Multi-objective radiomics model for predicting distant failure in lung SBRT

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.

Radiomics in stereotactic body radiotherapy for non-small cell lung cancer: a systematic review and radiomic quality score study

PURPOSE

Stereotactic body radiotherapy (SBRT) has been widely utilized for curative treatment of early-stage non-small cell lung cancer (NSCLC). It has achieved good local control rate comparable to surgery. Currently, no standard risk model exists for SBRT outcome or complication prediction. Radiomics has the potential to improve clinical outcome prognostication. Here, we reviewed the current literature on the radiomic analyses of thoracic SBRT through the use of radiomic quality score (RQS).

MATERIALS AND METHODS

Literature search was conducted on PubMed and Embase to retrieve radiomics studies on SBRT for early NSCLC. The literature search included studies up to June 2021. Only full papers published in peer reviewed journals were included. Studies that included metastatic lung cancers or non-lung cancers were excluded. Two independent investigators evaluated each study using the RQS and resolved discrepancies through discussion.

RESULTS

A total number of 25 studies were analysed. The mean RQS was 7.76 of a maximum score of 36. This corresponds to 21.56% of the maximum score. Lack of feature reduction strategies, external validation and open data sharing were identified as key limitations of the reviewed studies. Meanwhile, various common radiomic signatures across different studies such as gray level co-occurrence matrix Homogeneity and energy have been identified. Multiple robust radiomic models have also been reviewed that may improve outcome or complication prediction.

CONCLUSION

Radiomics in thoracic SBRT has a very promising future as a prognostication tool. However, larger multicenter prospective studies are required to confirm radiomic signatures. Improvement in future study methodologies can also facilitate its wider application.

ARMO: automated and reliable multi-objective model for lymph node metastasis prediction in head and neck cancer

Objective. Accurate diagnosis of lymph node metastasis (LNM) is critical in treatment management for patients with head and neck cancer. Positron emission tomography and computed tomography are routinely used for identifying LNM status. However, for small or less fluorodeoxyglucose (FDG) avid nodes, there are always uncertainties in LNM diagnosis. We are aiming to develop a reliable prediction model is for identifying LNM.Approach. In this study, a new automated and reliable multi-objective learning model (ARMO) is proposed. In ARMO, a multi-objective model is introduced to obtain balanced sensitivity and specificity. Meanwhile, confidence is calibrated by introducing individual reliability, whilst the model uncertainty is estimated by a newly defined overall reliability in ARMO. In the training stage, a Pareto-optimal model set is generated. Then all the Pareto-optimal models are used, and a reliable fusion strategy that introduces individual reliability is developed for calibrating the confidence of each output. The overall reliability is calculated to estimate the model uncertainty for each test sample.Main results. The experimental results demonstrated that ARMO obtained more promising results, which the area under the curve, accuracy, sensitivity and specificity can achieve 0.97, 0.93, 0.88 and 0.94, respectively. Meanwhile, based on calibrated confidence and overall reliability, clinicians could pay particular attention to highly uncertain predictions.Significance. In this study, we developed a unified model that can achieve balanced prediction, confidence calibration and uncertainty estimation simultaneously. The experimental results demonstrated that ARMO can obtain accurate and reliable prediction performance.

Towards reliable head and neck cancers locoregional recurrence prediction using delta-radiomics and learning with rejection option

PURPOSE

A reliable locoregional recurrence (LRR) prediction model is important for the personalized management of head and neck cancers (HNC) patients who received radiotherapy. This work aims to develop a delta-radiomics feature-based multi-classifier, multi-objective, and multi-modality (Delta-mCOM) model for post-treatment HNC LRR prediction. Furthermore, we aim to adopt a learning with rejection option (LRO) strategy to boost the reliability of Delta-mCOM model by rejecting prediction for samples with high prediction uncertainties.

METHODS

In this retrospective study, we collected PET/CT image and clinical data from 224 HNC patients who received radiotherapy (RT) at our institution. We calculated the differences between radiomics features extracted from PET/CT images acquired before and after radiotherapy and used them in conjunction with pre-treatment radiomics features as the input features. Using clinical parameters, PET radiomics features, and CT radiomics features, we built and optimized three separate single-modality models. We used multiple classifiers for model construction and employed sensitivity and specificity simultaneously as the training objectives for each of them. Then, for testing samples, we fused the output probabilities from all these single-modality models to obtain the final output probabilities of the Delta-mCOM model. In the LRO strategy, we estimated the epistemic and aleatoric uncertainties when predicting with a trained Delta-mCOM model and identified patients associated with prediction of higher reliability (low uncertainty estimates). The epistemic and aleatoric uncertainties were estimated using an AutoEncoder-style anomaly detection model and test-time augmentation (TTA) with predictions made from the Delta-mCOM model, respectively. Predictions with higher epistemic uncertainty or higher aleatoric uncertainty than given thresholds were deemed unreliable, and they were rejected before providing a final prediction. In this study, different thresholds corresponding to different low-reliability prediction rejection ratios were applied. Their values are based on the estimated epistemic and aleatoric uncertainties distribution of the validation data.

RESULTS

The Delta-mCOM model performed significantly better than the single-modality models, whether trained with pre-, post-treatment radiomics features or concatenated BaseLine and Delta-Radiomics Features (BL-DRFs). It was numerically superior to the PET and CT fused BL-DRF model (nonstatistically significant). Using the LRO strategy for the Delta-mCOM model, most of the evaluation metrics improved as the rejection ratio increased from 0% to around 25%. Utilizing both epistemic and aleatoric uncertainty for rejection yielded nonstatistically significant improved metrics compared to each alone at approximately a 25% rejection ratio. Metrics were significantly better than the no-rejection method when the reject ratio was higher than 50%.

CONCLUSIONS

The inclusion of the delta-radiomics feature improved the accuracy of HNC LRR prediction, and the proposed Delta-mCOM model can give more reliable predictions by rejecting predictions for samples of high uncertainty using the LRO strategy.

Predicting effectiveness of anti-VEGF injection through self-supervised learning in OCT images

Anti-vascular endothelial growth factor (Anti-VEGF) therapy has become a standard way for choroidal neovascularization (CNV) and cystoid macular edema (CME) treatment. However, anti-VEGF injection is a long-term therapy with expensive cost and may be not effective for some patients. Therefore, predicting the effectiveness of anti-VEGF injection before the therapy is necessary. In this study, a new optical coherence tomography (OCT) images based self-supervised learning (OCT-SSL) model for predicting the effectiveness of anti-VEGF injection is developed. In OCT-SSL, we pre-train a deep encoder-decoder network through self-supervised learning to learn the general features using a public OCT image dataset. Then, model fine-tuning is performed on our own OCT dataset to learn the discriminative features to predict the effectiveness of anti-VEGF. Finally, classifier trained by the features from fine-tuned encoder as a feature extractor is built to predict the response. Experimental results on our private OCT dataset demonstrated that the proposed OCT-SSL can achieve an average accuracy, area under the curve (AUC), sensitivity and specificity of 0.93, 0.98, 0.94 and 0.91, respectively. Meanwhile, it is found that not only the lesion region but also the normal region in OCT image is related to the effectiveness of anti-VEGF.

