A Deep Learning Model for Predicting Xerostomia Due to Radiation Therapy for Head and Neck Squamous Cell Carcinoma in the RTOG 0522 Clinical Trial

Imaging Assessment of Radiation Therapy-Related Normal Tissue Injury in Children: A PENTEC Visionary Statement

The Pediatric Normal Tissue Effects in the Clinic (PENTEC) consortium has made significant contributions to understanding and mitigating the adverse effects of childhood cancer therapy. This review addresses the role of diagnostic imaging in detecting, screening, and comprehending radiation therapy-related late effects in children, drawing insights from individual organ-specific PENTEC reports. We further explore how the development of imaging biomarkers for key organ systems, alongside technical advancements and translational imaging approaches, may enhance the systematic application of imaging evaluations in childhood cancer survivors. Moreover, the review critically examines knowledge gaps and identifies technical and practical limitations of existing imaging modalities in the pediatric population. Addressing these challenges may expand access to, minimize the risk of, and optimize the real-world application of, new imaging techniques. The PENTEC team envisions this document as a roadmap for the future development of imaging strategies in childhood cancer survivors, with the overarching goal of improving long-term health outcomes and quality of life for this vulnerable population.

Machine learning in the evaluation of voice and swallowing in the head and neck cancer patient

PURPOSE OF REVIEW

The purpose of this review is to present recent advances and limitations in machine learning applied to the evaluation of speech, voice, and swallowing in head and neck cancer.

RECENT FINDINGS

Novel machine learning models incorporating diverse data modalities with improved discriminatory capabilities have been developed for predicting toxicities following head and neck cancer therapy, including dysphagia, dysphonia, xerostomia, and weight loss as well as guiding treatment planning. Machine learning has been applied to the care of posttreatment voice and swallowing dysfunction by offering objective and standardized assessments and aiding innovative technologies for functional restoration. Voice and speech are also being utilized in machine learning algorithms to screen laryngeal cancer.

SUMMARY

Machine learning has the potential to help optimize, assess, predict, and rehabilitate voice and swallowing function in head and neck cancer patients as well as aid in cancer screening. However, existing studies are limited by the lack of sufficient external validation and generalizability, insufficient transparency and reproducibility, and no clear superior predictive modeling strategies. Algorithms and applications will need to be trained on large multiinstitutional data sets, incorporate sociodemographic data to reduce bias, and achieve validation through clinical trials for optimal performance and utility.

Machine Learning Approaches to Predict Symptoms in People With Cancer: Systematic Review

BACKGROUND

People with cancer frequently experience severe and distressing symptoms associated with cancer and its treatments. Predicting symptoms in patients with cancer continues to be a significant challenge for both clinicians and researchers. The rapid evolution of machine learning (ML) highlights the need for a current systematic review to improve cancer symptom prediction.

OBJECTIVE

This systematic review aims to synthesize the literature that has used ML algorithms to predict the development of cancer symptoms and to identify the predictors of these symptoms. This is essential for integrating new developments and identifying gaps in existing literature.

METHODS

We conducted this systematic review in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) checklist. We conducted a systematic search of CINAHL, Embase, and PubMed for English records published from 1984 to August 11, 2023, using the following search terms: cancer, neoplasm, specific symptoms, neural networks, machine learning, specific algorithm names, and deep learning. All records that met the eligibility criteria were individually reviewed by 2 coauthors, and key findings were extracted and synthesized. We focused on studies using ML algorithms to predict cancer symptoms, excluding nonhuman research, technical reports, reviews, book chapters, conference proceedings, and inaccessible full texts.

RESULTS

A total of 42 studies were included, the majority of which were published after 2017. Most studies were conducted in North America (18/42, 43%) and Asia (16/42, 38%). The sample sizes in most studies (27/42, 64%) typically ranged from 100 to 1000 participants. The most prevalent category of algorithms was supervised ML, accounting for 39 (93%) of the 42 studies. Each of the methods-deep learning, ensemble classifiers, and unsupervised ML-constituted 3 (3%) of the 42 studies. The ML algorithms with the best performance were logistic regression (9/42, 17%), random forest (7/42, 13%), artificial neural networks (5/42, 9%), and decision trees (5/42, 9%). The most commonly included primary cancer sites were the head and neck (9/42, 22%) and breast (8/42, 19%), with 17 (41%) of the 42 studies not specifying the site. The most frequently studied symptoms were xerostomia (9/42, 14%), depression (8/42, 13%), pain (8/42, 13%), and fatigue (6/42, 10%). The significant predictors were age, gender, treatment type, treatment number, cancer site, cancer stage, chemotherapy, radiotherapy, chronic diseases, comorbidities, physical factors, and psychological factors.

CONCLUSIONS

This review outlines the algorithms used for predicting symptoms in individuals with cancer. Given the diversity of symptoms people with cancer experience, analytic approaches that can handle complex and nonlinear relationships are critical. This knowledge can pave the way for crafting algorithms tailored to a specific symptom. In addition, to improve prediction precision, future research should compare cutting-edge ML strategies such as deep learning and ensemble methods with traditional statistical models.

Using meta-analysis and CNN-NLP to review and classify the medical literature for normal tissue complication probability in head and neck cancer

PURPOSE

The study aims to enhance the efficiency and accuracy of literature reviews on normal tissue complication probability (NTCP) in head and neck cancer patients using radiation therapy. It employs meta-analysis (MA) and natural language processing (NLP).

MATERIAL AND METHODS

The study consists of two parts. First, it employs MA to assess NTCP models for xerostomia, dysphagia, and mucositis after radiation therapy, using Python 3.10.5 for statistical analysis. Second, it integrates NLP with convolutional neural networks (CNN) to optimize literature search, reducing 3256 articles to 12. CNN settings include a batch size of 50, 50-200 epoch range and a 0.001 learning rate.

RESULTS

The study's CNN-NLP model achieved a notable accuracy of 0.94 after 200 epochs with Adamax optimization. MA showed an AUC of 0.67 for early-effect xerostomia and 0.74 for late-effect, indicating moderate to high predictive accuracy but with high variability across studies. Initial CNN accuracy of 66.70% improved to 94.87% post-tuning by optimizer and hyperparameters.

CONCLUSION

The study successfully merges MA and NLP, confirming high predictive accuracy for specific model-feature combinations. It introduces a time-based metric, words per minute (WPM), for efficiency and highlights the utility of MA and NLP in clinical research.

"Under the hood": artificial intelligence in personalized radiotherapy

This review presents and discusses the ways in which artificial intelligence (AI) tools currently intervene, or could potentially intervene in the future, to enhance the diverse tasks involved in the radiotherapy workflow. The radiotherapy framework is presented on 2 different levels for the personalization of the treatment, distinct in tasks and methodologies. The first level is the clinically well-established anatomy-based workflow, known as adaptive radiation therapy. The second level is referred to as biology-driven workflow, explored in the research literature and recently appearing in some preliminary clinical trials for personalized radiation treatments. A 2-fold role for AI is defined according to these 2 different levels. In the anatomy-based workflow, the role of AI is to streamline and improve the tasks in terms of time and variability reductions compared to conventional methodologies. The biology-driven workflow instead fully relies on AI, which introduces decision-making tools opening uncharted frontiers that were in the past deemed challenging to explore. These methodologies are referred to as radiomics and dosiomics, handling imaging and dosimetric information, or multiomics, when complemented by clinical and biological parameters (ie, biomarkers). The review explicitly highlights the methodologies that are currently incorporated into clinical practice or still in research, with the aim of presenting the AI's growing role in personalized radiotherapy.

Predicting nodal metastases in squamous cell carcinoma of the oral tongue using artificial intelligence

OBJECTIVE

The presence of occult nodal metastases in patients with squamous cell carcinoma (SCC) of the oral tongue has implications for treatment. Upwards of 30% of patients will have occult nodal metastases, yet a significant number of patients undergo unnecessary neck dissection to confirm nodal status. This study sought to predict the presence of nodal metastases in patients with SCC of the oral tongue using a convolutional neural network (CNN) that analyzed visual histopathology from the primary tumor alone.

