariate analysis and the least absolute shrinkage and selection operator, and then grouped into a radiomics model containing Positron Emission Tomography /Computed Tomography (PET/CT) radiomics features score, a deep model containing deep features score, and a combined model containing PET/CT radiomics features score +3D deep features score. TumorStage model was also constructed using initial patient tumor node metastasis stage to compare the performance of the combined model. A nomogram was constructed to analyze the influence of deep features on the performance of the model. The 10-fold cross-validation of the average area under the receiver operating characteristic curve and calibration curve were used to evaluate performance, and Shapley Additive exPlanations (SHAP) was developed for interpretation.

RESULTS

The TumorStage model, radiomics model, deep model, and the combined model achieved areas under the receiver operating characteristic curve of 0.604, 0.851, 0.840, and 0.895 on the train set and 0.571, 0.849, 0.832, and 0.900 on the test set. The combined model showed better performance of predicting the 5-year OS status of HNC patients than the radiomics model and deep model. The combined model was shown to provide a favorable fit in calibration curves and be clinically useful in decision curve analysis. SHAP summary plot and SHAP The SHAP summary plot and SHAP force plot visually interpreted the influence of deep features and radiomics features on the model results.

CONCLUSIONS

In predicting 5-year OS status in patients with HNC, 3D deep features could provide richer features for combined model, which showed outperformance compared with the radiomics model and deep model.

Impact of 18F-FDG PET Intensity Normalization on Radiomic Features of Oropharyngeal Squamous Cell Carcinomas and Machine Learning-Generated Biomarkers

We aimed to investigate the effects of 18F-FDG PET voxel intensity normalization on radiomic features of oropharyngeal squamous cell carcinoma (OPSCC) and machine learning-generated radiomic biomarkers. Methods: We extracted 1,037 18F-FDG PET radiomic features quantifying the shape, intensity, and texture of 430 OPSCC primary tumors. The reproducibility of individual features across 3 intensity-normalized images (body-weight SUV, reference tissue activity ratio to lentiform nucleus of brain and cerebellum) and the raw PET data was assessed using an intraclass correlation coefficient (ICC). We investigated the effects of intensity normalization on the features' utility in predicting the human papillomavirus (HPV) status of OPSCCs in univariate logistic regression, receiver-operating-characteristic analysis, and extreme-gradient-boosting (XGBoost) machine-learning classifiers. Results: Of 1,037 features, a high (ICC ≥ 0.90), medium (0.90 > ICC ≥ 0.75), and low (ICC < 0.75) degree of reproducibility across normalization methods was attained in 356 (34.3%), 608 (58.6%), and 73 (7%) features, respectively. In univariate analysis, features from the PET normalized to the lentiform nucleus had the strongest association with HPV status, with 865 of 1,037 (83.4%) significant features after multiple testing corrections and a median area under the receiver-operating-characteristic curve (AUC) of 0.65 (interquartile range, 0.62-0.68). Similar tendencies were observed in XGBoost models, with the lentiform nucleus-normalized model achieving the numerically highest average AUC of 0.72 (SD, 0.07) in the cross validation within the training cohort. The model generalized well to the validation cohorts, attaining an AUC of 0.73 (95% CI, 0.60-0.85) in independent validation and 0.76 (95% CI, 0.58-0.95) in external validation. The AUCs of the XGBoost models were not significantly different. Conclusion: Only one third of the features demonstrated a high degree of reproducibility across intensity-normalization techniques, making uniform normalization a prerequisite for interindividual comparability of radiomic markers. The choice of normalization technique may affect the radiomic features' predictive value with respect to HPV. Our results show trends that normalization to the lentiform nucleus may improve model performance, although more evidence is needed to draw a firm conclusion.

Utility of CT Radiomics and Delta Radiomics for Survival Evaluation in Locally Advanced Nasopharyngeal Carcinoma with Concurrent Chemoradiotherapy

BACKGROUND

A high incidence rate of nasopharyngeal carcinoma (NPC) has been observed in Southeast Asia compared to other parts of the world. Radiomics is a computational tool to predict outcomes and may be used as a prognostic biomarker for advanced NPC treated with concurrent chemoradiotherapy. Recently, radiomic analysis of the peripheral tumor microenvironment (TME), which is the region surrounding the gross tumor volume (GTV), has shown prognostic usefulness. In this study, not only was gross tumor volume (GTVt) analyzed but also tumor peripheral regions (GTVp) were explored in terms of the TME concept. Both radiomic features and delta radiomic features were analyzed using CT images acquired in a routine radiotherapy process.

METHODS

A total of 50 patients with NPC stages III, IVA, and IVB were enrolled between September 2004 and February 2014. Survival models were built using Cox regression with clinical factors (i.e., gender, age, overall stage, T stage, N stage, and treatment dose) and radiomic features. Radiomic features were extracted from GTVt and GTVp. GTVp was created surrounding GTVt for TME consideration. Furthermore, delta radiomics, which is the longitudinal change in quantitative radiomic features, was utilized for analysis. Finally, C-index values were computed using leave-one-out cross-validation (LOOCV) to evaluate the performances of all prognosis models.

RESULTS

Models were built for three different clinical outcomes, including overall survival (OS), local recurrence-free survival (LRFS), and progression-free survival (PFS). The range of the C-index in clinical factor models was (0.622, 0.729). All radiomics models, including delta radiomics models, were in the range of (0.718, 0.872). Among delta radiomics models, GTVt and GTVp were in the range of (0.833, 0.872) and (0.799, 0.834), respectively.

CONCLUSIONS

Radiomic analysis on the proximal region surrounding the gross tumor volume of advanced NPC patients for survival outcome evaluation was investigated, and preliminary positive results were obtained. Radiomic models and delta radiomic models demonstrated performance that was either superior to or comparable with that of conventional clinical models.

Radiomic biomarkers of locoregional recurrence: prognostic insights from oral cavity squamous cell carcinoma preoperative CT scans

Introduction

This study aimed to identify CT-based imaging biomarkers for locoregional recurrence (LR) in Oral Cavity Squamous Cell Carcinoma (OSCC) patients.

Methods

Computed tomography scans were collected from 78 patients with OSCC who underwent surgical treatment at a single medical center. We extracted 1,092 radiomic features from gross tumor volume in each patient's pre-treatment CT. Clinical characteristics were also obtained, including race, sex, age, tobacco and alcohol use, tumor staging, and treatment modality. A feature selection algorithm was used to eliminate the most redundant features, followed by a selection of the best subset of the Logistic regression model (LRM). The best LRM model was determined based on the best prediction accuracy in terms of the area under Receiver operating characteristic curve. Finally, significant radiomic features in the final LRM model were identified as imaging biomarkers.

Results and discussion

Two radiomics biomarkers, Large Dependence Emphasis (LDE) of the Gray Level Dependence Matrix (GLDM) and Long Run Emphasis (LRE) of the Gray Level Run Length Matrix (GLRLM) of the 3D Laplacian of Gaussian (LoG σ=3), have demonstrated the capability to preoperatively distinguish patients with and without LR, exhibiting exceptional testing specificity (1.00) and sensitivity (0.82). The group with LRE > 2.99 showed a 3-year recurrence-free survival rate of 0.81, in contrast to 0.49 for the group with LRE ≤ 2.99. Similarly, the group with LDE > 120 showed a rate of 0.82, compared to 0.49 for the group with LDE ≤ 120. These biomarkers broaden our understanding of using radiomics to predict OSCC progression, enabling personalized treatment plans to enhance patient survival.

Comparison of semi-automatic and manual segmentation methods for tumor delineation on head and neck squamous cell carcinoma (HNSCC) positron emission tomography (PET) images

Objective. Accurate and reproducible tumor delineation on positron emission tomography (PET) images is required to validate predictive and prognostic models based on PET radiomic features. Manual segmentation of tumors is time-consuming whereas semi-automatic methods are easily implementable and inexpensive. This study assessed the reliability of semi-automatic segmentation methods over manual segmentation for tumor delineation in head and neck squamous cell carcinoma (HNSCC) PET images.Approach. We employed manual and six semi-automatic segmentation methods (just enough interaction (JEI), watershed, grow from seeds (GfS), flood filling (FF), 30% SUVmax and 40%SUVmax threshold) using 3D slicer software to extract 128 radiomic features from FDG-PET images of 100 HNSCC patients independently by three operators. We assessed the distributional properties of all features and considered 92 log-transformed features for subsequent analysis. For each paired comparison of a feature, we fitted a separate linear mixed effect model using the method (two levels; manual versus one semi-automatic method) as a fixed effect and the subject and the operator as the random effects. We estimated different statistics-the intraclass correlation coefficient agreement (aICC), limits of agreement (LoA), total deviation index (TDI), coverage probability (CP) and coefficient of individual agreement (CIA)-to evaluate the agreement between the manual and semi-automatic methods.Main results. Accounting for all statistics across 92 features, the JEI method consistently demonstrated acceptable agreement with the manual method, with median values of aICC = 0.86, TDI = 0.94, CP = 0.66, and CIA = 0.91.Significance. This study demonstrated that JEI method is a reliable semi-automatic method for tumor delineation on HNSCC PET images.