Predicting recurrence risks in lung cancer patients using multimodal radiomics and random survival forests

Purpose

We developed a model integrating multimodal quantitative imaging features from tumor and nontumor regions, qualitative features, and clinical data to improve the risk stratification of patients with resectable non-small cell lung cancer (NSCLC).

Approach

We retrospectively analyzed 135 patients [mean age, 69 years (43 to 87, range); 100 male patients and 35 female patients] with NSCLC who underwent upfront surgical resection between 2008 and 2012. The tumor and peritumoral regions on both preoperative CT and FDG PET-CT and the vertebral bodies L3 to L5 on FDG PET were segmented to assess the tumor and bone marrow uptake, respectively. Radiomic features were extracted and combined with clinical and CT qualitative features. A random survival forest model was developed using the top-performing features to predict the time to recurrence/progression in the training cohort ( n = 101 ), validated in the testing cohort ( n = 34 ) using the concordance, and compared with a stage-only model. Patients were stratified into high- and low-risks of recurrence/progression using Kaplan-Meier analysis.

Results

The model, consisting of stage, three wavelet texture features, and three wavelet first-order features, achieved a concordance of 0.78 and 0.76 in the training and testing cohorts, respectively, significantly outperforming the baseline stage-only model results of 0.67 ( p < 0.005 ) and 0.60 ( p = 0.008 ), respectively. Patients at high- and low-risks of recurrence/progression were significantly stratified in both the training ( p < 0.005 ) and the testing ( p = 0.03 ) cohorts.

Conclusions

Our radiomic model, consisting of stage and tumor, peritumoral, and bone marrow features from CT and FDG PET-CT significantly stratified patients into low- and high-risk of recurrence/progression.

Predicting local persistence/recurrence after radiation therapy for head and neck cancer from PET/CT using a multi-objective, multi-classifier radiomics model

Objectives

Accurate identifying head and neck squamous cell cancer (HNSCC) patients at high risk of local persistence/recurrence (P/R) is of importance for personalized patient management. Here we developed a multi-objective, multi-classifier radiomics model for early HNSCC local P/R prediction based on post-treatment PET/CT scans and clinical data.

Materials and methods

We retrospectively identified 328 individuals (69 patients have local P/R) with HNSCC treated with definitive radiation therapy at our institution. The median follow-up from treatment completion to the first surveillance PET/CT imaging was 114 days (range: 82-159 days). Post-treatment PET/CT scans were reviewed and contoured for all patients. For each imaging modality, we extracted 257 radiomic features to build a multi-objective radiomics model with sensitivity, specificity, and feature sparsity as objectives for model training. Multiple representative classifiers were combined to construct the predictive model. The output probabilities of models built with features from various modalities were fused together to make the final prediction.

Results

We built and evaluated three single-modality models and two multi-modality models. The combination of PET, CT, and clinical data in the multi-objective, multi-classifier radiomics model trended towards the best prediction performance, with a sensitivity of 93%, specificity of 83%, accuracy of 85%, and AUC of 0.94.

Conclusion

Our study demonstrates the feasibility of employing a multi-objective, multi-classifier radiomics model with PET/CT radiomic features and clinical data to predict outcomes for patients with HNSCC after radiation therapy. The proposed prediction model shows the potential to detect cancer local P/R early after radiation therapy.

Applications of artificial intelligence in stereotactic body radiation therapy

This topical review focuses on the applications of artificial intelligence (AI) tools to stereotactic body radiation therapy (SBRT). The high dose per fraction and the limited number of fractions in SBRT require stricter accuracy than standard radiation therapy. The intent of this review is to describe the development and evaluate the possible benefit of AI tools integration into the radiation oncology workflow for SBRT automation. The selected papers were subdivided into four sections, representative of the whole radiotherapy process: ‘AI in SBRT target and organs at risk contouring’, ‘AI in SBRT planning’, ‘AI during the SBRT delivery’, and ‘AI for outcome prediction after SBRT’. Each section summarises the challenges, as well as limits and needs for improvement to achieve better integration of AI tools in the clinical workflow.

Radiomics-guided radiation therapy: opportunities and challenges

Radiomics is an advanced image-processing framework, which extracts image features and considers them as biomarkers towards personalized medicine. Applications include disease detection, diagnosis, prognosis, and therapy response assessment/prediction. As radiation therapy aims for further individualized treatments, radiomics could play a critical role in various steps before, during and after treatment. Elucidation of the concept of radiomics-guided radiation therapy (RGRT) is the aim of this review, attempting to highlight opportunities and challenges underlying the use of radiomics to guide clinicians and physicists towards more effective radiation treatments. This work identifies the value of RGRT in various steps of radiotherapy from patient selection to follow-up, and subsequently provides recommendations to improve future radiotherapy using quantitative imaging features.

[18F]FDG-PET/CT radiomics and artificial intelligence in lung cancer: Technical aspects and potential clinical applications

Lung cancer is the second most common cancer and the leading cause of cancer-related death worldwide. Molecular imaging using [¹⁸F]fluorodeoxyglucose Positron Emission Tomography and/or Computed Tomography ([¹⁸F]FDG-PET/CT) plays an essential role in the diagnosis, evaluation of response to treatment, and prediction of outcomes. The images are evaluated using qualitative and conventional quantitative indices. However, there is far more information embedded in the images, which can be extracted by sophisticated algorithms. Recently, the concept of uncovering and analyzing the invisible data extracted from medical images, called radiomics, is gaining more attention. Currently, [¹⁸F]FDG-PET/CT radiomics is growingly evaluated in lung cancer to discover if it enhances the diagnostic performance or implication of [¹⁸F]FDG-PET/CT in the management of lung cancer. In this review, we provide a short overview of the technical aspects, as they are discussed in different articles of this special issue. We mainly focus on the diagnostic performance of the [¹⁸F]FDG-PET/CT-based radiomics and the role of artificial intelligence in non-small cell lung cancer, impacting the early detection, staging, prediction of tumor subtypes, biomarkers, and patient's outcomes.