METHODS

Cases of SCC of the oral tongue were identified from the records of a single institution. Only patients with complete pathology data were included in the study. The primary tumors were randomized into 2 groups for training and testing, which was performed at 2 different levels of supervision. Board-certified pathologists annotated each slide. HALO-AI convolutional neural network and image software was used to perform training and testing. Receiver operator characteristic (ROC) curves and the Youden J statistic were used for primary analysis.

RESULTS

Eighty-nine cases of SCC of the oral tongue were included in the study. The best performing algorithm had a high level of supervision and a sensitivity of 65% and specificity of 86% when identifying nodal metastases. The area under the curve (AUC) of the ROC curve for this algorithm was 0.729.

CONCLUSION

A CNN can produce an algorithm that is able to predict nodal metastases in patients with squamous cell carcinoma of the oral tongue by analyzing the visual histopathology of the primary tumor alone.

NTCP modelling of xerostomia after radiotherapy for oropharyngeal cancer using the PRO-CTCAE and CTCAE scoring systems at different time-points post-RT

PURPOSE

This study aims at determining the parameter values of three normal tissue complication probability (NTCP) models for the contralateral parotid gland, contralateral submandibular gland (SMG) and contralateral salivary glands regarding the endpoint of xerostomia 6-24 months after radiotherapy for oropharynx cancer.

METHODS

The treatment and outcome data of 231 patients with favorable risk, HPV-associated oropharyngeal squamous cell carcinoma are analyzed. 60 Gy intensity modulated radiotherapy was delivered to all the patients. The presence and severity of xerostomia was recorded (pre- and post- radiotherapy) by the PRO-CTCAE and the CTCAE scoring systems. In both scoring systems, patients with a change in symptom severity (from baseline) of ≥ 2 were considered responders.

RESULTS

Xerostomia was observed in 61.3 %, 39.2 %, 28.6 % and 27.0 % of the patients based on the PRO-CTCAE scoring system at 6-, 12-, 18- and 24-months post-RT, respectively. The AUCs of the contralateral salivary glands ranged between 0.58-0.64 in the LKB model with the gEUD ranging between 20.3 Gy and 24.7 Gy.

CONCLUSIONS

Based on the PRO-CTCAE scores, mean dose < 22 Gy, V50 < 10 % for the contralateral salivary glands and mean dose < 18 Gy, V45 < 10 % for the contralateral parotid were found to significantly reduce by a factor of 2-3 the risk for radiation induced xerostomia that is observed at 6-24 months post-RT, respectively. Also, gEUD < 22 Gy to the contralateral salivary glands and < 18 Gy to the contralateral parotid was found to significantly reduce the risk for radiation induced xerostomia that is observed at 6-24 months post-RT by 2.0-2.3 times.

Radiation pneumonia predictive model for radiotherapy in esophageal carcinoma patients

BACKGROUND

The machine learning models with dose factors and the deep learning models with dose distribution matrix have been used to building lung toxics models for radiotherapy and achieve promising results. However, few studies have integrated clinical features into deep learning models. This study aimed to explore the role of three-dimension dose distribution and clinical features in predicting radiation pneumonitis (RP) in esophageal cancer patients after radiotherapy and designed a new hybrid deep learning network to predict the incidence of RP.

METHODS

A total of 105 esophageal cancer patients previously treated with radiotherapy were enrolled in this study. The three-dimension (3D) dose distributions within the lung were extracted from the treatment planning system, converted into 3D matrixes and used as inputs to predict RP with ResNet. In total, 15 clinical factors were normalized and converted into one-dimension (1D) matrixes. A new prediction model (HybridNet) was then built based on a hybrid deep learning network, which combined 3D ResNet18 and 1D convolution layers. Machine learning-based prediction models, which use the traditional dosiomic factors with and without the clinical factors as inputs, were also constructed and their predictive performance compared with that of HybridNet using tenfold cross validation. Accuracy and area under the receiver operator characteristic curve (AUC) were used to evaluate the model effect. DeLong test was used to compare the prediction results of the models.

RESULTS

The deep learning-based model achieved superior prediction results compared with machine learning-based models. ResNet performed best in the group that only considered dose factors (accuracy, 0.78 ± 0.05; AUC, 0.82 ± 0.25), whereas HybridNet performed best in the group that considered both dose factors and clinical factors (accuracy, 0.85 ± 0.13; AUC, 0.91 ± 0.09). HybridNet had higher accuracy than that of Resnet (p = 0.009).

CONCLUSION

Based on prediction results, the proposed HybridNet model could predict RP in esophageal cancer patients after radiotherapy with significantly higher accuracy, suggesting its potential as a useful tool for clinical decision-making. This study demonstrated that the information in dose distribution is worth further exploration, and combining multiple types of features contributes to predict radiotherapy response.

Enhanced prediction of postoperative radiotherapy-induced esophagitis in non-small cell lung cancer: Dosiomic model development in a real-world cohort and validation in the PORT-C randomized controlled trial

BACKGROUND

Radiotherapy-induced esophagitis (RE) diminishes the quality of life and interrupts treatment in patients with non-small cell lung cancer (NSCLC) undergoing postoperative radiotherapy. Dosimetric models showed limited capability in predicting RE. We aimed to develop dosiomic models to predict RE.

METHODS

Models were trained with a real-world cohort and validated with PORT-C randomized controlled trial cohort. Patients with NSCLC undergoing resection followed by postoperative radiotherapy between 2004 and 2015 were enrolled. The endpoint was grade ≥2 RE. Esophageal three-dimensional dose distribution features were extracted using handcrafted and convolutional neural network (CNN) methods, screened using an entropy-based method, and selected using minimum redundancy and maximum relevance. Prediction models were built using logistic regression. The areas under the receiver operating characteristic curve (AUC) and precision-recall curve were used to evaluate prediction model performance. A dosimetric model was built for comparison.

RESULTS

A total of 190 and 103 patients were enrolled in the training and validation sets, respectively. Using handcrafted and CNN methods, 107 and 4096 features were derived, respectively. Three handcrafted, four CNN-extracted and three dosimetric features were selected. AUCs of training and validation sets were 0.737 and 0.655 for the dosimetric features, 0.730 and 0.724 for handcrafted features, and 0.812 and 0.785 for CNN-extracted features, respectively. Precision-recall curves revealed that CNN-extracted features outperformed dosimetric and handcrafted features.

CONCLUSIONS

Prediction models may identify patients at high risk of developing RE. Dosiomic models outperformed the dosimetric-feature model in predicting RE. CNN-extracted features were more predictive but less interpretable than handcrafted features.

CT-based deep multi-label learning prediction model for outcome in patients with oropharyngeal squamous cell carcinoma

BACKGROUND

Personalized treatment is increasingly required for oropharyngeal squamous cell carcinoma (OPSCC) patients due to emerging new cancer subtypes and treatment options. Outcome prediction model can help identify low or high-risk patients who may be suitable to receive de-escalation or intensified treatment approaches.

PURPOSE

To develop a deep learning (DL)-based model for predicting multiple and associated efficacy endpoints in OPSCC patients based on computed tomography (CT).

METHODS

Two patient cohorts were used in this study: a development cohort consisting of 524 OPSCC patients (70% for training and 30% for independent testing) and an external test cohort of 396 patients. Pre-treatment CT-scans with the gross primary tumor volume contours (GTVt) and clinical parameters were available to predict endpoints, including 2-year local control (LC), regional control (RC), locoregional control (LRC), distant metastasis-free survival (DMFS), disease-specific survival (DSS), overall survival (OS), and disease-free survival (DFS). We proposed DL outcome prediction models with the multi-label learning (MLL) strategy that integrates the associations of different endpoints based on clinical factors and CT-scans.

