Artificial Intelligence for Radiation Oncology Applications Using Public Datasets

The Evolving Role of Artificial Intelligence in Radiotherapy Treatment Planning-A Literature Review

This paper examines the integration of artificial intelligence (AI) in radiotherapy for cancer treatment. The importance of radiotherapy in cancer management and its time-intensive planning process make AI adoption appealing especially with the escalating demand for radiotherapy. This review highlights the efficacy of AI across medical domains, where it surpasses human capabilities in areas such as cardiology and dermatology. Focusing on radiotherapy, the paper details AI's applications in target segmentation, dose optimization, and outcome prediction. It discusses adaptive radiotherapy's benefits and AI's potential to enhance patient outcomes with much improved treatment accuracy. The paper explores ethical concerns, including data privacy and bias, stressing the need for robust guidelines. Educating healthcare professionals and patients about AI's role is crucial as it acknowledges potential job-role changes and concerns about patients' trust in the use of AI. Overall, the integration of AI in radiotherapy holds transformative potential in streamlining processes, improving outcomes, and reducing costs. AI's potential to reduce healthcare costs underscores its significance with impactful change globally. However, successful implementation hinges on addressing ethical and logistical challenges and fostering collaboration among healthcare professionals and patient population data sets for its optimal utilization. Rigorous education, collaborative efforts, and global data sharing will be the compass guiding its' success in radiotherapy and healthcare.

Application of simultaneous uncertainty quantification and segmentation for oropharyngeal cancer use-case with Bayesian deep learning

BACKGROUND

Radiotherapy is a core treatment modality for oropharyngeal cancer (OPC), where the primary gross tumor volume (GTVp) is manually segmented with high interobserver variability. This calls for reliable and trustworthy automated tools in clinician workflow. Therefore, accurate uncertainty quantification and its downstream utilization is critical.

METHODS

Here we propose uncertainty-aware deep learning for OPC GTVp segmentation, and illustrate the utility of uncertainty in multiple applications. We examine two Bayesian deep learning (BDL) models and eight uncertainty measures, and utilize a large multi-institute dataset of 292 PET/CT scans to systematically analyze our approach.

RESULTS

We show that our uncertainty-based approach accurately predicts the quality of the deep learning segmentation in 86.6% of cases, identifies low performance cases for semi-automated correction, and visualizes regions of the scans where the segmentations likely fail.

CONCLUSIONS

Our BDL-based analysis provides a first-step towards more widespread implementation of uncertainty quantification in OPC GTVp segmentation.

Towards equitable AI in oncology

Artificial intelligence (AI) stands at the threshold of revolutionizing clinical oncology, with considerable potential to improve early cancer detection and risk assessment, and to enable more accurate personalized treatment recommendations. However, a notable imbalance exists in the distribution of the benefits of AI, which disproportionately favour those living in specific geographical locations and in specific populations. In this Perspective, we discuss the need to foster the development of equitable AI tools that are both accurate in and accessible to a diverse range of patient populations, including those in low-income to middle-income countries. We also discuss some of the challenges and potential solutions in attaining equitable AI, including addressing the historically limited representation of diverse populations in existing clinical datasets and the use of inadequate clinical validation methods. Additionally, we focus on extant sources of inequity including the type of model approach (such as deep learning, and feature engineering-based methods), the implications of dataset curation strategies, the need for rigorous validation across a variety of populations and settings, and the risk of introducing contextual bias that comes with developing tools predominantly in high-income countries.

Artificial Intelligence Uncertainty Quantification in Radiotherapy Applications - A Scoping Review

Background/purpose

The use of artificial intelligence (AI) in radiotherapy (RT) is expanding rapidly. However, there exists a notable lack of clinician trust in AI models, underscoring the need for effective uncertainty quantification (UQ) methods. The purpose of this study was to scope existing literature related to UQ in RT, identify areas of improvement, and determine future directions.

Methods

We followed the PRISMA-ScR scoping review reporting guidelines. We utilized the population (human cancer patients), concept (utilization of AI UQ), context (radiotherapy applications) framework to structure our search and screening process. We conducted a systematic search spanning seven databases, supplemented by manual curation, up to January 2024. Our search yielded a total of 8980 articles for initial review. Manuscript screening and data extraction was performed in Covidence. Data extraction categories included general study characteristics, RT characteristics, AI characteristics, and UQ characteristics.