Delta Radiomic Features Predict Resection Margin Status and Overall Survival in Neoadjuvant-Treated Pancreatic Cancer Patients

BACKGROUND

Neoadjuvant therapy (NAT) emerged as the standard of care for patients with pancreatic ductal adenocarcinoma (PDAC) who undergo surgery; however, surgery is morbid, and tools to predict resection margin status (RMS) and prognosis in the preoperative setting are needed. Radiomic models, specifically delta radiomic features (DRFs), may provide insight into treatment dynamics to improve preoperative predictions.

METHODS

We retrospectively collected clinical, pathological, and surgical data (patients with resectable, borderline, locally advanced, and metastatic disease), and pre/post-NAT contrast-enhanced computed tomography (CT) scans from PDAC patients at the University of Texas Southwestern Medical Center (UTSW; discovery) and Humanitas Hospital (validation cohort). Gross tumor volume was contoured from CT scans, and 257 radiomics features were extracted. DRFs were calculated by direct subtraction of pre/post-NAT radiomic features. Cox proportional models and binary prediction models, including/excluding clinical variables, were constructed to predict overall survival (OS), disease-free survival (DFS), and RMS.

RESULTS

The discovery and validation cohorts comprised 58 and 31 patients, respectively. Both cohorts had similar clinical characteristics, apart from differences in NAT (FOLFIRINOX vs. gemcitabine/nab-paclitaxel; p < 0.05) and type of surgery resections (pancreatoduodenectomy, distal or total pancreatectomy; p < 0.05). The model that combined clinical variables (pre-NAT carbohydrate antigen (CA) 19-9, the change in CA19-9 after NAT (∆CA19-9), and resectability status) and DRFs outperformed the clinical feature-based models and other radiomics feature-based models in predicting OS (UTSW: 0.73; Humanitas: 0.66), DFS (UTSW: 0.75; Humanitas: 0.64), and RMS (UTSW 0.73; Humanitas: 0.69).

CONCLUSIONS

Our externally validated, predictive/prognostic delta-radiomics models, which incorporate clinical variables, show promise in predicting the risk of predicting RMS in NAT-treated PDAC patients and their OS or DFS.

RADCURE: An open-source head and neck cancer CT dataset for clinical radiation therapy insights

PURPOSE

This manuscript presents RADCURE, one of the most extensive head and neck cancer (HNC) imaging datasets accessible to the public. Initially collected for clinical radiation therapy (RT) treatment planning, this dataset has been retrospectively reconstructed for use in imaging research.

ACQUISITION AND VALIDATION METHODS

RADCURE encompasses data from 3346 patients, featuring computed tomography (CT) RT simulation images with corresponding target and organ-at-risk contours. These CT scans were collected using systems from three different manufacturers. Standard clinical imaging protocols were followed, and contours were manually generated and reviewed at weekly RT quality assurance rounds. RADCURE imaging and structure set data was extracted from our institution's radiation treatment planning and oncology information systems using a custom-built data mining and processing system. Furthermore, images were linked to our clinical anthology of outcomes data for each patient and includes demographic, clinical and treatment information based on the 7th edition TNM staging system (Tumor-Node-Metastasis Classification System of Malignant Tumors). The median patient age is 63, with the final dataset including 80% males. Half of the cohort is diagnosed with oropharyngeal cancer, while laryngeal, nasopharyngeal, and hypopharyngeal cancers account for 25%, 12%, and 5% of cases, respectively. The median duration of follow-up is five years, with 60% of the cohort surviving until the last follow-up point.

DATA FORMAT AND USAGE NOTES

The dataset provides images and contours in DICOM CT and RT-STRUCT formats, respectively. We have standardized the nomenclature for individual contours-such as the gross primary tumor, gross nodal volumes, and 19 organs-at-risk-to enhance the RT-STRUCT files' utility. Accompanying demographic, clinical, and treatment data are supplied in a comma-separated values (CSV) file format. This comprehensive dataset is publicly accessible via The Cancer Imaging Archive.

POTENTIAL APPLICATIONS

RADCURE's amalgamation of imaging, clinical, demographic, and treatment data renders it an invaluable resource for a broad spectrum of radiomics image analysis research endeavors. Researchers can utilize this dataset to advance routine clinical procedures using machine learning or artificial intelligence, to identify new non-invasive biomarkers, or to forge prognostic models.

Shallow and deep learning classifiers in medical image analysis

An increasingly strong connection between artificial intelligence and medicine has enabled the development of predictive models capable of supporting physicians' decision-making. Artificial intelligence encompasses much more than machine learning, which nevertheless is its most cited and used sub-branch in the last decade. Since most clinical problems can be modeled through machine learning classifiers, it is essential to discuss their main elements. This review aims to give primary educational insights on the most accessible and widely employed classifiers in radiology field, distinguishing between "shallow" learning (i.e., traditional machine learning) algorithms, including support vector machines, random forest and XGBoost, and "deep" learning architectures including convolutional neural networks and vision transformers. In addition, the paper outlines the key steps for classifiers training and highlights the differences between the most common algorithms and architectures. Although the choice of an algorithm depends on the task and dataset dealing with, general guidelines for classifier selection are proposed in relation to task analysis, dataset size, explainability requirements, and available computing resources. Considering the enormous interest in these innovative models and architectures, the problem of machine learning algorithms interpretability is finally discussed, providing a future perspective on trustworthy artificial intelligence.Relevance statement The growing synergy between artificial intelligence and medicine fosters predictive models aiding physicians. Machine learning classifiers, from shallow learning to deep learning, are offering crucial insights for the development of clinical decision support systems in healthcare. Explainability is a key feature of models that leads systems toward integration into clinical practice. Key points • Training a shallow classifier requires extracting disease-related features from region of interests (e.g., radiomics).• Deep classifiers implement automatic feature extraction and classification.• The classifier selection is based on data and computational resources availability, task, and explanation needs.

Segmentation of Variants of Nuclei on Whole Slide Images by Using Radiomic Features

The histopathological segmentation of nuclear types is a challenging task because nuclei exhibit distinct morphologies, textures, and staining characteristics. Accurate segmentation is critical because it affects the diagnostic workflow for patient assessment. In this study, a framework was proposed for segmenting various types of nuclei from different organs of the body. The proposed framework improved the segmentation performance for each nuclear type using radiomics. First, we used distinct radiomic features to extract and analyze quantitative information about each type of nucleus and subsequently trained various classifiers based on the best input sub-features of each radiomic feature selected by a LASSO operator. Second, we inputted the outputs of the best classifier to various segmentation models to learn the variants of nuclei. Using the MoNuSAC2020 dataset, we achieved state-of-the-art segmentation performance for each category of nuclei type despite the complexity, overlapping, and obscure regions. The generalized adaptability of the proposed framework was verified by the consistent performance obtained in whole slide images of different organs of the body and radiomic features.

Quantitative US Delta Radiomics to Predict Radiation Response in Individuals with Head and Neck Squamous Cell Carcinoma

Purpose To investigate the role of quantitative US (QUS) radiomics data obtained after the 1st week of radiation therapy (RT) in predicting treatment response in individuals with head and neck squamous cell carcinoma (HNSCC). Materials and Methods This prospective study included 55 participants (21 with complete response [median age, 65 years {IQR: 47-80 years}, 20 male, one female; and 34 with incomplete response [median age, 59 years {IQR: 39-79 years}, 33 male, one female) with bulky node-positive HNSCC treated with curative-intent RT from January 2015 to October 2019. All participants received 70 Gy of radiation in 33-35 fractions over 6-7 weeks. US radiofrequency data from metastatic lymph nodes were acquired prior to and after 1 week of RT. QUS analysis resulted in five spectral maps from which mean values were extracted. We applied a gray-level co-occurrence matrix technique for textural analysis, leading to 20 QUS texture and 80 texture-derivative parameters. The response 3 months after RT was used as the end point. Model building and evaluation utilized nested leave-one-out cross-validation. Results Five delta (Δ) parameters had statistically significant differences (P < .05). The support vector machines classifier achieved a sensitivity of 71% (15 of 21), a specificity of 76% (26 of 34), a balanced accuracy of 74%, and an area under the receiver operating characteristic curve of 0.77 on the test set. For all the classifiers, the performance improved after the 1st week of treatment. Conclusion A QUS Δ-radiomics model using data obtained after the 1st week of RT from individuals with HNSCC predicted response 3 months after treatment completion with reasonable accuracy. Keywords: Computer-Aided Diagnosis (CAD), Ultrasound, Radiation Therapy/Oncology, Head/Neck, Radiomics, Quantitative US, Radiotherapy, Head and Neck Squamous Cell Carcinoma, Machine Learning Clinicaltrials.gov registration no. NCT03908684 Supplemental material is available for this article. © RSNA, 2024.