RESULTS

The multi-label learning models outperformed the models that were developed based on a single endpoint for all endpoints especially with high AUCs ≥ 0.80 for 2-year RC, DMFS, DSS, OS, and DFS in the internal independent test set and for all endpoints except 2-year LRC in the external test set. Furthermore, with the models developed, patients could be stratified into high and low-risk groups that were significantly different for all endpoints in the internal test set and for all endpoints except DMFS in the external test set.

CONCLUSION

MLL models demonstrated better discriminative ability for all 2-year efficacy endpoints than single outcome models in the internal test and for all endpoints except LRC in the external set.

Prediction of toxicity outcomes following radiotherapy using deep learning-based models: A systematic review

PURPOSE

This study aims to perform a comprehensive systematic review of deep learning (DL) models in predicting RT-induced toxicity.

MATERIALS AND METHODS

A literature review was performed following the PRISMA guidelines. Systematic searches were performed in PubMed, Scopus, Cochrane and Embase databases from the earliest record up to September 2022. Related studies on deep learning models for radiotherapy toxicity prediction were selected based on predefined PICOS criteria.

RESULTS

Fourteen studies of radiotherapy-treated patients on different types of cancer [prostate (n=2), HNC (n=4), liver (n=2), lung (n=4), cervical (n=1), and oesophagus (n=1)] were eligible for inclusion in the systematic review. Information regarding patient characteristics and model development was summarized. Several approaches, such as ensemble learning, data augmentation, and transfer learning, that were utilized by selected studies were discussed.

CONCLUSION

Deep learning techniques are able to produce a consistent performance for toxicity prediction. Future research using large and diverse datasets and standardization of the study methodologies are required to improve the consistency of the research output.

Multi-organ spatial stratification of 3-D dose distributions improves risk prediction of long-term self-reported severe symptoms in oropharyngeal cancer patients receiving radiotherapy: development of a pre-treatment decision support tool

Purpose

Identify Oropharyngeal cancer (OPC) patients at high-risk of developing long-term severe radiation-associated symptoms using dose volume histograms for organs-at-risk, via unsupervised clustering.

Material and methods

All patients were treated using radiation therapy for OPC. Dose-volume histograms of organs-at-risk were extracted from patients' treatment plans. Symptom ratings were collected via the MD Anderson Symptom Inventory (MDASI) given weekly during, and 6 months post-treatment. Drymouth, trouble swallowing, mucus, and vocal dysfunction were selected for analysis in this study. Patient stratifications were obtained by applying Bayesian Mixture Models with three components to patient's dose histograms for relevant organs. The clusters with the highest total mean doses were translated into dose thresholds using rule mining. Patient stratifications were compared against Tumor staging information using multivariate likelihood ratio tests. Model performance for prediction of moderate/severe symptoms at 6 months was compared against normal tissue complication probability (NTCP) models using cross-validation.

Results

A total of 349 patients were included for long-term symptom prediction. High-risk clusters were significantly correlated with outcomes for severe late drymouth (p <.0001, OR = 2.94), swallow (p = .002, OR = 5.13), mucus (p = .001, OR = 3.18), and voice (p = .009, OR = 8.99). Simplified clusters were also correlated with late severe symptoms for drymouth (p <.001, OR = 2.77), swallow (p = .01, OR = 3.63), mucus (p = .01, OR = 2.37), and voice (p <.001, OR = 19.75). Proposed cluster stratifications show better performance than NTCP models for severe drymouth (AUC.598 vs.559, MCC.143 vs.062), swallow (AUC.631 vs.561, MCC.20 vs -.030), mucus (AUC.596 vs.492, MCC.164 vs -.041), and voice (AUC.681 vs.555, MCC.181 vs -.019). Simplified dose thresholds also show better performance than baseline models for predicting late severe ratings for all symptoms.

Conclusion

Our results show that leveraging the 3-D dose histograms from radiation therapy plan improves stratification of patients according to their risk of experiencing long-term severe radiation associated symptoms, beyond existing NTPC models. Our rule-based method can approximate our stratifications with minimal loss of accuracy and can proactively identify risk factors for radiation-associated toxicity.

Application of Platelet-Rich Fibrin and Bone Morphogenetic Protein for Full-Mouth Implant-Based Oral Rehabilitation in a Case of Mandibular Osteoradionecrosis

Osteoradionecrosis (ORN) is a debilitating complication following radiation therapy, which in the head and neck region, occurs most frequently in the mandible. Although ORN is rare, it is complex and multifactorial and requires appropriate management. Manipulation of bone in patients with head and neck cancers before radiotherapy can cause ORN. In this report, we aim to present successful insertion of four dental implants in the interforaminal segment combined with application of platelet-rich fibrin and bone morphogenetic protein in a 60-year-old male with stable ORN in the posterior regions of the mandible.

Development and validation of a nomogram for prediction of recovery from moderate-severe xerostomia post-radiotherapy in nasopharyngeal carcinoma patients

PURPOSE

Aim to create and validate a comprehensive nomogram capable of accurately predicting the transition from moderate-severe to normal-mild xerostomia post-radiotherapy (postRT) in patients with nasopharyngeal carcinoma (NPC). Materials and methods We constructed and internally verified a prediction model using a primary cohort comprising 223 patients who were pathologically diagnosed with NPC from February 2016 to December 2019. LASSO regression model was used to identify the clinical factors and relevant variables (the pre-radiotherapy (XQ-preRT) and immediate post-radiotherapy (XQ-postRT) xerostomia questionnaire scores, as well as the mean dose (D_mean) delivered to the parotid gland (PG), submandibular gland (SMG), sublingual gland (SLG), tubarial gland (TG), and oral cavity). Cox proportional hazards regression analysis was performed to develop the prediction model, which was presented as a nomogram. The models' performance with regard to calibration, discrimination, and clinical usefulness was evaluated. The external validation cohort comprised 78 patients. Results Due to better discrimination and calibration in the training cohort, age, gender, XQ-postRT, and D_mean of PG, SMG, and TG were included in the individualized prediction model (C-index of 0.741 (95% CI:0.717 to 0.765). Verification of the nomogram's performance in internal and external validation cohorts revealed good discrimination (C-index of 0.729 (0.692 to 0.766) and 0.736 (0.702 to 0.770), respectively) and calibration. Decision curve analysis revealed that the nomogram was clinically useful. The 12-month and 24-month moderate-severe xerostomia rate was statistically lower in the SMG-spared arm (28.4% (0.230 to 35.2) and 5.2% (0.029 to 0.093), respectively) than that in SMG-unspared arm (56.8% (0.474 to 0.672) and 12.5% (0.070 to 0.223), respectively), with an HR of 1.84 (95%CI: 1.412 to 2.397, p= 0.000). The difference in restricted mean survival time for remaining moderate-severe xerostomia between the two arms at 24 months was 5.757 months (95% CI, 3.863 to 7.651; p=0.000). Conclusion The developed nomogram, incorporating age, gender, XQ-postRT, and D_mean to PG, SMG, and TG, can be used for predicting recovery from moderate-severe xerostomia post-radiotherapy in NPC patients. Sparing SMG is highly important for the patient's recovery.

Machine learning for the prediction of toxicities from head and neck cancer treatment: A systematic review with meta-analysis

INTRODUCTION

The aim of the present systematic review (SR) is to summarize Machine Learning (ML) models currently used to predict head and neck cancer (HNC) treatment-related toxicities, and to understand the impact of image biomarkers (IBMs) in prediction models (PMs). The present SR was conducted following the guidelines of the PRISMA 2022 and registered in PROSPERO database (CRD42020219304).

METHODS

The acronym PICOS was used to develop the focused review question (Can PMs accurately predict HNC treatment toxicities?) and the eligibility criteria. The inclusion criteria enrolled Prediction Model Studies (PMSs) with patient cohorts that were treated for HNC and developed toxicities. Electronic database search encompassed PubMed, EMBASE, Scopus, Cochrane Library, Web of Science, LILACS, and Gray Literature (Google Scholar and ProQuest). Risk of Bias (RoB) was assessed through PROBAST and the results were synthesized based on the data format (with and without IBMs) to allow comparison.