Computed tomography-based radiomic model predicts radiological response following stereotactic body radiation therapy in early-stage non-small-cell lung cancer and pulmonary oligo-metastases

Purpose

Radiomic models elaborate geometric and texture features of tumors extracted from imaging to develop predictors for clinical outcomes. Stereotactic body radiation therapy (SBRT) has been increasingly applied in the ablative treatment of thoracic tumors. This study aims to identify predictors of treatment responses in patients affected by early stage non-small cell lung cancer (NSCLC) or pulmonary oligo-metastases treated with SBRT and to develop an accurate machine learning model to predict radiological response to SBRT.

Materials and Methods

Computed tomography (CT) images of 85 tumors (stage I–II NSCLC and pulmonary oligo-metastases) from 69 patients treated with SBRT were analyzed. Gross tumor volumes (GTV) were contoured on CT images. Patients that achieved complete response (CR) or partial response (PR) were defined as responders. One hundred ten radiomic features were extracted using PyRadiomics module based on the GTV. The association of features with response to SBRT was evaluated. A model using support vector machine (SVM) was then trained to predict response based solely on the extracted radiomics features. Receiver operating characteristic curves were constructed to evaluate model performance of the identified radiomic predictors.

Results

Sixty-nine patients receiving thoracic SBRT from 2008 to 2018 were retrospectively enrolled. Skewness and root mean squared were identified as radiomic predictors of response to SBRT. The SVM machine learning model developed had an accuracy of 74.8%. The area under curves for CR, PR, and non-responder prediction were 0.86 (95% confidence interval [CI], 0.794–0.921), 0.946 (95% CI, 0.873–0.978), and 0.857 (95% CI, 0.789–0.915), respectively.

Conclusion

Radiomic analysis of pre-treatment CT scan is a promising tool that can predict tumor response to SBRT.

Radiomics for Predicting Lung Cancer Outcomes Following Radiotherapy: A Systematic Review

Lung cancer's radiomic phenotype may potentially inform clinical decision-making with respect to radical radiotherapy. At present there are no validated biomarkers available for the individualisation of radical radiotherapy in lung cancer and the mortality rate of this disease remains the highest of all other solid tumours. MEDLINE was searched using the terms 'radiomics' and 'lung cancer' according to the Preferred Reporting Items for Systematic Reviews and Met-Analyses (PRISMA) guidance. Radiomics studies were defined as those manuscripts describing the extraction and analysis of at least 10 quantifiable imaging features. Only those studies assessing disease control, survival or toxicity outcomes for patients with lung cancer following radical radiotherapy ± chemotherapy were included. Study titles and abstracts were reviewed by two independent reviewers. The Radiomics Quality Score was applied to the full text of included papers. Of 244 returned results, 44 studies met the eligibility criteria for inclusion. End points frequently reported were local (17%), regional (17%) and distant control (31%), overall survival (79%) and pulmonary toxicity (4%). Imaging features strongly associated with clinical outcomes include texture features belonging to the subclasses Gray level run length matrix, Gray level co-occurrence matrix and kurtosis. The median cohort size for model development was 100 (15-645); in the 11 studies with external validation in a separate independent population, the median cohort size was 84 (21-295). The median number of imaging features extracted was 184 (10-6538). The median Radiomics Quality Score was 11% (0-47). Patient-reported outcomes were not incorporated within any studies identified. No studies externally validated a radiomics signature in a registered prospective study. Imaging-derived indices attained through radiomic analyses could equip thoracic oncologists with biomarkers for treatment response, patterns of failure, normal tissue toxicity and survival in lung cancer. Based on routine scans, their non-invasive nature and cost-effectiveness are major advantages over conventional pathological assessment. Improved tools are required for the appraisal of radiomics studies, as significant barriers to clinical implementation remain, such as standardisation of input scan data, quality of reporting and external validation of signatures in randomised, interventional clinical trials.

Immunotherapy treatment outcome prediction in metastatic melanoma through an automated multi-objective delta-radiomics model

Based on recent studies, immunotherapy led by immune checkpoint inhibitors has significantly improved the patient survival rate and effectively reduced the recurrence risk. However, immunotherapy has different therapeutic effects for different patients, leading to difficulties in predicting the treatment response. Conversely, delta-radiomic features, which measure the difference between pre- and post-treatment through quantitative image features, have proven to be promising descriptors for treatment outcome prediction. Consequently, we developed an effective model termed as the automated multi-objective delta-radiomics (Auto-MODR) model for the prediction of immunotherapy response in metastatic melanoma. In Auto-MODR, delta-radiomic features and traditional radiomic features were used as inputs. Furthermore, a novel automated multi-objective model was developed to obtain more reliable and balanced results between sensitivity and specificity. We conducted extensive comparisons with existing studies on treatment outcome prediction. Our method achieved an area under the curve (AUC) of 0.86 in a cross-validation study and an AUC of 0.73 in an independent study. Compared with the model using conventional radiomic features (pre- and post-treatment) only, better performance can be obtained when conventional radiomic and delta-radiomic features are combined. Furthermore, Auto-MODR outperformed the currently available radiomic strategies.

Interobserver Variability in the Computed Tomography Assessment of Pulmonary Injury and Tumor Recurrence After Stereotactic Body Radiotherapy

PURPOSE

To evaluate the interobserver agreement of chest computed tomography (CT) findings in the diagnosis of expected changes and local recurrence after stereotactic body radiation therapy (SBRT) in patients with early-stage lung cancer or a single pulmonary metastasis.

MATERIALS AND METHODS

A total of 54 patients with early-stage lung cancer or pulmonary metastasis who were treated with SBRT from 2007 to 2015 were included. The exclusion criteria were patients who presented with pulmonary infection during follow-up and patients who underwent a single CT during follow-up. The imaging features on CT were assessed by 3 blinded radiologists at the following 2 time points after SBRT: (a) early follow-up and (b) late follow-up (≥6 mo). The radiologists classified the findings as expected changes after SBRT or recurrence. Interobserver agreement was assessed by kappa and Wilcoxon statistics.

RESULTS

A total of 13 women and 41 men with a mean age of 75.3 (±8.9) years were selected. The total and per fraction SBRT doses were 54 Gy (interquartile range: 45 to 54) and 18 Gy (interquartile range: 15 to 18), respectively. All expected changes and findings suggestive of recurrence had an almost perfect agreement (κ>0.85) among readers, except for diffuse consolidation in the early period (κ=0.65).

CONCLUSION

CT findings demonstrate high interobserver agreement for expected changes and for findings indicating recurrence after SBRT.

An MRI-based multi-objective radiomics model predicts lymph node status in patients with rectal cancer

PURPOSE

To apply a multi-objective radiomics model based on pre-operative magnetic resonance imaging (MRI) for improving diagnostic accuracy of LN metastasis in rectal cancer patients.