Does Dual-Energy Computed Tomography Material Decomposition Improve Radiomics Capability to Predict Survival in Head and Neck Squamous Cell Carcinoma Patients? A Preliminary Investigation

OBJECTIVE

Our study objective was to explore the additional value of dual-energy CT (DECT) material decomposition for squamous cell carcinoma of the head and neck (SCCHN) survival prognostication.

METHODS

A group of 50 SCCHN patients (male, 37; female, 13; mean age, 63.6 ± 10.82 years) with baseline head and neck DECT between September 2014 and August 2020 were retrospectively included. Primary tumors were segmented, radiomics features were extracted, and DECT material decomposition was performed. We used independent train and validation datasets with cross-validation and 100 independent iterations to identify prognostic signatures applying elastic net (EN) and random survival forest (RSF). Features were ranked and intercorrelated according to their prognostic importance. We benchmarked the models against clinical parameters. Intraclass correlation coefficients were used to analyze the interreader variation.

RESULTS

The exclusively radiomics-trained models achieved similar ( P = 0.947) prognostic performance of area under the curve (AUC) = 0.784 (95% confidence interval [CI], 0.775-0.812) (EN) and AUC = 0.785 (95% CI, 0.759-0.812) (RSF). The additional application of DECT material decomposition did not improve the model's performance (EN, P = 0.594; RSF, P = 0.198). In the clinical benchmark, the top averaged AUC value of 0.643 (95% CI, 0.611-0.675) was inferior to the quantitative imaging-biomarker models ( P < 0.001). A combined imaging and clinical model did not improve the imaging-based models ( P > 0.101). Shape features revealed high prognostic importance.

CONCLUSIONS

Radiomics AI applications may be used for SCCHN survival prognostication, but the spectral information of DECT material decomposition did not improve the model's performance in our preliminary investigation.

Radiomics analysis for the prediction of locoregional recurrence of locally advanced oropharyngeal cancer and hypopharyngeal cancer

PURPOSE

By radiomic analysis of the postcontrast CT images, this study aimed to predict locoregional recurrence (LR) of locally advanced oropharyngeal cancer (OPC) and hypopharyngeal cancer (HPC).

METHODS

A total of 192 patients with stage III-IV OPC or HPC from two independent cohort were randomly split into a training cohort with 153 cases and a testing cohort with 39 cases. Only primary tumor mass was manually segmented. Radiomic features were extracted using PyRadiomics, and then the support vector machine was used to build the radiomic model with fivefold cross-validation process in the training data set. For each case, a radiomics score was generated to indicate the probability of LR.

RESULTS

There were 94 patients with LR assigned in the progression group and 98 patients without LR assigned in the stable group. There was no significant difference of TNM staging, treatment strategies and common risk factors between these two groups. For the training data set, the radiomics model to predict LR showed 83.7% accuracy and 0.832 (95% CI 0.72, 0.87) area under the ROC curve (AUC). For the test data set, the accuracy and AUC slightly declined to 79.5% and 0.770 (95% CI 0.64, 0.80), respectively. The sensitivity/specificity of training and test data set for LR prediction were 77.6%/89.6%, and 66.7%/90.5%, respectively.

CONCLUSIONS

The image-based radiomic approach could provide a reliable LR prediction model in locally advanced OPC and HPC. Early identification of those prone to post-treatment recurrence would be helpful for appropriate adjustments to treatment strategies and post-treatment surveillance.

Reproducibility in Radiomics: A Comparison of Feature Extraction Methods and Two Independent Datasets

Radiomics involves the extraction of information from medical images that are not visible to the human eye. There is evidence that these features can be used for treatment stratification and outcome prediction. However, there is much discussion about the reproducibility of results between different studies. This paper studies the reproducibility of CT texture features used in radiomics, comparing two feature extraction implementations, namely the MATLAB toolkit and Pyradiomics, when applied to independent datasets of CT scans of patients: (i) the open access RIDER dataset containing a set of repeat CT scans taken 15 min apart for 31 patients (RIDER Scan 1 and Scan 2, respectively) treated for lung cancer; and (ii) the open access HN1 dataset containing 137 patients treated for head and neck cancer. Gross tumor volume (GTV), manually outlined by an experienced observer available on both datasets, was used. The 43 common radiomics features available in MATLAB and Pyradiomics were calculated using two intensity-level quantization methods with and without an intensity threshold. Cases were ranked for each feature for all combinations of quantization parameters, and the Spearman's rank coefficient, rs, calculated. Reproducibility was defined when a highly correlated feature in the RIDER dataset also correlated highly in the HN1 dataset, and vice versa. A total of 29 out of the 43 reported stable features were found to be highly reproducible between MATLAB and Pyradiomics implementations, having a consistently high correlation in rank ordering for RIDER Scan 1 and RIDER Scan 2 (rs > 0.8). 18/43 reported features were common in the RIDER and HN1 datasets, suggesting they may be agnostic to disease site. Useful radiomics features should be selected based on reproducibility. This study identified a set of features that meet this requirement and validated the methodology for evaluating reproducibility between datasets.

Robustness of [18F]FDG PET/CT radiomic analysis in the setting of drug-induced cardiotoxicity

BACKGROUND AND OBJECTIVES

Standardization of radiomic data acquisition protocols is still at a very early stage, revealing a strong need to work towards the definition of uniform image processing methodologies The aim of this study is to identify sources of variability in radiomic data derived from image discretization and resampling methodologies prior to image feature extraction. Furthermore, to identify robust potential image-based biomarkers for the early detection of cardiotoxicity.

METHODS

Image post-acquisition processing, interpolation, and volume of interest (VOI) segmentation were performed. Four experiments were conducted to assess the reliability in terms of the intraclass correlation coefficient (ICC) of the radiomic features and the effects of the variation of voxel size and gray level discretization. Statistical analysis was performed separating the patients according to cardiotoxicity diagnosis. Differences of texture features were studied with Mann-Whitney U test. P-values <0.05 after multiple testing correction were considered statistically significant. Additionally, a non-supervised k-Means clustering algorithm was evaluated.

RESULTS

The effect of the variation in the voxel size demonstrated a non-dependency relationship with the values of the radiomic features, regardless of the chosen discretization method. The median ICC values were 0.306 and 0.872 for absolute agreement and consistency, respectively, when varying the discretization bin number. The median ICC values were 0.678 and 0.878 for absolute agreement and consistency, respectively, when varying the discretization bin size. A total of 16 first order, 6 Gray Level Co-occurrence Matrix (GLCM), 4 Gray Level Dependence Matrix (GLDM) and 4 Gray Level Run Length Matrix (GLRLM) features demonstrated statistically significant differences between the diagnosis groups for interim scans (P<0.05) for the fixed bin size (FBS) discretization methodology. However, no statistically significant differences between diagnostic groups were found for the fixed bin number (FBN) discretization methodology. Two clusters based on the radiomic features were identified.

CONCLUSIONS

Gray level discretization has a major impact on the repeatability of the radiomic features. The selection of the optimal processing methodology has led to the identification of texture-based patterns for the differentiation of early cardiac damage profiles.

Semisupervised 3D segmentation of pancreatic tumors in positron emission tomography/computed tomography images using a mutual information minimization and cross-fusion strategy

Background

Accurate segmentation of pancreatic cancer tumors using positron emission tomography/computed tomography (PET/CT) multimodal images is crucial for clinical diagnosis and prognosis evaluation. However, deep learning methods for automated medical image segmentation require a substantial amount of manually labeled data, making it time-consuming and labor-intensive. Moreover, addition or simple stitching of multimodal images leads to redundant information, failing to fully exploit the complementary information of multimodal images. Therefore, we developed a semisupervised multimodal network that leverages limited labeled samples and introduces a cross-fusion and mutual information minimization (MIM) strategy for PET/CT 3D segmentation of pancreatic tumors.

Methods

Our approach combined a cross multimodal fusion (CMF) module with a cross-attention mechanism. The complementary multimodal features were fused to form a multifeature set to enhance the effectiveness of feature extraction while preserving specific features of each modal image. In addition, we designed an MIM module to mitigate redundant high-level modal information and compute the latent loss of PET and CT. Finally, our method employed the uncertainty-aware mean teacher semi-supervised framework to segment regions of interest from PET/CT images using a small amount of labeled data and a large amount of unlabeled data.

Results

We evaluated our combined MIM and CMF semisupervised segmentation network (MIM-CMFNet) on a private dataset of pancreatic cancer, yielding an average Dice coefficient of 73.14%, an average Jaccard index score of 60.56%, and an average 95% Hausdorff distance (95HD) of 6.30 mm. In addition, to verify the broad applicability of our method, we used a public dataset of head and neck cancer, yielding an average Dice coefficient of 68.71%, an average Jaccard index score of 57.72%, and an average 95HD of 7.88 mm.

Conclusions

The experimental results demonstrate the superiority of our MIM-CMFNet over existing semisupervised techniques. Our approach can achieve a performance similar to that of fully supervised segmentation methods while significantly reducing the data annotation cost by 80%, suggesting it is highly practicable for clinical application.