RESULTS

A total of 28 studies and 4,713 patients were included. Xerostomia was the most frequently investigated toxicity (17; 60.71 %). Sixteen (57.14 %) studies reported using radiomics features in combination with clinical or dosimetrics/dosiomics for modelling. High RoB was identified in 23 studies. Meta-analysis (MA) showed an area under the receiver operating characteristics curve (AUROC) of 0.82 for models with IBMs and 0.81 for models without IBMs (p value < 0.001), demonstrating no difference among IBM- and non-IBM-based models.

DISCUSSION

The development of a PM based on sample-specific features represents patient selection bias and may affect a model's performance. Heterogeneity of the studies as well as non-standardized metrics prevent proper comparison of studies, and the absence of an independent/external test does not allow the evaluation of the model's generalization ability.

CONCLUSION

IBM-featured PMs are not superior to PMs based on non-IBM predictors. The evidence was appraised as of low certainty.

Remote assessment of cognition and quality of life following radiotherapy for nasopharyngeal carcinoma: deep-learning-based predictive models and MRI correlates

PURPOSE

Irradiation of the brain regions from nasopharyngeal carcinoma (NPC) radiotherapy (RT) is frequently unavoidable, which may result in radiation-induced cognitive deficit. Using deep learning (DL), the study aims to develop prediction models in predicting compromised cognition in patients following NPC RT using remote assessments and determine their relation to the quality of life (QoL) and MRI changes.

METHODS

Seventy patients (20-76 aged) with MRI imaging (pre- and post-RT (6 months-1 year)) and complete cognitive assessments were recruited. Hippocampus, temporal lobes (TLs), and cerebellum were delineated and dosimetry parameters were extracted. Assessments were given post-RT via telephone (Telephone Interview Cognitive Status (TICS), Telephone Montreal Cognitive Assessment (T-MoCA), Telephone Mini Addenbrooke's Cognitive Examination (Tele-MACE), and QLQ-H&N 43). Regression and deep neural network (DNN) models were used to predict post-RT cognition using anatomical and treatment dose features.

RESULTS

Remote cognitive assessments were inter-correlated (r > 0.9). TLs showed significance in pre- and post-RT volume differences and cognitive deficits, that are correlated with RT-associated volume atrophy and dose distribution. Good classification accuracy based on DNN area under receiver operating curve (AUROC) for cognitive prediction (T-MoCA AUROC = 0.878, TICS AUROC = 0.89, Tele-MACE AUROC = 0.919).

CONCLUSION

DL-based prediction models assessed using remote assessments can assist in predicting cognitive deficit following NPC RT. Comparable results of remote assessments in assessing cognition suggest its possibility in replacing standard assessments.

IMPLICATIONS FOR CANCER SURVIVORS

Application of prediction models in individual patient enables tailored interventions to be provided in managing cognitive changes following NPC RT.

Deep learning-based precise prediction and early detection of radiation-induced temporal lobe injury for nasopharyngeal carcinoma

Background

Radiotherapy is the mainstay of treatment for nasopharyngeal carcinoma. Radiation-induced temporal lobe injury (TLI) can regress or resolve in the early phase, but it is irreversible at a later stage. However, no study has proposed a risk-based follow-up schedule for its early detection. Planning evaluation is difficult when dose-volume histogram (DVH) parameters are similar and optimization is terminated.

Methods

This multicenter retrospective study included 6065 patients between 2014 and 2018. A 3D ResNet-based deep learning model was developed in training and validation cohorts and independently tested using concordance index in internal and external test cohorts. Accordingly, the patients were stratified into risk groups, and the model-predicted risks were used to develop risk-based follow-up schedules. The schedule was compared with the Radiation Therapy Oncology Group (RTOG) recommendation (every 3 months during the first 2 years and every 6 months in 3-5 years). Additionally, the model was used to evaluate plans with similar DVH parameters.

Findings

Our model achieved concordance indexes of 0.831, 0.818, and 0.804, respectively, which outperformed conventional prediction models (all P < 0.001). The temporal lobes in all the cohorts were stratified into three groups with discrepant TLI-free survival. Personalized follow-up schedules developed for each risk group could detect TLI 1.9 months earlier than the RTOG recommendation. According to a higher median predicted 3-year TLI-free survival (99.25% vs. 99.15%, P < 0.001), the model identified a better plan than previous models.

Interpretation

The deep learning model predicted TLI more precisely. The model-determined risk-based follow-up schedule detected the TLI earlier. The planning evaluation was refined because the model identified a better plan with a lower risk of TLI.

Funding

The Sun Yat-sen University Clinical Research 5010 Program (2015020), Guangdong Basic and Applied Basic Research Foundation (2022A1515110356), Medical Scientific Research Foundation of Guangdong Province (A2022367), and Guangzhou Science and Technology Program (2023A04J1788).

Artificial Intelligence and Laryngeal Cancer: From Screening to Prognosis: A State of the Art Review

OBJECTIVE

This state of the art review aims to examine contemporary advances in applications of artificial intelligence (AI) to the screening, detection, management, and prognostication of laryngeal cancer (LC).

DATA SOURCES

Four bibliographic databases were searched: PubMed, EMBASE, Cochrane, and IEEE.

REVIEW METHODS

A structured review of the current literature (up to January 2022) was performed. Search terms related to topics of AI in LC were identified and queried by 2 independent reviewers. Citations of selected studies and review articles were also evaluated to ensure comprehensiveness.

CONCLUSIONS

AI applications in LC have encompassed a variety of data modalities, including radiomics, genomics, acoustics, clinical data, and videomics, to support screening, diagnosis, therapeutic decision making, and prognosis. However, most studies remain at the proof-of-concept level, as AI algorithms are trained on single-institution databases with limited data sets and a single data modality.

IMPLICATIONS FOR PRACTICE

AI algorithms in LC will need to be trained on large multi-institutional data sets and integrate multimodal data for optimal performance and clinical utility from screening to prognosis. Out of the data types reviewed, genomics has the most potential to provide generalizable models thanks to available large multi-institutional open access genomic data sets. Voice acoustic data represent an inexpensive and accurate biomarker, which is easy and noninvasive to capture, offering a unique opportunity for screening and monitoring of LA, especially in low-resource settings.

Toxicity Prediction in Pelvic Radiotherapy Using Multiple Instance Learning and Cascaded Attention Layers

Modern radiotherapy delivers treatment plans optimised on an individual patient level, using CT-based 3D models of patient anatomy. This optimisation is fundamentally based on simple assumptions about the relationship between radiation dose delivered to the cancer (increased dose will increase cancer control) and normal tissue (increased dose will increase rate of side effects). The details of these relationships are still not well understood, especially for radiation-induced toxicity. We propose a convolutional neural network based on multiple instance learning to analyse toxicity relationships for patients receiving pelvic radiotherapy. A dataset comprising of 315 patients were included in this study; with 3D dose distributions, pre-treatment CT scans with annotated abdominal structures, and patient-reported toxicity scores provided for each participant. In addition, we propose a novel mechanism for segregating the attentions over space and dose/imaging features independently for a better understanding of the anatomical distribution of toxicity. Quantitative and qualitative experiments were performed to evaluate the network performance. The proposed network could predict toxicity with 80% accuracy. Attention analysis over space demonstrated that there was a significant association between radiation dose to the anterior and right iliac of the abdomen and patient-reported toxicity. Experimental results showed that the proposed network had outstanding performance for toxicity prediction, localisation and explanation with the ability of generalisation for an unseen dataset.

Contouring the accessory parotid gland and major parotid glands as a single organ at risk during nasopharyngeal carcinoma radiotherapy

Background and purpose

No research currently exists on the role of the accessory parotid gland (APG) in nasopharyngeal carcinoma (NPC). We thereby aimed to assess the effects of APG on the dosimetry of the parotid glands (PGs) during NPC radiotherapy and evaluate its predictive value for late xerostomia.