METHODS

This study consisted of 91 patients diagnosed with rectal cancer from April 2018 to March 2019. All patients underwent rectal MRI before surgery without any other treatment. Clinical data, subjective radiologist assessments, and radiomic features of LNs were obtained. A total of 1409 radiomic features were extracted from T2WI LN images. Multi-objective optimization with the iterative multi-objective immune algorithm (IMIA) was used to select radiomic features to build prediction models. Predictive performances of radiomic, radiologist, and combined radiomic and radiologist models were assessed for accuracy by receiver operating characteristics (ROC) curves.

RESULTS

For the radiologist analysis, heterogeneity was the only significant independent predictor of LN status. The sensitivity, specificity, and accuracy of the subjective radiologist analysis were 72.09%, 73.81%, and 78.12%, respectively. The sensitivity, specificity, and accuracy of the solitary radiomic model consisting of 10 features were 89.81%, 82.57%, and 87.77%, respectively. The sensitivity, specificity, and accuracy of the combined model, which consisted of 12 radiomic and radiologist features, were 92.23%, 84.69%, and 89.88%, respectively. The combined model had the best prediction performance with an AUC of 0.94.

CONCLUSIONS

The multi-objective radiomics model based on T2WI images was very useful in predicting pre-operative LN status in rectal cancer patients.

Comparison and Fusion of Machine Learning Algorithms for Prospective Validation of PET/CT Radiomic Features Prognostic Value in Stage II-III Non-Small Cell Lung Cancer

Machine learning (ML) algorithms for selecting and combining radiomic features into multiparametric prediction models have become popular; however, it has been shown that large variations in performance can be obtained by relying on different approaches. The purpose of this study was to evaluate the potential benefit of combining different algorithms into an improved consensus for the final prediction, as it has been shown in other fields. Methods: The evaluation was carried out in the context of the use of radiomics from ¹⁸F-FDG PET/CT images for predicting outcome in stage II-III Non-Small Cell Lung Cancer. A cohort of 138 patients was exploited for the present analysis. Eighty-seven patients had been previously recruited retrospectively for another study and were used here for training and internal validation. We also used data from prospectively recruited patients (n = 51) for testing. Three different machine learning pipelines relying on embedded feature selection were trained to predict overall survival (OS) as a binary classification: Support Vector machines (SVMs), Random Forests (RFs), and Logistic Regression (LR). Two different clinical endpoints were investigated: median OS or OS shorter than 6 months. The fusion of the three approaches was implemented using two different strategies: majority voting on the binary outputs or averaging of the output probabilities. Results: Our results confirm previous findings, highlighting that different ML pipelines select different sets of features and reach different classification performances (accuracy in the testing set ranging between 63% and 67% for median OS, and between 75% and 80% for OS < 6 months). Generating a consensus improved the performance for both endpoints; with the probabilities averaging strategy outperforming the majority voting (accuracy of 78% vs. 71% for median OS and 89 vs. 84% for OS < 6 months). Overall, the performance of these radiomic-based models outperformed the standard clinical staging in both endpoints (accuracy of 58% and 53% accuracy in the testing set for each endpoint). Conclusion: Although obtained in a small cohort of patients, our results suggest that a consensus of machine learning algorithms can improve performance in the context of radiomics. The resulting prognostic stratification in the prospective testing cohort is higher than when relying on the clinical stage. This could be of interest for clinical practice as it could help to identify patients with higher risk amongst stage II and III patients, who could benefit from intensified treatment and/or more frequent follow-up after treatment.

Quantitative PET Imaging and Clinical Parameters as Predictive Factors for Patients With Cervical Carcinoma: Implications of a Prediction Model Generated Using Multi-Objective Support Vector Machine Learning

Purpose:

Quantitative features from pre-treatment positron emission tomography (PET) have been used to predict treatment outcomes for patients with cervical carcinoma. The purpose of this study is to use quantitative PET imaging features and clinical parameters to construct a multi-objective machine learning predictive model.

Materials/Methods:

Seventy-five patients with stage IB2-IVA disease treated at our institution from 2009–2012 were analyzed. Models predicting locoregional and distant failure were generated using clinical parameters (age, race, stage, histology, tumor size, nodal status) and imaging features (12 textural, 9 intensity, 8 geometric features, 2 additional imaging features) from pre-treatment PET. Model features were selected based on a multi-objective evolutionary algorithm to maximize specificity given a fixed moderately high sensitivity using support vector machine learning methods. Model 1 used clinical parameters only (C), Model 2 used imaging features only (I), and Model 3 used clinical and imaging features (C+I). Sensitivity, specificity, area under a receiver-operating characteristic curve (AUC), and p-values were compared to assess ability to predict locoregional and distant failure.

Results:

C+I had the highest performance for both locoregional failure (AUC 0.84, p < 0.01; specificity: 0.86; sensitivity: 0.79) and distant failure (AUC 0.75, p < 0.01; specificity: 0.75; sensitivity: 0.75).

Conclusions:

Based on a moderately high fixed sensitivity and optimized for specificity, the model using both clinical parameters and imaging features (C+I) had the best performance in predicting both locoregional failure and distant failure.

Ultrasonography-Based Radiomics of Screening-Detected Ductal Carcinoma In Situ According to Visibility on Mammography

ABSTRACT

Ductal carcinoma in situ (DCIS) has different prognostic factors according to the detection modality. The purpose of this study was to compare parameters from a radiomic analysis of ultrasonography (US) images for DCIS detected on screening mammography (MMG) and US and detected on screening US only. A total of 154 surgically confirmed DCIS visible on US were included. Regions of interest were drawn onto US images of DCIS, and texture analysis was performed. Lesions were classified into those detected by both US and MMG (the US-MMG group) and those detected by US only (the US group). Analysis parameters were compared between the US-MMG group and the US group. Ninety-six lesions were included in the US-MMG group and 58 lesions in the US group. Energy, entropy, maximum, mean absolute deviation, range, SD, and variance were significantly higher in the US-MMG group than the US group. Kurtosis, skewness, and uniformity were significantly lower in the US-MMG group than the US group. Among the 22 gray-level cooccurrence matrix parameters, 18, 21, 22, 20, and 21 parameters were significantly different between the 2 groups in 0, 45, 90, and 135 degrees and the average value. Among the 11 gray-level run-length matrix parameters, 6, 6, 7, 7, and 6 parameters were significantly different in 0, 45, 90, and 135 degrees and the average value. Inverse variance and gray-level nonuniformity were the most different features between the 2 groups. Screening-detected DCIS showed different radiomic features according to the detection modality.