Development and validation of survival prognostic models for head and neck cancer patients using machine learning and dosiomics and CT radiomics features: a multicentric study

BACKGROUND

This study aimed to investigate the value of clinical, radiomic features extracted from gross tumor volumes (GTVs) delineated on CT images, dose distributions (Dosiomics), and fusion of CT and dose distributions to predict outcomes in head and neck cancer (HNC) patients.

METHODS

A cohort of 240 HNC patients from five different centers was obtained from The Cancer Imaging Archive. Seven strategies, including four non-fusion (Clinical, CT, Dose, DualCT-Dose), and three fusion algorithms (latent low-rank representation referred (LLRR),Wavelet, weighted least square (WLS)) were applied. The fusion algorithms were used to fuse the pre-treatment CT images and 3-dimensional dose maps. Overall, 215 radiomics and Dosiomics features were extracted from the GTVs, alongside with seven clinical features incorporated. Five feature selection (FS) methods in combination with six machine learning (ML) models were implemented. The performance of the models was quantified using the concordance index (CI) in one-center-leave-out 5-fold cross-validation for overall survival (OS) prediction considering the time-to-event.

RESULTS

The mean CI and Kaplan-Meier curves were used for further comparisons. The CoxBoost ML model using the Minimal Depth (MD) FS method and the glmnet model using the Variable hunting (VH) FS method showed the best performance with CI = 0.73 ± 0.15 for features extracted from LLRR fused images. In addition, both glmnet-Cindex and Coxph-Cindex classifiers achieved a CI of 0.72 ± 0.14 by employing the dose images (+ incorporated clinical features) only.

CONCLUSION

Our results demonstrated that clinical features, Dosiomics and fusion of dose and CT images by specific ML-FS models could predict the overall survival of HNC patients with acceptable accuracy. Besides, the performance of ML methods among the three different strategies was almost comparable.

Radiomic Biomarkers of Locoregional Recurrence: Prognostic Insights from Oral Cavity Squamous Cell Carcinoma preoperative CT scans

This study aimed to identify CT-based imaging biomarkers for locoregional recurrence (LR) in Oral Cavity Squamous Cell Carcinoma (OSCC) patients. Our study involved a retrospective review of 78 patients with OSCC who underwent surgical treatment at a single medical center. An approach involving feature selection and statistical model diagnostics was utilized to identify biomarkers. Two radiomics biomarkers, Large Dependence Emphasis (LDE) of the Gray Level Dependence Matrix (GLDM) and Long Run Emphasis (LRE) of the Gray Level Run Length Matrix (GLRLM) of the 3D Laplacian of Gaussian (LoG σ = 3), have demonstrated the capability to preoperatively distinguish patients with and without LR, exhibiting exceptional testing specificity (1.00) and sensitivity (0.82). The group with LRE > 2.99 showed a 3-year recurrence-free survival rate of 0.81, in contrast to 0.49 for the group with LRE ≤ 2.99. Similarly, the group with LDE > 120 showed a rate of 0.82, compared to 0.49 for the group with LDE ≤ 120. These biomarkers broaden our understanding of using radiomics to predict OSCC progression, enabling personalized treatment plans to enhance patient survival.

Patterns of cortical oxygenation may predict the response to stenting in subjects with renal artery stenosis: A radiomics-based model

BACKGROUND

Percutaneous-transluminal renal angioplasty (PTRA) and stenting aim to halt the progression of kidney disease in patients with renal artery stenosis (RAS), but its outcome is often suboptimal. We hypothesized that a model incorporating markers of renal function and oxygenation extracted using radiomics analysis of blood oxygenation-level dependent (BOLD)-MRI images may predict renal response to PTRA in swine RAS.

MATERIALS AND METHODS

Twenty domestic pigs with RAS were scanned with CT and BOLD MRI before and 4 weeks after PTRA. Stenotic (STK) and contralateral (CLK) kidney volume, blood flow (RBF), and glomerular filtration rate (GFR) were determined, and BOLD-MRI R2 * maps were generated before and after administration of furosemide, a tubular reabsorption inhibitor. Radiomics features were extracted from pre-PTRA BOLD maps and Robust features were determined by Intraclass correlation coefficients (ICC). Prognostic models were developed to predict post-PTRA renal function based on the baseline functional and BOLD-radiomics features, using Lasso-regression for training, and testing with resampling.

RESULTS

Twenty-six radiomics features passed the robustness test. STK oxygenation distribution pattern did not respond to furosemide, whereas in the CLK radiomics features sensitive to oxygenation heterogeneity declined. Radiomics-based model predictions of post-PTRA GFR (r = 0.58, p = 0.007) and RBF (r = 0.68; p = 0.001) correlated with actual measurements with sensitivity and specificity of 92% and 67%, respectively. Models were unsuccessful in predicting post-PTRA systemic measures of renal function.

CONCLUSIONS

Several radiomics features are sensitive to cortical oxygenation patterns and permit estimation of post-PTRA renal function, thereby distinguishing subjects likely to respond to PTRA and stenting.

Deep-learning models for image-based gynecological cancer diagnosis: a systematic review and meta- analysis

Introduction

Gynecological cancers pose a significant threat to women worldwide, especially those in resource-limited settings. Human analysis of images remains the primary method of diagnosis, but it can be inconsistent and inaccurate. Deep learning (DL) can potentially enhance image-based diagnosis by providing objective and accurate results. This systematic review and meta-analysis aimed to summarize the recent advances of deep learning (DL) techniques for gynecological cancer diagnosis using various images and explore their future implications.

Methods

The study followed the PRISMA-2 guidelines, and the protocol was registered in PROSPERO. Five databases were searched for articles published from January 2018 to December 2022. Articles that focused on five types of gynecological cancer and used DL for diagnosis were selected. Two reviewers assessed the articles for eligibility and quality using the QUADAS-2 tool. Data was extracted from each study, and the performance of DL techniques for gynecological cancer classification was estimated by pooling and transforming sensitivity and specificity values using a random-effects model.

Results

The review included 48 studies, and the meta-analysis included 24 studies. The studies used different images and models to diagnose different gynecological cancers. The most popular models were ResNet, VGGNet, and UNet. DL algorithms showed more sensitivity but less specificity compared to machine learning (ML) methods. The AUC of the summary receiver operating characteristic plot was higher for DL algorithms than for ML methods. Of the 48 studies included, 41 were at low risk of bias.

Conclusion

This review highlights the potential of DL in improving the screening and diagnosis of gynecological cancer, particularly in resource-limited settings. However, the high heterogeneity and quality of the studies could affect the validity of the results. Further research is necessary to validate the findings of this study and to explore the potential of DL in improving gynecological cancer diagnosis.

Prognostic Value of Radiomic Analysis Using Pre- and Post-Treatment 18F-FDG-PET/CT in Patients with Laryngeal Cancer and Hypopharyngeal Cancer

BACKGROUND

The prognostic value of conducting 18F-FDG PET/CT imaging has yielded different results in patients with laryngeal cancer and hypopharyngeal cancer, but these results are controversial, and there is a lack of dedicated studies on each type of cancer. This study aimed to evaluate whether combining radiomic analysis of pre- and post-treatment 18F-FDG PET/CT imaging features and clinical parameters has additional prognostic value in patients with laryngeal cancer and hypopharyngeal cancer.

METHODS

From 2008 to 2016, data on patients diagnosed with cancer of the larynx and hypopharynx were retrospectively collected. The patients underwent pre- and post-treatment 18F-FDG PET/CT imaging. The values of ΔPre-Post PET were measured from the texture features. Least absolute shrinkage and selection operator (LASSO) Cox regression was used to select the most predictive features to formulate a Rad-score for both progression-free survival (PFS) and overall survival (OS). Kaplan-Meier curve analysis and Cox regression were employed to assess PFS and OS. Then, the concordance index (C-index) and calibration plot were used to evaluate the performance of the radiomics nomogram.

RESULTS

Study data were collected for a total of 91 patients. The mean follow-up period was 71.5 mo. (8.4-147.3). The Rad-score was formulated based on the texture parameters and was significantly associated with both PFS (p = 0.024) and OS (p = 0.009). When predicting PFS, only the Rad-score demonstrated a significant association (HR 2.1509, 95% CI [1.100-4.207], p = 0.025). On the other hand, age (HR 1.116, 95% CI [1.041-1.197], p = 0.002) and Rad-score (HR 33.885, 95% CI [2.891-397.175], p = 0.005) exhibited associations with OS. The Rad-score value showed good discrimination when it was combined with clinical parameters in both PFS (C-index 0.802-0.889) and OS (C-index 0.860-0.958). The calibration plots also showed a good agreement between the observed and predicted survival probabilities.

CONCLUSIONS

Combining clinical parameters with radiomics analysis of pre- and post-treatment 18F-FDG PET/CT parameters in patients with laryngeal cancer and hypopharyngeal cancer might have additional prognostic value.

Information fusion for fully automated segmentation of head and neck tumors from PET and CT images

BACKGROUND

PET/CT images combining anatomic and metabolic data provide complementary information that can improve clinical task performance. PET image segmentation algorithms exploiting the multi-modal information available are still lacking.