Material and methods

The clinical data of 32 NPC patients with radiological evidence of the APG treated at Sun Yat-sen Memorial Hospital between November 2020 and February 2021 were retrospectively reviewed. Clinically approved treatment plans consisted of only the PGs as an organ at risk (OAR) (Plan1), while Plan2 was designed by considering the APG as a single organ at risk (OAR). The APG on Plan1 was delineated, and dose–volume parameters of the PGs alone (PG-only) and of the combined structure (PG+APG) were analyzed in both plans. The association of such dosimetric parameters in Plan1 with xerostomia at 6–9 months post-radiotherapy was further explored.

Results

Fifty APGs were found, with a mean volume of 3.3 ± 0.2 ml. Significant differences were found in all dosimetric parameters between Plan1 and Plan2. The mean dose and percentage of OAR volumes receiving more than 30 Gy significantly reduced in Plan1 itself (PG-only vs. PG+APG, 39.55 ± 0.83 Gy vs. 37.71 ± 0.75 Gy, and 62.00 ± 2.00% vs. 57.41 ± 1.56%, respectively; p &lt; 001) and reduced further in Plan2 (PG+APG, 36.40 ± 0.74 Gy, and 55.54 ± 1.61%, respectively; p &lt; 0.001). Three additional patients met the dose constraint in Plan1, which increased to seven in Plan2. With APG included, the predictive power of the dosimetric parameters for xerostomia tended to improve, although no significant differences were observed.

Conclusion

APG is anatomically similar to the PGs. Our findings suggest the potential benefits of treating the APG and PGs as a single OAR during radiotherapy (RT) of NPC by improving PG sparing.

Artificial intelligence for multimodal data integration in oncology

In oncology, the patient state is characterized by a whole spectrum of modalities, ranging from radiology, histology, and genomics to electronic health records. Current artificial intelligence (AI) models operate mainly in the realm of a single modality, neglecting the broader clinical context, which inevitably diminishes their potential. Integration of different data modalities provides opportunities to increase robustness and accuracy of diagnostic and prognostic models, bringing AI closer to clinical practice. AI models are also capable of discovering novel patterns within and across modalities suitable for explaining differences in patient outcomes or treatment resistance. The insights gleaned from such models can guide exploration studies and contribute to the discovery of novel biomarkers and therapeutic targets. To support these advances, here we present a synopsis of AI methods and strategies for multimodal data fusion and association discovery. We outline approaches for AI interpretability and directions for AI-driven exploration through multimodal data interconnections. We examine challenges in clinical adoption and discuss emerging solutions.

Predicting radiotherapy-induced xerostomia in head and neck cancer patients using day-to-day kinetics of radiomics features

Background and purpose

The images acquired during radiotherapy for image-guidance purposes could be used to monitor patient-specific response to irradiation and improve treatment personalisation. We investigated whether the kinetics of radiomics features from daily mega-voltage CT image-guidance scans (MVCT) improve prediction of moderate-to-severe xerostomia compared to dose/volume parameters in radiotherapy of head-and-neck cancer (HNC).

Materials and Methods

All included HNC patients (N = 117) received 30 or more fractions of radiotherapy with daily MVCTs. Radiomics features were calculated on the contra-lateral parotid glands of daily MVCTs. Their variations over time after each complete week of treatment were used to predict moderate-to-severe xerostomia (CTCAEv4.03 grade ≥ 2) at 6, 12 and 24 months post-radiotherapy. After dimensionality reduction, backward/forward selection was used to generate combinations of predictors.Three types of logistic regression model were generated for each follow-up time: 1) a pre-treatment reference model using dose/volume parameters, 2) a combination of dose/volume and radiomics-based predictors, and 3) radiomics-based predictors. The models were internally validated by cross-validation and bootstrapping and their performance evaluated using Area Under the Curve (AUC) on separate training and testing sets.

Results

Moderate-to-severe xerostomia was reported by 46 %, 33 % and 26 % of the patients at 6, 12 and 24 months respectively. The selected models using radiomics-based features extracted at or before mid-treatment outperformed the dose-based models with an AUCtrain/AUCtest of 0.70/0.69, 0.76/0.74, 0.86/0.86 at 6, 12 and 24 months, respectively.

Conclusion

Our results suggest that radiomics features calculated on MVCTs from the first half of the radiotherapy course improve prediction of moderate-to-severe xerostomia in HNC patients compared to a dose-based pre-treatment model.

Transfer learning approach based on computed tomography images for predicting late xerostomia after radiotherapy in patients with oropharyngeal cancer

Background and purpose

Although the latest breakthroughs in radiotherapy (RT) techniques have led to a decrease in adverse event rates, these techniques are still associated with substantial toxicity, including xerostomia. Imaging biomarkers could be useful to predict the toxicity risk related to each individual patient. Our preliminary work aims to develop a radiomic-based support tool exploiting pre-treatment CT images to predict late xerostomia risk in 3 months after RT in patients with oropharyngeal cancer (OPC).

Materials and methods

We performed a multicenter data collection. We enrolled 61 patients referred to three care centers in Apulia, Italy, out of which 22 patients experienced at least mild xerostomia 3 months after the end of the RT cycle. Pre-treatment CT images, clinical and dose features, and alcohol-smoking habits were collected. We proposed a transfer learning approach to extract quantitative imaging features from CT images by means of a pre-trained convolutional neural network (CNN) architecture. An optimal feature subset was then identified to train an SVM classifier. To evaluate the robustness of the proposed model with respect to different manual contouring practices on CTs, we repeated the same image analysis pipeline on “fake” parotid contours.

Results

The best performances were achieved by the model exploiting the radiomic features alone. On the independent test, the model reached median AUC, accuracy, sensitivity, and specificity values of 81.17, 83.33, 71.43, and 90.91%, respectively. The model was robust with respect to diverse manual parotid contouring procedures.

Conclusion

Radiomic analysis could help to develop a valid support tool for clinicians in planning radiotherapy treatment, by providing a risk score of the toxicity development for each individual patient, thus improving the quality of life of the same patient, without compromising patient care.

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.

Extending the relative seriality formalism for interpretable deep learning of normal tissue complication probability models

We formally demonstrate that the relative seriality (RS) model of normal tissue complication probability (NTCP) can be recast as a simple neural network with one convolutional and one pooling layer. This approach enables us to systematically construct deep relative seriality networks (DRSNs), a new class of mechanistic generalizations of the RS model with radiobiologically interpretable parameters amenable to deep learning. To demonstrate the utility of this formulation, we analyze a simplified example of xerostomia due to irradiation of the parotid gland during alpha radiopharmaceutical therapy. Using a combination of analytical calculations and numerical simulations, we show for both the RS and DRSN cases that the ability of the neural network to generalize without overfitting is tied to ‘stiff’ and ‘sloppy’ directions in the parameter space of the mechanistic model. These results serve as proof-of-concept for radiobiologically interpretable deep learning of NTCP, while simultaneously yielding insight into how such techniques can robustly generalize beyond the training set despite uncertainty in individual parameters.

Cluster model incorporating heterogeneous dose distribution of partial parotid irradiation for radiotherapy induced xerostomia prediction with machine learning methods

PURPOSE

A cluster model incorporating heterogeneous dose distribution within the parotid gland was developed and validated retrospectively for radiotherapy (RT) induced xerostomia prediction with machine learning (ML) techniques.

METHODS

Sixty clusters were obtained at 1 Gy step size with threshold doses ranging from 1 to 60 Gy, for each of the enrolled 155 patients with HNC from three institutions. Feature clusters were selected with the neighborhood component analysis (NCA) and subsequently fed into four supervised ML models for xerostomia prediction comparison: support vector machines (SVM), k-nearest neighbor (kNN), naïve Bayes (NB), and random forest (RF). The predictive performance of each model was evaluated using cross validation resampling with the area-under-the-curves (AUC) of the receiver-operating-characteristic (ROC). The xerostomia predicting capacity using testing data was assessed with accuracy, sensitivity, and specificity for these models and three cluster connectivity choices. Mean dose based logistic regression served as the benchmark for evaluation.