Machine Learning-Based Differentiation of Nontuberculous Mycobacteria Lung Disease and Pulmonary Tuberculosis Using CT Images

An increasing number of patients infected with nontuberculous mycobacteria (NTM) are observed worldwide. However, it is challenging to identify NTM lung diseases from pulmonary tuberculosis (PTB) due to considerable overlap in classic manifestations and clinical and radiographic characteristics. This study quantifies both cavitary and bronchiectasis regions in CT images and explores a machine learning approach for the differentiation of NTM lung diseases and PTB. It involves 116 patients and 103 quantitative features. After the selection of informative features, a linear support vector machine performs disease classification, and simultaneously, discriminative features are recognized. Experimental results indicate that bronchiectasis is relatively more informative, and two features are figured out due to promising prediction performance (area under the curve, 0.84 ± 0.06; accuracy, 0.85 ± 0.06; sensitivity, 0.88 ± 0.07; and specificity, 0.80 ± 0.12). This study provides insight into machine learning-based identification of NTM lung diseases from PTB, and more importantly, it makes early and quick diagnosis of NTM lung diseases possible that can facilitate lung disease management and treatment planning.

Pilot Study of CT-Based Radiomics Model for Early Evaluation of Response to Immunotherapy in Patients With Metastatic Melanoma

Objective: Determine the performance of a computed tomography (CT) -based radiomics model in predicting early response to immunotherapy in patients with metastatic melanoma. Methods: This retrospective study examined 50 patients with metastatic melanoma who received immunotherapy treatment in our hospital with an anti-programmed cell death-1 (PD-1) agent or an inhibitor of cytotoxic T lymphocyte antigen-4 (CTLA-4). Thirty-four patients who received an anti-PD-1 agent were in the training sample and 16 patients who received a CTLA-4 inhibitor were in the validation sample. Patients with true progressive disease (PD) were in the poor response group, and those with pseudoprogression, complete response (CR), partial response (PR), or stable disease (SD) were in the good response group. CT images were examined at baseline and after the first and second cycles of treatment, and the imaging data were extracted for radiomics modeling. Results: The radiomics model based on pre-treatment, post-treatment, and delta features provided the best results for predicting response to immunotherapy. Receiver operating characteristic (ROC) analysis for good response indicated an area under the curve (AUC) of 0.882 for the training group and an AUC of 0.857 for the validation group. The sensitivity, specificity, and accuracy of model were 85.70% (6/7), 66.70% (6/9), and 75% (12/16) for predicting a good response. Conclusion: A CT-based radiomics model for metastatic melanoma has the potential to predict early response to immunotherapy and to identify pseudoprogression.

Multifaceted radiomics for distant metastasis prediction in head & neck cancer

Accurately predicting distant metastasis in head & neck cancer has the potential to improve patient survival by allowing early treatment intensification with systemic therapy for high-risk patients. By extracting large amounts of quantitative features and mining them, radiomics has achieved success in predicting treatment outcomes for various diseases. However, there are several challenges associated with conventional radiomic approaches, including: (1) how to optimally combine information extracted from multiple modalities; (2) how to construct models emphasizing different objectives for different clinical applications; and (3) how to utilize and fuse output obtained by multiple classifiers. To overcome these challenges, we propose a unified model termed as multifaceted radiomics (M-radiomics). In M-radiomics, a deep learning with stacked sparse autoencoder is first utilized to fuse features extracted from different modalities into one representation feature set. A multi-objective optimization model is then introduced into M-radiomics where probability-based objective functions are designed to maximize the similarity between the probability output and the true label vector. Finally, M-radiomics employs multiple base classifiers to get a diverse Pareto-optimal model set and then fuses the output probabilities of all the Pareto-optimal models through an evidential reasoning rule fusion (ERRF) strategy in the testing stage to obtain the final output probability. Experimental results show that M-radiomics with the stacked autoencoder outperforms the model without the autoencoder. M-radiomics obtained more accurate results with a better balance between sensitivity and specificity than other single-objective or single-classifier-based models.

Artificial Intelligence in Lung Cancer: Bridging the Gap Between Computational Power and Clinical Decision-Making

Lung cancer remains the most common cause of cancer death worldwide. Recent advances in lung cancer screening, radiotherapy, surgical techniques, and systemic therapy have led to increasing complexity in diagnosis, treatment decision-making, and assessment of recurrence. Artificial intelligence (AI)-based prediction models are being developed to address these issues and may have a future role in screening, diagnosis, treatment selection, and decision-making around salvage therapy. Imaging plays an essential role in all components of lung cancer management and has the potential to play a key role in AI applications. Artificial intelligence has demonstrated value in prognostic biomarker discovery in lung cancer diagnosis, treatment, and response assessment, putting it at the forefront of the next phase of personalized medicine. However, although exploratory studies demonstrate potential utility, there is a need for rigorous validation and standardization before AI can be utilized in clinical decision-making. In this review, we will provide a summary of the current literature implementing AI for outcome prediction in lung cancer. We will describe the anticipated impact of AI on the management of patients with lung cancer and discuss the challenges of clinical implementation of these techniques.

A Reliable Multi-classifier Multi-objective Model for Predicting Recurrence in Triple Negative Breast Cancer

Recurrence is a significant prognostic factor in patients with triple negative breast cancer, and the ability to accurately predict it is essential for treatment optimization. Machine learning is a preferred strategy for recurrence prediction. Most current predictive models are built based on single classifier and trained through a single objective. However, since many classifiers are available, selecting an optimal model is challenging. On the other hand, a single objective may not be a good measure to guide model training. We proposed a new multi-classifier multi-objective (MCMO) recurrence predictive model. Specifically, new similarity-based sensitivity and specificity were defined and considered as the two objective functions simultaneously during training. Also the evidential reasoning (ER) approach was used for fusing the output of each classifier to obtain more reliable outcome. Using the proposed MCMO model, we achieved a predictive area under the receiver operating characteristic curve (AUC) of 0.9 with balanced sensitivity and specificity. Furthermore, MCMO outperformed all the individual classifiers, and yielded more reliable results than other commonly used optimization and fusion methods.

Interdisciplinarity: An essential requirement for translation of radiomics research into clinical practice -a systematic review focused on thoracic oncology

BACKGROUND

Recently, evidence has accumulated that demonstrates the potential for future applications of radiomics in many clinical settings, including thoracic oncology. Methodological reasons for the immaturity of image mining (radiomics and artificial intelligence-based) studies have been identified. However, data on the influence of the composition of the research team on the quality of investigations in radiomics are lacking.

AIM

This review aims to evaluate the interdisciplinarity within studies on radiomics in thoracic oncology in order to assess its influence on the quality of research (QUADAS-2 score) in the image mining field.

METHODS

We considered for inclusion radiomics investigations with objectives relating to clinical practice in thoracic oncology. Subsequently, we interviewed the corresponding authors. The field of expertise and/or educational degree was then used to assess interdisciplinarity. Subsequently, all studies were evaluated applying the QUADAS-2 score and assigned to a research phase from 0 to IV.