PURPOSE

Our study aimed to assess the performance of PET and CT image fusion for gross tumor volume (GTV) segmentations of head and neck cancers (HNCs) utilizing conventional, deep learning (DL), and output-level voting-based fusions.

METHODS

The current study is based on a total of 328 histologically confirmed HNCs from six different centers. The images were automatically cropped to a 200 × 200 head and neck region box, and CT and PET images were normalized for further processing. Eighteen conventional image-level fusions were implemented. In addition, a modified U2-Net architecture as DL fusion model baseline was used. Three different input, layer, and decision-level information fusions were used. Simultaneous truth and performance level estimation (STAPLE) and majority voting to merge different segmentation outputs (from PET and image-level and network-level fusions), that is, output-level information fusion (voting-based fusions) were employed. Different networks were trained in a 2D manner with a batch size of 64. Twenty percent of the dataset with stratification concerning the centers (20% in each center) were used for final result reporting. Different standard segmentation metrics and conventional PET metrics, such as SUV, were calculated.

RESULTS

In single modalities, PET had a reasonable performance with a Dice score of 0.77 ± 0.09, while CT did not perform acceptably and reached a Dice score of only 0.38 ± 0.22. Conventional fusion algorithms obtained a Dice score range of [0.76-0.81] with guided-filter-based context enhancement (GFCE) at the low-end, and anisotropic diffusion and Karhunen-Loeve transform fusion (ADF), multi-resolution singular value decomposition (MSVD), and multi-level image decomposition based on latent low-rank representation (MDLatLRR) at the high-end. All DL fusion models achieved Dice scores of 0.80. Output-level voting-based models outperformed all other models, achieving superior results with a Dice score of 0.84 for Majority_ImgFus, Majority_All, and Majority_Fast. A mean error of almost zero was achieved for all fusions using SUVpeak , SUVmean and SUVmedian .

CONCLUSION

PET/CT information fusion adds significant value to segmentation tasks, considerably outperforming PET-only and CT-only methods. In addition, both conventional image-level and DL fusions achieve competitive results. Meanwhile, output-level voting-based fusion using majority voting of several algorithms results in statistically significant improvements in the segmentation of HNC.

PAHPhysRAD: A Digital Imaging and Communications in Medicine Research Tool for Segmentation and Radiomic Feature Extraction

Introduction

Segmentation and analysis of organs at risks (OARs) and tumor volumes are integral concepts in the development of radiotherapy treatment plans and prediction of patients' treatment outcomes.

Aims

To develop a research tool, PAHPhysRAD, that can be used to semi- and fully automate segmentation of OARs. In addition, the proposed software seeks to extract 3214 radiomic features from tumor volumes and user-specified dose-volume parameters.

Materials and Methods

Developed within MATLAB, PAHPhysRAD provides a comprehensive suite of segmentation tools, including manual, semi-automatic, and automatic options. For semi-autosegmentation, meta AI's Segment Anything Model was incorporated using the bounding box methods. Autosegmentation of OARs and tumor volume are implemented through a module that enables the addition of models in Open Neural Network Exchange format. To validate the radiomic feature extraction module in PAHPhysRAD, radiomic features extracted from gross tumor volume of 15 non-small cell lung carcinoma patients were compared against the features extracted from 3D Slicer™. The dose-volume parameters extraction module was validated using the dose volume data extracted from 28 tangential field-based breast treatment planning datasets. The volume receiving ≥20 Gy (V20) for ipsilateral lung and the mean doses received by the heart and ipsilateral lung, were compared against the parameters extracted from Eclipse.

Results

The Wilcoxon signed-rank test revealed no significant difference between the majority of the radiomic features derived from PAHPhysRAD and 3D Slicer. The average mean lung and heart doses calculated in Eclipse were 5.51 ± 2.28 Gy and 1.64 ± 1.98 Gy, respectively. Similarly, the average mean lung and heart doses calculated in PAHPhysRAD were 5.45 ± 2.89 Gy and 1.67 ± 2.08 Gy, respectively.

Conclusion

The MATLAB-based graphical user interface, PAHPhysRAD, offers a user-friendly platform for viewing and analyzing medical scans with options to extract radiomic features and dose-volume parameters. Its versatility, compatibility, and potential for further development make it an asset in medical image analysis.

A statistical deformation model-based data augmentation method for volumetric medical image segmentation

The accurate delineation of organs-at-risk (OARs) is a crucial step in treatment planning during radiotherapy, as it minimizes the potential adverse effects of radiation on surrounding healthy organs. However, manual contouring of OARs in computed tomography (CT) images is labor-intensive and susceptible to errors, particularly for low-contrast soft tissue. Deep learning-based artificial intelligence algorithms surpass traditional methods but require large datasets. Obtaining annotated medical images is both time-consuming and expensive, hindering the collection of extensive training sets. To enhance the performance of medical image segmentation, augmentation strategies such as rotation and Gaussian smoothing are employed during preprocessing. However, these conventional data augmentation techniques cannot generate more realistic deformations, limiting improvements in accuracy. To address this issue, this study introduces a statistical deformation model-based data augmentation method for volumetric medical image segmentation. By applying diverse and realistic data augmentation to CT images from a limited patient cohort, our method significantly improves the fully automated segmentation of OARs across various body parts. We evaluate our framework on three datasets containing tumor OARs from the head, neck, chest, and abdomen. Test results demonstrate that the proposed method achieves state-of-the-art performance in numerous OARs segmentation challenges. This innovative approach holds considerable potential as a powerful tool for various medical imaging-related sub-fields, effectively addressing the challenge of limited data access.

Multi-Modal Ensemble Deep Learning in Head and Neck Cancer HPV Sub-Typing

Oropharyngeal Squamous Cell Carcinoma (OPSCC) is one of the common forms of heterogeneity in head and neck cancer. Infection with human papillomavirus (HPV) has been identified as a major risk factor for OPSCC. Therefore, differentiating the HPV-positive and negative cases in OPSCC patients is an essential diagnostic factor influencing future treatment decisions. In this study, we investigated the accuracy of a deep learning-based method for image interpretation and automatically detected the HPV status of OPSCC in routinely acquired Computed Tomography (CT) and Positron Emission Tomography (PET) images. We introduce a 3D CNN-based multi-modal feature fusion architecture for HPV status prediction in primary tumor lesions. The architecture is composed of an ensemble of CNN networks and merges image features in a softmax classification layer. The pipeline separately learns the intensity, contrast variation, shape, texture heterogeneity, and metabolic assessment from CT and PET tumor volume regions and fuses those multi-modal features for final HPV status classification. The precision, recall, and AUC scores of the proposed method are computed, and the results are compared with other existing models. The experimental results demonstrate that the multi-modal ensemble model with soft voting outperformed single-modality PET/CT, with an AUC of 0.76 and F1 score of 0.746 on publicly available TCGA and MAASTRO datasets. In the MAASTRO dataset, our model achieved an AUC score of 0.74 over primary tumor volumes of interest (VOIs). In the future, more extensive cohort validation may suffice for better diagnostic accuracy and provide preliminary assessment before the biopsy.

Clinical performance of a multiparametric MRI-based post concussive syndrome index

Introduction

Diffusion Tensor Imaging (DTI) has revealed measurable changes in the brains of patients with persistent post-concussive syndrome (PCS). Because of inconsistent results in univariate DTI metrics among patients with mild traumatic brain injury (mTBI), there is currently no single objective and reliable MRI index for clinical decision-making in patients with PCS.

Purpose

This study aimed to evaluate the performance of a newly developed PCS Index (PCSI) derived from machine learning of multiparametric magnetic resonance imaging (MRI) data to classify and differentiate subjects with mTBI and PCS history from those without a history of mTBI.

Materials and methods

Data were retrospectively extracted from 139 patients aged between 18 and 60 years with PCS who underwent MRI examinations at 2 weeks to 1-year post-mTBI, as well as from 336 subjects without a history of head trauma. The performance of the PCS Index was assessed by comparing 69 patients with a clinical diagnosis of PCS with 264 control subjects. The PCSI values for patients with PCS were compared based on the mechanism of injury, time interval from injury to MRI examination, sex, history of prior concussion, loss of consciousness, and reported symptoms.

Results

Injured patients had a mean PCSI value of 0.57, compared to the control group, which had a mean PCSI value of 0.12 (p = 8.42e-23) with accuracy of 88%, sensitivity of 64%, and specificity of 95%, respectively. No statistically significant differences were found in the PCSI values when comparing the mechanism of injury, sex, or loss of consciousness.

Conclusion

The PCSI for individuals aged between 18 and 60 years was able to accurately identify patients with post-concussive injuries from 2 weeks to 1-year post-mTBI and differentiate them from the controls. The results of this study suggest that multiparametric MRI-based PCSI has great potential as an objective clinical tool to support the diagnosis, treatment, and follow-up care of patients with post-concussive syndrome. Further research is required to investigate the replicability of this method using other types of clinical MRI scanners.