RESULTS

Feature clusters identified by NCA fell in three threshold dose ranges: 5-15Gy, 25-35Gy, and 45-50Gy. Mean dose predictive power was 15% lower than that of the cluster model using the logistic regression classifier. Model validation demonstrated that kNN model outperformed slightly other three models but no substantial difference was observed. Applying the fine-tuned models to testing data yielded that the mean accuracy from SVM, kNN and NB models were between 0.68 and 0.7 while that of RF was ∼0.6. SVM model yielded the best sensitivity (0.76) and kNN model delivered consistent sensitivity and specificity. This is consistent with cross validation. Clusters calculated with three connectivity choices exhibited minimally different predictions.

CONCLUSION

Compared to mean dose, the proposed cluster model has shown its improvement as the xerostomia predictor. When combining with ML techniques, it could provide a clinically useful tool for xerostomia prediction and facilitate decision making during radiotherapy planning for patients with HNC.

Intelligent systems for precision dental diagnosis and treatment planning – A review

Machines have changed the course of mankind. Simple machines were the basis of human civilization. Today with humongous technological development, machines are intelligent enough to carry out very complex nerve-racking tasks. The ability of a machine to learn from algorithms changed eventually into, the machine learning by itself, which constitutes artificial intelligence. Literature has plausible evidence for the use of intelligent systems in medical field. Artificial intelligence has been used in the multiple denominations of dentistry. These machines are used in the precision diagnosis, interpretation of medical images, accumulation of data, classification and compilation of records, determination of treatment and construction of a personalized treatment plan. Artificial intelligence can help in timely diagnosis of complex dental diseases which would ultimately aid in rapid commencement of treatment. Research helps us understand the effectiveness and challenges in the use of this technology. The apt use of intelligent systems could transform the entire medical system for the better.

Deep Learning for Radiotherapy Outcome Prediction Using Dose Data - A Review

Artificial intelligence, and in particular deep learning using convolutional neural networks, has been used extensively for image classification and segmentation, including on medical images for diagnosis and prognosis prediction. Use in radiotherapy prognostic modelling is still limited, however, especially as applied to toxicity and tumour response prediction from radiation dose distributions. We review and summarise studies that applied deep learning to radiotherapy dose data, in particular studies that utilised full three-dimensional dose distributions. Ten papers have reported on deep learning models for outcome prediction utilising spatial dose information, whereas four studies used reduced dimensionality (dose volume histogram) information for prediction. Many of these studies suffer from the same issues that plagued early normal tissue complication probability modelling, including small, single-institutional patient cohorts, lack of external validation, poor data and model reporting, use of late toxicity data without taking time-to-event into account, and nearly exclusive focus on clinician-reported complications. They demonstrate, however, how radiation dose, imaging and clinical data may be technically integrated in convolutional neural networks-based models; and some studies explore how deep learning may help better understand spatial variation in radiosensitivity. In general, there are a number of issues specific to the intersection of radiotherapy outcome modelling and deep learning, for example translation of model developments into treatment plan optimisation, which will require further combined effort from the radiation oncology and artificial intelligence communities.

Locoregional Recurrence Prediction Using a Deep Neural Network of Radiological and Radiotherapy Images

Radiation therapy (RT) is an important and potentially curative modality for head and neck squamous cell carcinoma (HNSCC). Locoregional recurrence (LR) of HNSCC after RT is ranging from 15% to 50% depending on the primary site and stage. In addition, the 5-year survival rate of patients with LR is low. To classify high-risk patients who might develop LR, a deep learning model for predicting LR needs to be established. In this work, 157 patients with HNSCC who underwent RT were analyzed. Based on the National Cancer Institute’s multi-institutional TCIA data set containing FDG-PET/CT/dose, a 3D deep learning model was proposed to predict LR without time-consuming segmentation or feature extraction. Our model achieved an averaged area under the curve (AUC) of 0.856. Adding clinical factors into the model improved the AUC to an average of 0.892 with the highest AUC of up to 0.974. The 3D deep learning model could perform individualized risk quantification of LR in patients with HNSCC without time-consuming tumor segmentation.

Application of machine learning in oral and maxillofacial surgery

Oral and maxillofacial anatomy is extremely complex, and medical imaging is critical in the diagnosis and treatment of soft and bone tissue lesions. Hence, there exists accumulating imaging data without being properly utilized over the last decades. As a result, problems are emerging regarding how to integrate and interpret a large amount of medical data and alleviate clinicians’ workload. Recently, artificial intelligence has been developing rapidly to analyze complex medical data, and machine learning is one of the specific methods of achieving this goal, which is based on a set of algorithms and previous results. Machine learning has been considered useful in assisting early diagnosis, treatment planning, and prognostic estimation through extracting key features and building mathematical models by computers. Over the past decade, machine learning techniques have been applied to the field of oral and maxillofacial surgery and increasingly achieved expert-level performance. Thus, we hold a positive attitude towards developing machine learning for reducing the number of medical errors, improving the quality of patient care, and optimizing clinical decision-making in oral and maxillofacial surgery. In this review, we explore the clinical application of machine learning in maxillofacial cysts and tumors, maxillofacial defect reconstruction, orthognathic surgery, and dental implant and discuss its current problems and solutions.

Improving robustness of a deep learning-based lung-nodule classification model of CT images with respect to image noise

Objective. Robustness is an important aspect to consider, when developing methods for medical image analysis. This study investigated robustness properties of deep neural networks (DNNs) for a lung nodule classification problem based on CT images and proposed a solution to improve robustness.Approach. We firstly constructed a class of four DNNs with different widths, each predicting an output label (benign or malignant) for an input CT image cube containing a lung nodule. These networks were trained to achieve Area Under the Curve of 0.891-0.914 on a testing dataset. We then added to the input CT image cubes noise signals generated randomly using a realistic CT image noise model based on a noise power spectrum at 100 mAs, and monitored the DNNs output change. We definedSAR₅(%) to quantify the robustness of the trained DNN model, indicating that for 5% of CT image cubes, the noise can change the prediction results with a chance of at leastSAR₅(%). To understand robustness, we viewed the information processing pipeline by the DNN as a two-step process, with the first step using all but the last layers to extract representations of the input CT image cubes in a latent space, and the second step employing the last fully-connected layer as a linear classifier to determine the position of the sample representations relative to a decision plane. To improve robustness, we proposed to retrain the last layer of the DNN with a Supporting Vector Machine (SVM) hinge loss function to enforce the desired position of the decision plane.Main results.SAR₅ranged in 47.0%-62.0% in different DNNs. The unrobustness behavior may be ascribed to the unfavorable placement of the decision plane in the latent representation space, which made some samples be perturbed to across the decision plane and hence susceptible to noise. The DNN-SVM model improved robustness over the DNN model and reducedSAR₅by 8.8%-21.0%.Significance. This study provided insights about the potential reason for the unrobustness behavior of DNNs and the proposed DNN-SVM model improved model robustness.

Deep Learning for Clinical Image Analyses in Oral Squamous Cell Carcinoma: A Review

Importance

Oral squamous cell carcinoma (SCC) is a lethal malignant neoplasm with a high rate of tumor metastasis and recurrence. Accurate diagnosis, prognosis prediction, and metastasis detection can improve patient outcomes. Deep learning for clinical image analysis can be used for diagnosis and prognosis in cancers, including oral SCC; its use in these areas can improve patient care and outcome.

Observations

This review is a summary of the use of deep learning models for diagnosis, prognosis, and metastasis detection for oral SCC by analyzing information from pathological and radiographic images. Specifically, deep learning has been used to classify different cell types, to differentiate cancer cells from nonmalignant cells, and to identify oral SCC from other cancer types. It can also be used to predict survival, to differentiate between tumor grades, and to detect lymph node metastasis. In general, the performance of these deep learning models has an accuracy ranging from 77.89% to 97.51% and 76% to 94.2% with the use of pathological and radiographic images, respectively. The review also discusses the importance of using good-quality clinical images in sufficient quantity on model performance.