RESULTS

Overall, 27 studies were included. The study quality according to the QUADAS-2 score was low (score ≤5) in 8, moderate (=6) in 12, and high (≥7) in 7 papers. An interdisciplinary team (at least 3 different expertise categories) was involved in half of the papers without any type of validation and in all papers with independent validation. Clinicians were not involved in phase 0 studies while they contributed to all papers classified as phase I and to 4/5 papers classified as phase II with independent validation.

CONCLUSIONS

The composition of the research team influences the quality of investigations in radiomics. Also, growth in interdisciplinarity appears to reflect research development from the early phase to a more mature, clinically oriented stage of investigation.

[Prognosis factors after lung stereotactic body radiotherapy for non-small cell lung carcinoma]

Lung cancer is the fourth most common cancer in France with a prevalence of 30,000 new cases per year. Lobectomy surgery with dissection is the gold standard treatment for T1-T2 localized non-small cell lung carcinoma. A segmentectomy may be proposed to operable patients but fragile from a respiratory point of view. For inoperable patients or patients with unsatisfactory pulmonary function tests, local treatment with stereotactic radiotherapy may be proposed to achieve local control rates ranging from 85 to 95% at 3-5 years. Several studies have examined prognostic factors after stereotaxic pulmonary radiotherapy. We conducted a general review of the literature to identify factors affecting local control.

Multi-objective ensemble deep learning using electronic health records to predict outcomes after lung cancer radiotherapy

Accurately predicting treatment outcome is crucial for creating personalized treatment plans and follow-up schedules. Electronic health records (EHRs) contain valuable patient-specific information that can be leveraged to improve outcome prediction. We propose a reliable multi-objective ensemble deep learning (MoEDL) method that uses features extracted from EHRs to predict high risk of treatment failure after radiotherapy in patients with lung cancer. The dataset used in this study contains EHRs of 814 patients who had not achieved disease-free status and 193 patients who were disease-free with at least one year follow-up time after lung cancer radiation therapy. The proposed MoEDL consists of three phases: (1) training with dynamic ensemble deep learning; (2) model selection with adaptive multi-objective optimization; and (3) testing with evidential reasoning (ER) fusion. Specifically, in the training phase, we employ deep perceptron networks as base learners to handle various issues with EHR data. The architecture and key hyper-parameters of each base learner are dynamically adjusted to increase the diversity of learners while reducing the time spent tuning hyper-parameters. Furthermore, we integrate the snapshot ensembles (SE) restarting strategy, multi-objective optimization, and ER fusion to improve the prediction robustness and accuracy of individual networks. The SE restarting strategy can yield multiple candidate models at no additional training cost in the training stage. The multi-objective model simultaneously considers sensitivity, specificity, and AUC as objective functions, overcoming the limitations of single-objective-based model selection. For the testing stage, we utilized an analytic ER rule to fuse the output scores from each optimal model to obtain reliable and robust predictive results. Our experimental results demonstrate that MoEDL can perform better than other conventional methods.

Advanced 4-dimensional cone-beam computed tomography reconstruction by combining motion estimation, motion-compensated reconstruction, biomechanical modeling and deep learning

4-Dimensional cone-beam computed tomography (4D-CBCT) offers several key advantages over conventional 3D-CBCT in moving target localization/delineation, structure de-blurring, target motion tracking, treatment dose accumulation and adaptive radiation therapy. However, the use of the 4D-CBCT in current radiation therapy practices has been limited, mostly due to its sub-optimal image quality from limited angular sampling of cone-beam projections. In this study, we summarized the recent developments of 4D-CBCT reconstruction techniques for image quality improvement, and introduced our developments of a new 4D-CBCT reconstruction technique which features simultaneous motion estimation and image reconstruction (SMEIR). Based on the original SMEIR scheme, biomechanical modeling-guided SMEIR (SMEIR-Bio) was introduced to further improve the reconstruction accuracy of fine details in lung 4D-CBCTs. To improve the efficiency of reconstruction, we recently developed a U-net-based deformation-vector-field (DVF) optimization technique to leverage a population-based deep learning scheme to improve the accuracy of intra-lung DVFs (SMEIR-Unet), without explicit biomechanical modeling. Details of each of the SMEIR, SMEIR-Bio and SMEIR-Unet techniques were included in this study, along with the corresponding results comparing the reconstruction accuracy in terms of CBCT images and the DVFs. We also discussed the application prospects of the SMEIR-type techniques in image-guided radiation therapy and adaptive radiation therapy, and presented potential schemes on future developments to achieve faster and more accurate 4D-CBCT imaging.

Predicting lung nodule malignancies by combining deep convolutional neural network and handcrafted features

To predict lung nodule malignancy with a high sensitivity and specificity for low dose CT (LDCT) lung cancer screening, we propose a fusion algorithm that combines handcrafted features (HF) into the features learned at the output layer of a 3D deep convolutional neural network (CNN). First, we extracted twenty-nine HF, including nine intensity features, eight geometric features, and twelve texture features based on grey-level co-occurrence matrix (GLCM). We then trained 3D CNNs modified from three 2D CNN architectures (AlexNet, VGG-16 Net and Multi-crop Net) to extract the CNN features learned at the output layer. For each 3D CNN, the CNN features combined with the 29 HF were used as the input for the support vector machine (SVM) coupled with the sequential forward feature selection (SFS) method to select the optimal feature subset and construct the classifiers. The fusion algorithm takes full advantage of the HF and the highest level CNN features learned at the output layer. It can overcome the disadvantage of the HF that may not fully reflect the unique characteristics of a particular lesion by combining the intrinsic CNN features. Meanwhile, it also alleviates the requirement of a large scale annotated dataset for the CNNs based on the complementary of HF. The patient cohort includes 431 malignant nodules and 795 benign nodules extracted from the LIDC/IDRI database. For each investigated CNN architecture, the proposed fusion algorithm achieved the highest AUC, accuracy, sensitivity, and specificity scores among all competitive classification models.

[Prediction of rectal toxicity of radiotherapy for prostate cancer based on multi-modality feature and multi-classifiers]

OBJECTIVE

To evaluate rectal toxicity of radiotherapy for prostate cancer using a novel predictive model based on multi-modality and multi-classifier fusion.

METHODS

We retrospectively collected the clinical data from 44 prostate cancer patients receiving external beam radiation (EBRT), including the treatment data, clinical parameters, planning CT data and the treatment plans. The clinical parameter features and dosimetric features were extracted as two different modality features, and a subset of features was selected to train the 5 base classifiers (SVM, Decision Tree, K-nearest-neighbor, Random forests and XGBoost). To establish the multi-modality and multi-classifier fusion model, a multi-criteria decision-making based weight assignment algorithm was used to assign weights for each base classifier under the same modality. A repeat 5-fold cross-validation and the 4 indexes including the area under ROC curve (AUC), accuracy, sensitivity and specificity were used to evaluate the proposed model. In addition, the proposed model was compared quantitatively with different feature selection methods, different weight allocation algorithms, the model based on single mode single classifier, and two integrated models using other fusion methods.