Tensor radiomics: paradigm for systematic incorporation of multi-flavoured radiomics features

Background

Radiomics features hold significant value as quantitative imaging biomarkers for diagnosis, prognosis, and treatment response assessment. To generate radiomics features and ultimately develop signatures, various factors can be manipulated, including image discretization parameters (e.g., bin number or size), convolutional filters, segmentation perturbation, or multi-modality fusion levels. Typically, only one set of parameters is employed, resulting in a single value or "flavour" for each radiomics feature. In contrast, we propose "tensor radiomics" (TR) where tensors of features calculated using multiple parameter combinations (i.e., flavours) are utilized to optimize the creation of radiomics signatures.

Methods

We provide illustrative instances of TR implementation in positron emission tomography-computed tomography (PET-CT), magnetic resonance imaging (MRI), and CT by leveraging machine learning (ML) and deep learning (DL) methodologies, as well as reproducibility analyses: (I) to predict overall survival (OS) in lung cancer (CT) and head and neck cancer (PET-CT), TR was employed by varying bin sizes. This approach involved use of a hybrid deep neural network called 'TR-Net' and two ML-based techniques for combining different flavours. (II) TR was constructed by incorporating different segmentation perturbations and various bin sizes to classify the response of late-stage lung cancer to first-line immunotherapy using CT images. (III) In MRI of glioblastoma (GBM), TR was implemented to generate multi-flavour radiomics features, enabling enhanced analysis and interpretation. (IV) TR was employed via multiple PET-CT fusions in head and neck cancer. Flavours based on different fusions were created using Laplacian pyramids and wavelet transforms.

Results

Our findings demonstrated that TR outperformed conventional radiomics features in lung cancer CT and head and neck cancer PET-CT images, significantly enhancing OS prediction accuracy. TR also improved classification of lung cancer response to therapy and exhibited notable advantages in reproducibility compared to single-flavour features in MR imaging of GBM. Moreover, in head and neck cancer, TR through multiple PET-CT fusions exhibited improved performance in predicting OS.

Conclusions

We conclude that the proposed TR paradigm has significant potential to improve performance in different medical imaging tasks. By incorporating multiple flavours of radiomics features, TR overcomes limitations associated with individual features and shows promise in enhancing prognostic capabilities in clinical settings.

Automatic Head and Neck Tumor segmentation and outcome prediction relying on FDG-PET/CT images: Findings from the second edition of the HECKTOR challenge

By focusing on metabolic and morphological tissue properties respectively, FluoroDeoxyGlucose (FDG)-Positron Emission Tomography (PET) and Computed Tomography (CT) modalities include complementary and synergistic information for cancerous lesion delineation and characterization (e.g. for outcome prediction), in addition to usual clinical variables. This is especially true in Head and Neck Cancer (HNC). The goal of the HEad and neCK TumOR segmentation and outcome prediction (HECKTOR) challenge was to develop and compare modern image analysis methods to best extract and leverage this information automatically. We present here the post-analysis of HECKTOR 2nd edition, at the 24th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2021. The scope of the challenge was substantially expanded compared to the first edition, by providing a larger population (adding patients from a new clinical center) and proposing an additional task to the challengers, namely the prediction of Progression-Free Survival (PFS). To this end, the participants were given access to a training set of 224 cases from 5 different centers, each with a pre-treatment FDG-PET/CT scan and clinical variables. Their methods were subsequently evaluated on a held-out test set of 101 cases from two centers. For the segmentation task (Task 1), the ranking was based on a Borda counting of their ranks according to two metrics: mean Dice Similarity Coefficient (DSC) and median Hausdorff Distance at 95th percentile (HD95). For the PFS prediction task, challengers could use the tumor contours provided by experts (Task 3) or rely on their own (Task 2). The ranking was obtained according to the Concordance index (C-index) calculated on the predicted risk scores. A total of 103 teams registered for the challenge, for a total of 448 submissions and 29 papers. The best method in the segmentation task obtained an average DSC of 0.759, and the best predictions of PFS obtained a C-index of 0.717 (without relying on the provided contours) and 0.698 (using the expert contours). An interesting finding was that best PFS predictions were reached by relying on DL approaches (with or without explicit tumor segmentation, 4 out of the 5 best ranked) compared to standard radiomics methods using handcrafted features extracted from delineated tumors, and by exploiting alternative tumor contours (automated and/or larger volumes encompassing surrounding tissues) rather than relying on the expert contours. This second edition of the challenge confirmed the promising performance of fully automated primary tumor delineation in PET/CT images of HNC patients, although there is still a margin for improvement in some difficult cases. For the first time, the prediction of outcome was also addressed and the best methods reached relatively good performance (C-index above 0.7). Both results constitute another step forward toward large-scale outcome prediction studies in HNC.

A Radiomics Approach to Identify Immunologically Active Tumor in Patients with Head and Neck Squamous Cell Carcinomas

BACKGROUND

We recently developed a gene-expression-based HOT score to identify the hot/cold phenotype of head and neck squamous cell carcinomas (HNSCCs), which is associated with the response to immunotherapy. Our goal was to determine whether radiomic profiling from computed tomography (CT) scans can distinguish hot and cold HNSCC.

METHOD

We included 113 patients from The Cancer Genome Atlas (TCGA) and 20 patients from the Groupe Hospitalier Pitié-Salpêtrière (GHPS) with HNSCC, all with available pre-treatment CT scans. The hot/cold phenotype was computed for all patients using the HOT score. The IBEX software (version 4.11.9, accessed on 30 march 2020) was used to extract radiomic features from the delineated tumor region in both datasets, and the intraclass correlation coefficient (ICC) was computed to select robust features. Machine learning classifier models were trained and tested in the TCGA dataset and validated using the area under the receiver operator characteristic curve (AUC) in the GHPS cohort.

RESULTS

A total of 144 radiomic features with an ICC >0.9 was selected. An XGBoost model including these selected features showed the best performance prediction of the hot/cold phenotype with AUC = 0.86 in the GHPS validation dataset.

CONCLUSIONS AND RELEVANCE

We identified a relevant radiomic model to capture the overall hot/cold phenotype of HNSCC. This non-invasive approach could help with the identification of patients with HNSCC who may benefit from immunotherapy.

Comprehensive evaluation of similarity between synthetic and real CT images for nasopharyngeal carcinoma

BACKGROUND

Although magnetic resonance imaging (MRI)-to-computed tomography (CT) synthesis studies based on deep learning have significantly progressed, the similarity between synthetic CT (sCT) and real CT (rCT) has only been evaluated in image quality metrics (IQMs). To evaluate the similarity between synthetic CT (sCT) and real CT (rCT) comprehensively, we comprehensively evaluated IQMs and radiomic features for the first time.

METHODS

This study enrolled 127 patients with nasopharyngeal carcinoma who underwent CT and MRI scans. Supervised-learning (Unet) and unsupervised-learning (CycleGAN) methods were applied to build MRI-to-CT synthesis models. The regions of interest (ROIs) included nasopharynx gross tumor volume (GTVnx), brainstem, parotid glands, and temporal lobes. The peak signal-to-noise ratio (PSNR), mean absolute error (MAE), root mean square error (RMSE), and structural similarity (SSIM) were used to evaluate image quality. Additionally, 837 radiomic features were extracted for each ROI, and the correlation was evaluated using the concordance correlation coefficient (CCC).

RESULTS

The MAE, RMSE, SSIM, and PSNR of the body were 91.99, 187.12, 0.97, and 51.15 for Unet and 108.30, 211.63, 0.96, and 49.84 for CycleGAN. For the metrics, Unet was superior to CycleGAN (P < 0.05). For the radiomic features, the percentage of four levels (i.e., excellent, good, moderate, and poor, respectively) were as follows: GTVnx, 8.5%, 14.6%, 26.5%, and 50.4% for Unet and 12.3%, 25%, 38.4%, and 24.4% for CycleGAN; other ROIs, 5.44% ± 3.27%, 5.56% ± 2.92%, 21.38% ± 6.91%, and 67.58% ± 8.96% for Unet and 5.16% ± 1.69%, 3.5% ± 1.52%, 12.68% ± 7.51%, and 78.62% ± 8.57% for CycleGAN.

CONCLUSIONS

Unet-sCT was superior to CycleGAN-sCT for the IQMs. However, neither exhibited absolute superiority in radiomic features, and both were far less similar to rCT. Therefore, further work is required to improve the radiomic similarity for MRI-to-CT synthesis.

TRIAL REGISTRATION

This study was a retrospective study, so it was free from registration.