Conclusions and Relevance

Applying pathological and radiographic images in deep learning models for diagnosis and prognosis of oral SCC has been explored, and most studies report results showing good classification accuracy. The successful use of deep learning in these areas has a high clinical translatability in the improvement of patient care.

Dosiomics-based prediction of radiation-induced hypothyroidism in nasopharyngeal carcinoma patients

PURPOSE

To predict the incidence of radiation-induced hypothyroidism (RHT) in nasopharyngeal carcinoma (NPC) patients, dosiomics features based prediction models were established.

MATERIALS AND METHODS

A total of 145 NPC patients treated with radiotherapy from January 2012 to January 2015 were included. Dosiomics features of the dose distribution within thyroid gland were extracted. The minimal-redundancy-maximal-relevance (mRMR) criterion was used to rank the extracted features and selected the most relevant features. Machine learning (ML) algorithms including logistic regression (LR), support vector machine (SVM), random forest (RF), and k-nearest neighbor (KNN) were utilized to establish prediction models, respectively. Nested sampling and hyper-tuning methods were adopted to train and validate the prediction models. The dosiomics-based (DO) prediction models were evaluated through comparing with the dose-volume factor-based (DV) models in terms of the area under the receiver operating characteristic (ROC) curve (AUC). The demographics factors (age and gender) were included in both DO model and DV model.

RESULTS

Age, V45 and 37 dosiomics features exhibited significant correlations with RHT in univariate analysis. For prediction performance, DO prediction models exhibited better results with the best AUC value 0.7 while DV prediction models 0.61. In DO prediction models, the AUC values displayed a trend from ascending to descending with the increasing of selected features. The highest AUC value was achieved when the number of selected features was 3. In DV prediction model, similar trend was not observed.

CONCLUSION

This study established a prediction model based on the dosiomics features with better performance than conventional dose-volume factors, leading to early predict the possible RHT among NPC patients who had received radiotherapy and take precaution measures for NPC patients.

The Application and Development of Deep Learning in Radiotherapy: A Systematic Review

With the massive use of computers, the growth and explosion of data has greatly promoted the development of artificial intelligence (AI). The rise of deep learning (DL) algorithms, such as convolutional neural networks (CNN), has provided radiation oncologists with many promising tools that can simplify the complex radiotherapy process in the clinical work of radiation oncology, improve the accuracy and objectivity of diagnosis, and reduce the workload, thus enabling clinicians to spend more time on advanced decision-making tasks. As the development of DL gets closer to clinical practice, radiation oncologists will need to be more familiar with its principles to properly evaluate and use this powerful tool. In this paper, we explain the development and basic concepts of AI and discuss its application in radiation oncology based on different task categories of DL algorithms. This work clarifies the possibility of further development of DL in radiation oncology.

Spatial descriptions of radiotherapy dose: normal tissue complication models and statistical associations

For decades, dose-volume information for segmented anatomy has provided the essential data for correlating radiotherapy dosimetry with treatment-induced complications. Dose-volume information has formed the basis for modelling those associations via normal tissue complication probability (NTCP) models and for driving treatment planning. Limitations to this approach have been identified. Many studies have emerged demonstrating that the incorporation of information describing the spatial nature of the dose distribution, and potentially its correlation with anatomy, can provide more robust associations with toxicity and seed more general NTCP models. Such approaches are culminating in the application of computationally intensive processes such as machine learning and the application of neural networks. The opportunities these approaches have for individualising treatment, predicting toxicity and expanding the solution space for radiation therapy are substantial and have clearly widespread and disruptive potential. Impediments to reaching that potential include issues associated with data collection, model generalisation and validation. This review examines the role of spatial models of complication and summarises relevant published studies. Sources of data for these studies, appropriate statistical methodology frameworks for processing spatial dose information and extracting relevant features are described. Spatial complication modelling is consolidated as a pathway to guiding future developments towards effective, complication-free radiotherapy treatment.

Artificial Intelligence and Radiomics in Head and Neck Cancer Care: Opportunities, Mechanics, and Challenges

The advent of large-scale high-performance computing has allowed the development of machine-learning techniques in oncologic applications. Among these, there has been substantial growth in radiomics (machine-learning texture analysis of images) and artificial intelligence (which uses deep-learning techniques for "learning algorithms"); however, clinical implementation has yet to be realized at scale. To improve implementation, opportunities, mechanics, and challenges, models of imaging-enabled artificial intelligence approaches need to be understood by clinicians who make the treatment decisions. This article aims to convey the basic conceptual premises of radiomics and artificial intelligence using head and neck cancer as a use case. This educational overview focuses on approaches for head and neck oncology imaging, detailing current research efforts and challenges to implementation.

Machine learning in dental, oral and craniofacial imaging: a review of recent progress

Artificial intelligence has been emerging as an increasingly important aspect of our daily lives and is widely applied in medical science. One major application of artificial intelligence in medical science is medical imaging. As a major component of artificial intelligence, many machine learning models are applied in medical diagnosis and treatment with the advancement of technology and medical imaging facilities. The popularity of convolutional neural network in dental, oral and craniofacial imaging is heightening, as it has been continually applied to a broader spectrum of scientific studies. Our manuscript reviews the fundamental principles and rationales behind machine learning, and summarizes its research progress and its recent applications specifically in dental, oral and craniofacial imaging. It also reviews the problems that remain to be resolved and evaluates the prospect of the future development of this field of scientific study.

Comparison of machine learning methods for prediction of osteoradionecrosis incidence in patients with head and neck cancer

Objectives:

Mandible osteoradionecrosis (ORN) is one of the most severe toxicities in patients with head and neck cancer (HNC) undergoing radiotherapy (RT). The existing literature focuses on the correlation of mandible ORN and clinical and dosimetric factors. This study proposes the use of machine learning (ML) methods as prediction models for mandible ORN incidence.

Methods:

A total of 96 patients (ORN incidence ratio of 1:1) treated between 2011 and 2015 were selected from the local HNC toxicity database. Demographic, clinical and dosimetric data (based on the mandible dose–volume histogram) were considered as model variables. Prediction accuracy (measured using a stratified fivefold nested cross-validation), sensitivity, specificity, precision and negative predictive value were used to evaluate the prediction performance of a multivariate logistic regression (LR) model, a support vector machine (SVM) model, a random forest (RF) model, an adaptive boosting (AdaBoost) model and an artificial neural network (ANN) model. The different models were compared based on their prediction accuracy and using the McNemar’s hypothesis test.

Results:

The ANN model (77% accuracy), closely followed by the SVM (76%), AdaBoost (75%) and LR (75%) models, showed the highest overall prediction accuracy. The RF model (71%) showed the lowest prediction accuracy. However, based on the McNemar’s test applied to all model pair combinations, no statistically significant difference between the models was found.

Conclusion:

Based on our results, we encourage the use of ML-based prediction models for ORN incidence as has already been done for other HNC toxicity end points.

Advances in knowledge:

This research opens a new path towards personalised RT for HNC using ML to predict mandible ORN incidence.

Artificial intelligence in radiotherapy: a technological review

Radiation therapy (RT) is widely used to treat cancer. Technological advances in RT have occurred in the past 30 years. These advances, such as three-dimensional image guidance, intensity modulation, and robotics, created challenges and opportunities for the next breakthrough, in which artificial intelligence (AI) will possibly play important roles. AI will replace certain repetitive and labor-intensive tasks and improve the accuracy and consistency of others, particularly those with increased complexity because of technological advances. The improvement in efficiency and consistency is important to manage the increasing cancer patient burden to the society. Furthermore, AI may provide new functionalities that facilitate satisfactory RT. The functionalities include superior images for real-time intervention and adaptive and personalized RT. AI may effectively synthesize and analyze big data for such purposes. This review describes the RT workflow and identifies areas, including imaging, treatment planning, quality assurance, and outcome prediction, that benefit from AI. This review primarily focuses on deep-learning techniques, although conventional machine-learning techniques are also mentioned.