RESULTS

Repeated (5 times) 5-fold cross validation of the proposed model showed an accuracy of 0.78 for distinguishing toxicity from non-toxicity with an AUC of 0.83, a specificity of 0.79 and a sensitivity of 0.76.

CONCLUSIONS

Compared with the models based on a single mode or a single classifier and other fusion models, the proposed model can more accurately predict rectal toxicity of radiotherapy for prostate cancer.

A machine-learning-based prediction model of fistula formation after interstitial brachytherapy for locally advanced gynecological malignancies

PURPOSE

External beam radiotherapy combined with interstitial brachytherapy is commonly used to treat patients with bulky, advanced gynecologic cancer. However, the high radiation dose needed to control the tumor may result in fistula development. There is a clinical need to identify patients at high risk for fistula formation such that treatment may be managed to prevent this toxic side effect. This work aims to develop a fistula prediction model framework using machine learning based on patient, tumor, and treatment features.

METHODS AND MATERIALS

This retrospective study included 35 patients treated at our institution using interstitial brachytherapy for various gynecological malignancies. Five patients developed rectovaginal fistula and two developed both rectovaginal and vesicovaginal fistula. For each patient, 31 clinical features of multiple data types were collected to develop a fistula prediction framework. A nonlinear support vector machine was used to build the prediction model. Sequential backward feature selection and sequential floating backward feature selection methods were used to determine optimal feature sets. To overcome data imbalance issues, the synthetic minority oversampling technique was used to generate synthetic fistula cases for model training.

RESULTS

Seven mixed data features were selected by both sequential backward selection and sequential floating backward selection methods. Our prediction model using these features achieved a high prediction accuracy, that is, 0.904 area under the curve, 97.1% sensitivity, and 88.5% specificity.

CONCLUSIONS

A machine-learning-based prediction model of fistula formation has been developed for patients with advanced gynecological malignancies treated using interstitial brachytherapy. This model may be clinically impactful pending refinement and validation in a larger series.

Combining many-objective radiomics and 3D convolutional neural network through evidential reasoning to predict lymph node metastasis in head and neck cancer

Lymph node metastasis (LNM) is a significant prognostic factor in patients with head and neck cancer, and the ability to predict it accurately is essential to optimizing treatment. Positron emission tomography (PET) and computed tomography (CT) imaging are routinely used to identify LNM. Although large or highly active lymph nodes (LNs) have a high probability of being positive, identifying small or less reactive LNs is challenging. The accuracy of LNM identification strongly depends on the physician's experience, so an automatic prediction model for LNM based on CT and PET images is warranted to assist LMN identification across care providers and facilities. Radiomics and deep learning are the two promising imaging-based strategies for node malignancy prediction. Radiomics models are built based on handcrafted features, while deep learning learns the features automatically. To build a more reliable model, we proposed a hybrid predictive model that takes advantages of both radiomics and deep learning based strategies. We designed a new many-objective radiomics (MaO-radiomics) model and a 3D convolutional neural network (3D-CNN) that fully utilizes spatial contextual information, and we fused their outputs through an evidential reasoning (ER) approach. We evaluated the performance of the hybrid method for classifying normal, suspicious and involved LNs. The hybrid method achieves an accuracy (ACC) of 0.88 while XmasNet and Radiomics methods achieve 0.81 and 0.75, respectively. The hybrid method provides a more accurate way for predicting LNM using PET and CT.

Multi-Objective-Based Radiomic Feature Selection for Lesion Malignancy Classification

OBJECTIVE

accurately classifying the malignancy of lesions detected in a screening scan is critical for reducing false positives. Radiomics holds great potential to differentiate malignant from benign tumors by extracting and analyzing a large number of quantitative image features. Since not all radiomic features contribute to an effective classifying model, selecting an optimal feature subset is critical.

METHODS

this work proposes a new multi-objective based feature selection (MO-FS) algorithm that considers sensitivity and specificity simultaneously as the objective functions during feature selection. For MO-FS, we developed a modified entropy-based termination criterion that stops the algorithm automatically rather than relying on a preset number of generations. We also designed a solution selection methodology for multi-objective learning that uses the evidential reasoning approach (SMOLER) to automatically select the optimal solution from the Pareto-optimal set. Furthermore, we developed an adaptive mutation operation to generate the mutation probability in MO-FS automatically.

RESULTS

we evaluated the MO-FS for classifying lung nodule malignancy in low-dose CT and breast lesion malignancy in digital breast tomosynthesis.

CONCLUSION

the experimental results demonstrated that the feature set selected by MO-FS achieved better classification performance than features selected by other commonly used methods.

SIGNIFICANCE

the proposed method is general and more effective radiomic feature selection strategy.

A pilot study using kernelled support tensor machine for distant failure prediction in lung SBRT

We developed a kernelled support tensor machine (KSTM)-based model with tumor tensors derived from pre-treatment PET and CT imaging as input to predict distant failure in early stage non-small cell lung cancer (NSCLC) treated with stereotactic body radiation therapy (SBRT). The patient cohort included 110 early stage NSCLC patients treated with SBRT, 25 of whom experienced failure at distant sites. Three-dimensional tumor tensors were constructed and used as input for the KSTM-based classifier. A KSTM iterative algorithm with a convergent proof was developed to train the weight vectors for every mode of the tensor for the classifier. In contrast to conventional radiomics approaches that rely on handcrafted imaging features, the KSTM-based classifier uses 3D imaging as input, taking full advantage of the imaging information. The KSTM-based classifier preserves the intrinsic 3D geometry structure of the medical images and the correlation in the original images and trains the classification hyper-plane in an adaptive feature tensor space. The KSTM-based predictive algorithm was compared with three conventional machine learning models and three radiomics approaches. For PET and CT, the KSTM-based predictive method achieved the highest prediction results among the seven methods investigated in this study based on 10-fold cross validation and independent testing.