A common [18F]-FDG PET radiomic signature to predict survival in patients with HPV-induced cancers

Locally advanced cervical cancer (LACC) and anal and oropharyngeal squamous cell carcinoma (ASCC and OPSCC) are mostly caused by oncogenic human papillomaviruses (HPV). In this paper, we developed machine learning (ML) models based on clinical, biological, and radiomic features extracted from pre-treatment fluorine-18-fluorodeoxyglucose positron emission tomography ([18F]-FDG PET) images to predict the survival of patients with HPV-induced cancers. For this purpose, cohorts from five institutions were used: two cohorts of patients treated for LACC including 104 patients from Gustave Roussy Campus Cancer (Center 1) and 90 patients from Leeds Teaching Hospitals NHS Trust (Center 2), two datasets of patients treated for ASCC composed of 66 patients from Institut du Cancer de Montpellier (Center 3) and 67 patients from Oslo University Hospital (Center 4), and one dataset of 45 OPSCC patients from the University Hospital of Zurich (Center 5). Radiomic features were extracted from baseline [18F]-FDG PET images. The ComBat technique was applied to mitigate intra-scanner variability. A modified consensus nested cross-validation for feature selection and hyperparameter tuning was applied on four ML models to predict progression-free survival (PFS) and overall survival (OS) using harmonized imaging features and/or clinical and biological variables as inputs. Each model was trained and optimized on Center 1 and Center 3 cohorts and tested on Center 2, Center 4, and Center 5 cohorts. The radiomic-based CoxNet model achieved C-index values of 0.75 and 0.78 for PFS and 0.76, 0.74, and 0.75 for OS on the test sets. Radiomic feature-based models had superior performance compared to the bioclinical ones, and combining radiomic and bioclinical variables did not improve the performances. Metabolic tumor volume (MTV)-based models obtained lower C-index values for a majority of the tested configurations but quite equivalent performance in terms of time-dependent AUCs (td-AUC). The results demonstrate the possibility of identifying common PET-based image signatures for predicting the response of patients with induced HPV pathology, validated on multi-center multiconstructor data.

Contour subregion error detection methodology using deep learning auto-segmentation

BACKGROUND

Inaccurate manual organ delineation is one of the high-risk failure modes in radiation treatment. Numerous automated contour quality assurance (QA) systems have been developed to assess contour acceptability; however, manual inspection of flagged cases is a time-consuming and challenging process, and can lead to users overlooking the exact error location.

PURPOSE

Our aim is to develop and validate a contour QA system that can effectively detect and visualize subregional contour errors, both qualitatively and quantitatively.

METHODS/MATERIALS

A novel contour subregion error detection (CSED) system was developed using subregional surface distance discrepancies between manual and deep learning auto-segmentation (DLAS) contours. A validation study was conducted using a head and neck public dataset containing 339 cases and evaluated according to knowledge-based pass criteria derived from a clinical training dataset of 60 cases. A blind qualitative evaluation was conducted, comparing the results from the CSED system with manual labels. Subsequently, the CSED-flagged cases were re-examined by a radiation oncologist.

RESULTS

The CSED system could visualize the diverse types of subregional contour errors qualitatively and quantitatively. In the validation dataset, the CSED system resulted in true positive rates (TPR) of 0.814, 0.800, and 0.771; false positive rates (FPR) of 0.310, 0.267, and 0.298; and accuracies of 0.735, 0.759, and 0.730, for brainstem and left and right parotid contours, respectively. The CSED-assisted manual review caught 13 brainstem, 19 left parotid, and 21 right parotid contour errors missed by conventional human review. The TPR/FPR/accuracy of the CSED-assisted manual review improved to 0.836/0.253/0.784, 0.831/0.171/0.830, and 0.808/0.193/0.807 for each structure, respectively. Further, the time savings achieved through CSED-assisted review improved by 75%, with the time for review taking 24.81 ± 12.84, 26.75 ± 10.41, and 28.71 ± 13.72 s for each structure, respectively.

CONCLUSIONS

The CSED system enables qualitative and quantitative detection, localization, and visualization of manual segmentation subregional errors utilizing DLAS contours as references. The use of this system has been shown to help reduce the risk of high-risk failure modes resulting from inaccurate organ segmentation.

Risk Stratification for Oropharyngeal Squamous Cell Carcinoma Using Texture Analysis on CT - A Step Beyond HPV Status

Background and Purpose: Human papillomavirus-associated oropharyngeal squamous cell carcinoma (OPSCC) is increasingly prevalent. Despite the overall more favorable outcome, the observed heterogeneous treatment response within this patient group highlights the need for additional means to prognosticate and guide clinical decision-making. Promising prediction models using radiomics from primary OPSCC have been derived. However, no model/s using metastatic lymphadenopathy exist to allow prognostication in those instances when the primary tumor is not seen. The aim of our study was to evaluate whether radiomics using metastatic lymphadenopathy allows for the development of a useful risk assessment model comparable to the primary tumor and whether additional knowledge of the HPV status further improves its prognostic efficacy. Materials and Methods: 80 consecutive patients diagnosed with stage III-IV OPSCC between February 2009 and October 2015, known human papillomavirus status, and pre-treatment CT images were retrospectively identified. Manual segmentation of primary tumor and metastatic lymphadenopathy was performed and the extracted texture features were used to develop multivariate assessment models to prognosticate treatment response. Results: Texture analysis of either the primary or metastatic lymphadenopathy from pre-treatment enhanced CT images can be used to develop models for the stratification of treatment outcomes in OPSCC patients. AUCs range from .78 to .85 for the various OPSCC groups tested, indicating high predictive capability of the models. Conclusions: This preliminary study can form the basis multi-centre trial that may help optimize treatment and improve quality of life in patients with OPSCC in the era of personalized medicine.

Multitask Learning with Convolutional Neural Networks and Vision Transformers Can Improve Outcome Prediction for Head and Neck Cancer Patients

Neural-network-based outcome predictions may enable further treatment personalization of patients with head and neck cancer. The development of neural networks can prove challenging when a limited number of cases is available. Therefore, we investigated whether multitask learning strategies, implemented through the simultaneous optimization of two distinct outcome objectives (multi-outcome) and combined with a tumor segmentation task, can lead to improved performance of convolutional neural networks (CNNs) and vision transformers (ViTs). Model training was conducted on two distinct multicenter datasets for the endpoints loco-regional control (LRC) and progression-free survival (PFS), respectively. The first dataset consisted of pre-treatment computed tomography (CT) imaging for 290 patients and the second dataset contained combined positron emission tomography (PET)/CT data of 224 patients. Discriminative performance was assessed by the concordance index (C-index). Risk stratification was evaluated using log-rank tests. Across both datasets, CNN and ViT model ensembles achieved similar results. Multitask approaches showed favorable performance in most investigations. Multi-outcome CNN models trained with segmentation loss were identified as the optimal strategy across cohorts. On the PET/CT dataset, an ensemble of multi-outcome CNNs trained with segmentation loss achieved the best discrimination (C-index: 0.29, 95% confidence interval (CI): 0.22-0.36) and successfully stratified patients into groups with low and high risk of disease progression (p=0.003). On the CT dataset, ensembles of multi-outcome CNNs and of single-outcome ViTs trained with segmentation loss performed best (C-index: 0.26 and 0.26, CI: 0.18-0.34 and 0.18-0.35, respectively), both with significant risk stratification for LRC in independent validation (p=0.002 and p=0.011). Further validation of the developed multitask-learning models is planned based on a prospective validation study, which has recently completed recruitment.

Multi-institutional PET/CT image segmentation using federated deep transformer learning

BACKGROUND AND OBJECTIVE

Generalizable and trustworthy deep learning models for PET/CT image segmentation necessitates large diverse multi-institutional datasets. However, legal, ethical, and patient privacy issues challenge sharing of datasets between different centers. To overcome these challenges, we developed a federated learning (FL) framework for multi-institutional PET/CT image segmentation.

METHODS

A dataset consisting of 328 FL (HN) cancer patients who underwent clinical PET/CT examinations gathered from six different centers was enrolled. A pure transformer network was implemented as fully core segmentation algorithms using dual channel PET/CT images. We evaluated different frameworks (single center-based, centralized baseline, as well as seven different FL algorithms) using 68 PET/CT images (20% of each center data). In particular, the implemented FL algorithms include clipping with the quantile estimator (ClQu), zeroing with the quantile estimator (ZeQu), federated averaging (FedAvg), lossy compression (LoCo), robust aggregation (RoAg), secure aggregation (SeAg), and Gaussian differentially private FedAvg with adaptive quantile clipping (GDP-AQuCl).

RESULTS

The Dice coefficient was 0.80±0.11 for both centralized and SeAg FL algorithms. All FL approaches achieved centralized learning model performance with no statistically significant differences. Among the FL algorithms, SeAg and GDP-AQuCl performed better than the other techniques. However, there was no statistically significant difference. All algorithms, except the center-based approach, resulted in relative errors less than 5% for SUVmax and SUVmean for all FL and centralized methods. Centralized and FL algorithms significantly outperformed the single center-based baseline.

CONCLUSIONS

The developed FL-based (with centralized method performance) algorithms exhibited promising performance for HN tumor segmentation from PET/CT images.