Detecting spatial susceptibility to cardiac toxicity of radiation therapy for lung cancer

Radiation therapy (RT) is a commonly used approach for treating lung cancer. Because the lungs are close to the heart, radiation dose may inevitably spill to the heart, causing heart damage and diminishing treatment outcomes. There is an urgent need to better understand how treatment outcomes are affected by radiation dose spilled to the heart in order to optimize RT planning. However, despite the fact that dose distribution on the heart is 3-D, most existing research collapses the 3-D dose map into a 1-D histogram to be linked with outcomes. This ignores the spatial information. We propose a novel method that automatically searches for subregions of the heart that are susceptible to radiation toxicity, called Toxicity-Susceptible Subregions (TSSs), based on the 3-D dose distribution. We apply the proposed method to a real-world dataset and find TSSs that harbor the sinoatrial node of the electronic conduction system of the heart. Damage of the sinoatrial node by radiation toxicity disrupts the crucial function of the heart, leading to shortening of the overall survival. Our finding suggests that protective strategies may be developed to spare the TSSs, and thus helping RT planning achieve optimal results in treating lung cancer patients.

Introduction to machine and deep learning for medical physicists

Recent years have witnessed tremendous growth in the application of machine learning (ML) and deep learning (DL) techniques in medical physics. Embracing the current big data era, medical physicists equipped with these state-of-the-art tools should be able to solve pressing problems in modern radiation oncology. Here, a review of the basic aspects involved in ML/DL model building, including data processing, model training, and validation for medical physics applications is presented and discussed. Machine learning can be categorized based on the underlying task into supervised learning, unsupervised learning, or reinforcement learning; each of these categories has its own input/output dataset characteristics and aims to solve different classes of problems in medical physics ranging from automation of processes to predictive analytics. It is recognized that data size requirements may vary depending on the specific medical physics application and the nature of the algorithms applied. Data processing, which is a crucial step for model stability and precision, should be performed before training the model. Deep learning as a subset of ML is able to learn multilevel representations from raw input data, eliminating the necessity for hand crafted features in classical ML. It can be thought of as an extension of the classical linear models but with multilayer (deep) structures and nonlinear activation functions. The logic of going "deeper" is related to learning complex data structures and its realization has been aided by recent advancements in parallel computing architectures and the development of more robust optimization methods for efficient training of these algorithms. Model validation is an essential part of ML/DL model building. Without it, the model being developed cannot be easily trusted to generalize to unseen data. Whenever applying ML/DL, one should keep in mind, according to Amara's law, that humans may tend to overestimate the ability of a technology in the short term and underestimate its capability in the long term. To establish ML/DL role into standard clinical workflow, models considering balance between accuracy and interpretability should be developed. Machine learning/DL algorithms have potential in numerous radiation oncology applications, including automatizing mundane procedures, improving efficiency and safety of auto-contouring, treatment planning, quality assurance, motion management, and outcome predictions. Medical physicists have been at the frontiers of technology translation into medicine and they ought to be prepared to embrace the inevitable role of ML/DL in the practice of radiation oncology and lead its clinical implementation.

Artificial Intelligence in radiotherapy: state of the art and future directions

Recent advances in computing capability allowed the development of sophisticated predictive models to assess complex relationships within observational data, described as Artificial Intelligence. Medicine is one of the several fields of application and Radiation oncology could benefit from these approaches, particularly in patients' medical records, imaging, baseline pathology, planning or instrumental data. Artificial Intelligence systems could simplify many steps of the complex workflow of radiotherapy such as segmentation, planning or delivery. However, Artificial Intelligence could be considered as a "black box" in which human operator may only understand input and output predictions and its application to the clinical practice remains a challenge. The low transparency of the overall system is questionable from manifold points of view (ethical included). Given the complexity of this issue, we collected the basic definitions to help the clinician to understand current literature, and overviewed experiences regarding implementation of AI within radiotherapy clinical workflow, aiming to describe this field from the clinician perspective.

Automatic kidney segmentation in ultrasound images using subsequent boundary distance regression and pixelwise classification networks

It remains challenging to automatically segment kidneys in clinical ultrasound (US) images due to the kidneys' varied shapes and image intensity distributions, although semi-automatic methods have achieved promising performance. In this study, we propose subsequent boundary distance regression and pixel classification networks to segment the kidneys automatically. Particularly, we first use deep neural networks pre-trained for classification of natural images to extract high-level image features from US images. These features are used as input to learn kidney boundary distance maps using a boundary distance regression network and the predicted boundary distance maps are classified as kidney pixels or non-kidney pixels using a pixelwise classification network in an end-to-end learning fashion. We also adopted a data-augmentation method based on kidney shape registration to generate enriched training data from a small number of US images with manually segmented kidney labels. Experimental results have demonstrated that our method could automatically segment the kidney with promising performance, significantly better than deep learning-based pixel classification networks.

Robust Collaborative Clustering of Subjects and Radiomic Features for Cancer Prognosis

Feature dimensionality reduction plays an important role in radiomic studies with a large number of features. However, conventional radiomic approaches may suffer from noise, and feature dimensionality reduction techniques are not equipped to utilize latent supervision information of patient data under study, such as differences in patients, to learn discriminative low dimensional representations. To achieve robustness to noise and feature dimensionality reduction with improved discriminative power, we develop a robust collaborative clustering method to simultaneously cluster patients and radiomic features into distinct groups respectively under adaptive sparse regularization. Our method is built upon matrix tri-factorization enhanced by adaptive sparsity regularization for simultaneous feature dimensionality reduction and denoising. Particularly, latent grouping information of patients with distinct radiomic features is learned and utilized as supervision information to guide the feature dimensionality reduction, and noise in radiomic features is adaptively isolated in a Bayesian framework under a general assumption of Laplacian distributions of transform-domain coefficients. Experiments on synthetic data have demonstrated the effectiveness of the proposed approach in data clustering, and evaluation results on an FDG-PET/CT dataset of rectal cancer patients have demonstrated that the proposed method outperforms alternative methods in terms of both patient stratification and prediction of patient clinical outcomes.

Radiomic features from MRI distinguish myxomas from myxofibrosarcomas

BACKGROUND

Myxoid tumors pose diagnostic challenges for radiologists and pathologists. All myxoid tumors can be differentiated from each other using fluorescent in-situ hybridization (FISH) or immunohistochemical markers, except for myxomas and myxofibrosarcomas. Myxomas and myxofibrosarcomas are rare tumors. Myxomas are benign and histologically bland, whereas myxofibrosarcomas are malignant and histologically heterogenous. Because of the histological heterogeneity, low grade myxofibrosarcomas may be mistaken for myxomas on core needle biopsies. We evaluated the performance of T1-weighted signal intensity (T1SI), tumor volume, and radiomic features extracted from magnetic resonance imaging (MRI) to differentiate myxomas from myxofibrosarcomas.

METHODS

The MRIs of 56 patients (29 with myxomas, 27 with myxofibrosarcomas) were analyzed. We extracted 89 radiomic features. Random forests based classifiers using the T1SI, volume features, and radiomic features were used to differentiate myxomas from myxofibrosarcomas. The classifiers were validated using a leave-one-out cross-validation. The performances of the classifiers were then compared.

RESULTS

Myxomas had lower normalized T1SI than myxofibrosaromas (p = 0.006) and the AUC using the T1SI was 0.713. However, the classification model using radiomic features had an AUC of 0.885 (accuracy = 0.839, sensitivity = 0.852, specificity = 0.828), and outperformed the classification models using T1SI (AUC = 0.713) and tumor volume (AUC = 0.838). The classification model using radiomic features was significantly better than the classifier using T1SI values (p = 0.039).

CONCLUSIONS

Myxofibrosarcomas are on average higher in T1-weighted signal intensity than myxomas. Myxofibrosarcomas are larger and have shape differences compared to myxomas. Radiomic features performed best for differentiating myxomas from myxofibrosarcomas compared to T1-weighted signal intensity and tumor volume features.