Developing Predictive or Prognostic Biomarkers for Charged Particle Radiotherapy

The response to radiotherapy can vary greatly among individuals, even though advances in technology allow for the highly localized placement of therapeutic doses of radiation to a tumor. This variability in patient response to radiation is biologically driven, but the individuality of tumor and healthy tissue biology are not used to create individual treatment plans. Biomarkers of radiosensitivity, whether intrinsic or from hypoxia, would move radiation oncology from precision medicine to precise, personalized medicine. Charged particle radiotherapy allows for even greater dose conformity, but the biological advantages of charged particle radiotherapy have not yet been cultivated. The development of biomarkers that would drive biologically based clinical trials, identify patients for whom charged particles are most appropriate, or aid in particle-selection strategies could be envisioned with appropriate biomarkers. Initially, biomarkers for low-linear energy transfer (LET) radiation responses should be tested against charged particles. Biomarkers of tumor radioresistance to low-LET radiations could be used to identify patients for whom the enhanced relative biological effectiveness (RBE) of charged particles would be more effective compared with low-LET radiations and those for whom specific DNA-repair inhibitors, in combination with charged particles, may also be appropriate. Furthermore, heavy charged particles can overcome the radioresistance of hypoxic tumors when used at the appropriate LET. Biomarkers for hypoxia could identify hypoxic tumors and, in combination with imaging, define hypoxic regions of a tumor for specific ion selection. Moreover, because of the enhanced RBE for charged particles, the risk for adverse healthy tissue effects may be greater, even though charged particles have greater tumor conformality. There are many validated healthy-tissue biomarkers available to test against charged particle exposures. Lastly, newer biological techniques, as well as newer bioinformatic and computational methods, are rapidly changing the landscape for biomarker identification, validation, and clinical trial design.

Reliable gene mutation prediction in clear cell renal cell carcinoma through multi-classifier multi-objective radiogenomics model

Genetic studies have identified associations between gene mutations and clear cell renal cell carcinoma (ccRCC). Since the complete gene mutational landscape cannot be characterized through biopsy and sequencing assays for each patient, non-invasive tools are needed to determine the mutation status for tumors. Radiogenomics may be an attractive alternative tool to identify disease genomics by analyzing amounts of features extracted from medical images. Most current radiogenomics predictive models are built based on a single classifier and trained through a single objective. However, since many classifiers are available, selecting an optimal model is challenging. On the other hand, a single objective may not be a good measure to guide model training. We proposed a new multi-classifier multi-objective (MCMO) radiogenomics predictive model. To obtain more reliable prediction results, similarity-based sensitivity and specificity were defined and considered as the two objective functions simultaneously during training. To take advantage of different classifiers, the evidential reasoning (ER) approach was used for fusing the output of each classifier. Additionally, a new similarity-based multi-objective optimization algorithm (SMO) was developed for training the MCMO to predict ccRCC related gene mutations (VHL, PBRM1 and BAP1) using quantitative CT features. Using the proposed MCMO model, we achieved a predictive area under the receiver operating characteristic curve (AUC) over 0.85 for VHL, PBRM1 and BAP1 genes with balanced sensitivity and specificity. Furthermore, MCMO outperformed all the individual classifiers, and yielded more reliable results than other optimization algorithms and commonly used fusion strategies.

Predicting Lymph Node Metastasis in Head and Neck Cancer by Combining Many-objective Radiomics and 3-dimensioal Convolutional Neural Network through Evidential Reasoning

Lymph node metastasis (LNM) is a significant prognostic factor in patients with head and neck cancer, and the ability to predict it accurately is essential for treatment optimization. PET and CT imaging are routinely used for LNM identification. However, uncertainties of LNM always exist especially for small size or reactive nodes. Radiomics and deep learning are the two preferred imaging-based strategies for node malignancy prediction. Radiomics models are built based on handcrafted features, and deep learning can learn the features automatically. We proposed a hybrid predictive model that combines many-objective radiomics (MO-radiomics) and 3-dimensional convolutional neural network (3D-CNN) through evidential reasoning (ER) approach. To build a more reliable model, we proposed a new many-objective radiomics model. Meanwhile, we designed a 3D-CNN that fully utilizes spatial contextual information. Finally, the outputs were fused through the ER approach. To study the predictability of the two modalities, three models were built for PET, CT, and PET& CT. The results showed that the model performed best when the two modalities were combined. Moreover, we showed that the quantitative results obtained from the hybrid model were better than those obtained from MO-radiomics and 3D-CNN.

Radiomics in Nuclear Medicine Applied to Radiation Therapy: Methods, Pitfalls, and Challenges

Radiomics is a recent area of research in precision medicine and is based on the extraction of a large variety of features from medical images. In the field of radiation oncology, comprehensive image analysis is crucial to personalization of treatments. A better characterization of local heterogeneity and the shape of the tumor, depicting individual cancer aggressiveness, could guide dose planning and suggest volumes in which a higher dose is needed for better tumor control. In addition, noninvasive imaging features that could predict treatment outcome from baseline scans could help the radiation oncologist to determine the best treatment strategies and to stratify patients as at low risk or high risk of recurrence. Nuclear medicine molecular imaging reflects information regarding biological processes in the tumor thanks to a wide range of radiotracers. Many studies involving ¹⁸F-fluorodeoxyglucose positron emission tomography suggest an added value of radiomics compared with the use of conventional PET metrics such as standardized uptake value for both tumor diagnosis and prediction of recurrence or treatment outcome. However, these promising results should not hide technical difficulties that still currently prevent the approach from being widely studied or clinically used. These difficulties mostly pertain to the variability of the imaging features as a function of the acquisition device and protocol, the robustness of the models with respect to that variability, and the interpretation of the radiomic models. Addressing the impact of the variability in acquisition and reconstruction protocols is needed, as is harmonizing the radiomic feature calculation methods, to ensure the reproducibility of studies in a multicenter context and their implementation in a clinical workflow. In this review, we explain the potential impact of positron emission tomography radiomics for radiation therapy and underline the various aspects that need to be carefully addressed to make the most of this promising approach.

Shell feature: a new radiomics descriptor for predicting distant failure after radiotherapy in non-small cell lung cancer and cervix cancer

Distant failure is the main cause of human cancer-related mortalities. To develop a model for predicting distant failure in non-small cell lung cancer (NSCLC) and cervix cancer (CC) patients, a shell feature, consisting of outer voxels around the tumor boundary, was constructed using pre-treatment positron emission tomography (PET) images from 48 NSCLC patients received stereotactic body radiation therapy and 52 CC patients underwent external beam radiation therapy and concurrent chemotherapy followed with high-dose-rate intracavitary brachytherapy. The hypothesis behind this feature is that non-invasive and invasive tumors may have different morphologic patterns in the tumor periphery, in turn reflecting the differences in radiological presentations in the PET images. The utility of the shell was evaluated by the support vector machine classifier in comparison with intensity, geometry, gray level co-occurrence matrix-based texture, neighborhood gray tone difference matrix-based texture, and a combination of these four features. The results were assessed in terms of accuracy, sensitivity, specificity, and AUC. Collectively, the shell feature showed better predictive performance than all the other features for distant failure prediction in both NSCLC and CC cohorts.