Comparison of computed tomography image features extracted by radiomics, self-supervised learning and end-to-end deep learning for outcome prediction of oropharyngeal cancer

Background and purpose

To compare the prediction performance of image features of computed tomography (CT) images extracted by radiomics, self-supervised learning and end-to-end deep learning for local control (LC), regional control (RC), locoregional control (LRC), distant metastasis-free survival (DMFS), tumor-specific survival (TSS), overall survival (OS) and disease-free survival (DFS) of oropharyngeal squamous cell carcinoma (OPSCC) patients after (chemo)radiotherapy.

Methods and materials

The OPC-Radiomics dataset was used for model development and independent internal testing and the UMCG-OPC set for external testing. Image features were extracted from the Gross Tumor Volume contours of the primary tumor (GTVt) regions in CT scans when using radiomics or a self-supervised learning-based method (autoencoder). Clinical and combined (radiomics, autoencoder or end-to-end) models were built using multivariable Cox proportional-hazard analysis with clinical features only and both clinical and image features for LC, RC, LRC, DMFS, TSS, OS and DFS prediction, respectively.

Results

In the internal test set, combined autoencoder models performed better than clinical models and combined radiomics models for LC, RC, LRC, DMFS, TSS and DFS prediction (largest improvements in C-index: 0.91 vs. 0.76 in RC and 0.74 vs. 0.60 in DMFS). In the external test set, combined radiomics models performed better than clinical and combined autoencoder models for all endpoints (largest improvements in LC, 0.82 vs. 0.71). Furthermore, combined models performed better in risk stratification than clinical models and showed good calibration for most endpoints.

Conclusions

Image features extracted using self-supervised learning showed best internal prediction performance while radiomics features have better external generalizability.

Haralick texture analysis for microdosimetry: characterization of Monte Carlo generated 3D specific energy distributions

Objective.Explore the application of Haralick textural analysis to 3D distributions of specific energy (energy imparted per unit mass) scored in cell-scale targets considering varying mean specific energy (absorbed dose), target volume, and incident spectrum.Approach.Monte Carlo simulations are used to generate specific energy distributions in cell-scale water voxels ((1μm)3-(15μm)3) irradiated by photon sources (mean energies: 0.02-2 MeV) to varying mean specific energies (10-400 mGy). Five Haralick features (homogeneity, contrast, entropy, correlation, local homogeneity) are calculated using an implementation of Haralick analysis designed to reduce sensitivity to grey level quantization and are interpreted using fundamental radiation physics.Main results.Haralick measures quantify differences in 3D specific energy distributions observed with varying voxel volume, absorbed dose magnitude, and source spectrum. For example, specific energy distributions in small (1-3μm) voxels with low magnitudes of absorbed dose (10 mGy) have relatively high measures of homogeneity and local homogeneity and relatively low measures of contrast and entropy (all relative to measures for larger voxels), reflecting the many voxels with zero specific energy in an otherwise sporadic distribution. With increasing target size, energy is shared across more target voxels, and trends in Haralick measures, such as decreasing homogeneity and increasing contrast and entropy, reflect characteristics of each 3D specific energy distribution. Specific energy distributions for sources of differing mean energy are characterized by Haralick measures, e.g. contrast generally decreases with increasing source energy, correlation and homogeneity are often (not always) higher for higher energy sources.Significance.Haralick texture analysis successfully quantifies spatial trends in 3D specific energy distributions characteristic of radiation source, target size, and absorbed dose magnitude, thus offering new avenues to quantify microdosimetric data beyond first order histogram features. Promising future directions include investigations of multiscale tissue models, targeted radiation therapy techniques, and biological response to radiation.

From coarse to fine: a deep 3D probability volume contours framework for tumour segmentation and dose painting in PET images

With the increasing integration of functional imaging techniques like Positron Emission Tomography (PET) into radiotherapy (RT) practices, a paradigm shift in cancer treatment methodologies is underway. A fundamental step in RT planning is the accurate segmentation of tumours based on clinical diagnosis. Furthermore, novel tumour control methods, such as intensity modulated radiation therapy (IMRT) dose painting, demand the precise delineation of multiple intensity value contours to ensure optimal tumour dose distribution. Recently, convolutional neural networks (CNNs) have made significant strides in 3D image segmentation tasks, most of which present the output map at a voxel-wise level. However, because of information loss in subsequent downsampling layers, they frequently fail to precisely identify precise object boundaries. Moreover, in the context of dose painting strategies, there is an imperative need for reliable and precise image segmentation techniques to delineate high recurrence-risk contours. To address these challenges, we introduce a 3D coarse-to-fine framework, integrating a CNN with a kernel smoothing-based probability volume contour approach (KsPC). This integrated approach generates contour-based segmentation volumes, mimicking expert-level precision and providing accurate probability contours crucial for optimizing dose painting/IMRT strategies. Our final model, named KsPC-Net, leverages a CNN backbone to automatically learn parameters in the kernel smoothing process, thereby obviating the need for user-supplied tuning parameters. The 3D KsPC-Net exploits the strength of KsPC to simultaneously identify object boundaries and generate corresponding probability volume contours, which can be trained within an end-to-end framework. The proposed model has demonstrated promising performance, surpassing state-of-the-art models when tested against the MICCAI 2021 challenge dataset (HECKTOR).

Long-term reproducibility of opportunistically assessed vertebral bone mineral density and texture features in routine clinical multi-detector computed tomography using an automated segmentation framework

Background

To investigate reproducibility of texture features and volumetric bone mineral density (vBMD) extracted from trabecular bone in the thoracolumbar spine in routine clinical multi-detector computed tomography (MDCT) data in a single scanner environment.

Methods

Patients who underwent two routine clinical thoraco-abdominal MDCT exams at a single scanner with a time interval of 6 to 26 months (n=203, 131 males; time interval mean, 13 months; median, 12 months) were included in this observational study. Exclusion criteria were metabolic and hematological disorders, bone metastases, use of bone-active medications, and history of osteoporotic vertebral fractures (VFs) or prior diagnosis of osteoporosis. A convolutional neural network (CNN)-based framework was used for automated spine labeling and segmentation (T5-L5), asynchronous Hounsfield unit (HU)-to-BMD calibration, and correction for the intravenous contrast medium phase. Vertebral vBMD and six texture features [varianceglobal, entropy, short-run emphasis (SRE), long-run emphasis (LRE), run-length non-uniformity (RLN), and run percentage (RP)] were extracted for mid- (T5-T8) and lower thoracic (T9-T12), and lumbar vertebrae (L1-L5), respectively. Relative annual changes were calculated in texture features and vBMD for each vertebral level and sorted by sex, and changes were checked for statistical significance (P<0.05) using paired t-tests. Root mean square coefficient of variation (RMSCV) and root mean square error (RMSE) were calculated as measures of variability.

Results

SRE, LRE, RLN, and RP exhibited substantial reproducibility with RMSCV-values below 2%, for both sexes and at all spine levels, while vBMD was less reproducible (RMSCV =11.9-16.2%). Entropy showed highest variability (RMSCV =4.34-7.69%) due to statistically significant increases [range, mean ± standard deviation: (4.40±5.78)% to (8.36±8.66)%, P<0.001]. RMSCV of varianceglobal ranged from 1.60% to 3.03%.

Conclusions

Opportunistic assessment of texture features in a single scanner environment using the presented CNN-based framework yields substantial reproducibility, outperforming vBMD reproducibility. Lowest scan-rescan variability was found for higher-order texture features. Further studies are warranted to determine, whether microarchitectural changes to the trabecular bone may be assessed through texture features.

Spatially-aware clustering improves AJCC-8 risk stratification performance in oropharyngeal carcinomas

OBJECTIVE

Evaluate the effectiveness of machine learning tools that incorporate spatial information such as disease location and lymph node metastatic patterns-of-spread, for prediction of survival and toxicity in HPV+ oropharyngeal cancer (OPC).

MATERIALS & METHODS

675 HPV+ OPC patients that were treated at MD Anderson Cancer Center between 2005 and 2013 with curative intent IMRT were retrospectively collected under IRB approval. Risk stratifications incorporating patient radiometric data and lymph node metastasis patterns via an anatomically-adjacent representation with hierarchical clustering were identified. These clusterings were combined into a 3-level patient stratification and included along with other known clinical features in a Cox model for predicting survival outcomes, and logistic regression for toxicity, using independent subsets for training and validation.

RESULTS

Four groups were identified and combined into a 3-level stratification. The inclusion of patient stratifications in predictive models for 5-yr Overall survival (OS), 5-year recurrence free survival, (RFS) and Radiation-associated dysphagia (RAD) consistently improved model performance measured using the area under the curve (AUC). Test set AUC improvements over models with clinical covariates, was 9 % for predicting OS, and 18 % for predicting RFS, and 7 % for predicting RAD. For models with both clinical and AJCC covariates, AUC improvement was 7 %, 9 %, and 2 % for OS, RFS, and RAD, respectively.

CONCLUSION

Including data-driven patient stratifications considerably improve prognosis for survival and toxicity outcomes over the performance achieved by clinical staging and clinical covariates alone. These stratifications generalize well to across cohorts, and sufficient information for reproducing these clusters is included.