Results

We identified 56 articles published from 2015-2024. 10 domains of RT applications were represented; most studies evaluated auto-contouring (50%), followed by image-synthesis (13%), and multiple applications simultaneously (11%). 12 disease sites were represented, with head and neck cancer being the most common disease site independent of application space (32%). Imaging data was used in 91% of studies, while only 13% incorporated RT dose information. Most studies focused on failure detection as the main application of UQ (60%), with Monte Carlo dropout being the most commonly implemented UQ method (32%) followed by ensembling (16%). 55% of studies did not share code or datasets.

Conclusion

Our review revealed a lack of diversity in UQ for RT applications beyond auto-contouring. Moreover, there was a clear need to study additional UQ methods, such as conformal prediction. Our results may incentivize the development of guidelines for reporting and implementation of UQ in RT.

The Integration of Deep Learning in Radiotherapy: Exploring Challenges, Opportunities, and Future Directions through an Umbrella Review

This study investigates, through a narrative review, the transformative impact of deep learning (DL) in the field of radiotherapy, particularly in light of the accelerated developments prompted by the COVID-19 pandemic. The proposed approach was based on an umbrella review following a standard narrative checklist and a qualification process. The selection process identified 19 systematic review studies. Through an analysis of current research, the study highlights the revolutionary potential of DL algorithms in optimizing treatment planning, image analysis, and patient outcome prediction in radiotherapy. It underscores the necessity of further exploration into specific research areas to unlock the full capabilities of DL technology. Moreover, the study emphasizes the intricate interplay between digital radiology and radiotherapy, revealing how advancements in one field can significantly influence the other. This interdependence is crucial for addressing complex challenges and advancing the integration of cutting-edge technologies into clinical practice. Collaborative efforts among researchers, clinicians, and regulatory bodies are deemed essential to effectively navigate the evolving landscape of DL in radiotherapy. By fostering interdisciplinary collaborations and conducting thorough investigations, stakeholders can fully leverage the transformative power of DL to enhance patient care and refine therapeutic strategies. Ultimately, this promises to usher in a new era of personalized and optimized radiotherapy treatment for improved patient outcomes.

The diagnostic, therapeutic, and ethical impact of artificial intelligence in modern medicine

In the evolution of modern medicine, artificial intelligence (AI) has been proven to provide an integral aspect of revolutionizing clinical diagnosis, drug discovery, and patient care. With the potential to scrutinize colossal amounts of medical data, radiological and histological images, and genomic data in healthcare institutions, AI-powered systems can recognize, determine, and associate patterns and provide impactful insights that would be strenuous and challenging for clinicians to detect during their daily clinical practice. The outcome of AI-mediated search offers more accurate, personalized patient diagnoses, guides in research for new drug therapies, and provides a more effective multidisciplinary treatment plan that can be implemented for patients with chronic diseases. Among the many promising applications of AI in modern medicine, medical imaging stands out distinctly as an area with tremendous potential. AI-powered algorithms can now accurately and sensitively identify cancer cells and other lesions in medical images with greater accuracy and sensitivity. This allows for earlier diagnosis and treatment, which can significantly impact patient outcomes. This review provides a comprehensive insight into diagnostic, therapeutic, and ethical issues with the advent of AI in modern medicine.

Accuracy and usability of artificial intelligence chatbot generated chemotherapy protocols

Background: Medical practitioners are increasingly using artificial intelligence (AI) chatbots for easier and faster access to information. To our knowledge, the accuracy and availability of AI-generated chemotherapy protocols has not yet been studied. Methods: Nine simulated cancer patient cases were designed and AI chatbots, ChatGPT version 3.5 (OpenAI) and Bing (Microsoft), were used to generate chemotherapy protocols for each case. Results: Generated chemotherapy protocols were compared with the original protocols for nine simulated cancer patients. ChatGPT's overall performance was 5 out of 9 on protocol generation, and Bing's was 4 out of 9; this was statistically nonsignificant (p = 1). Conclusion: AI chatbots show both potential and limitations in generating chemotherapy protocols. The overall performance is low, and they should be used carefully in oncological practice.

Artificial intelligence in biology and medicine, and radioprotection research: perspectives from Jerusalem

While AI is widely used in biomedical research and medical practice, its use is constrained to few specific practical areas, e.g., radiomics. Participants of the workshop on "Artificial Intelligence in Biology and Medicine" (Jerusalem, Feb 14-15, 2023), both researchers and practitioners, aimed to build a holistic picture by exploring AI advancements, challenges and perspectives, as well as to suggest new fields for AI applications. Presentations showcased the potential of large language models (LLMs) in generating molecular structures, predicting protein-ligand interactions, and promoting democratization of AI development. Ethical concerns in medical decision making were also addressed. In biological applications, AI integration of multi-omics and clinical data elucidated the health relevant effects of low doses of ionizing radiation. Bayesian latent modeling identified statistical associations between unobserved variables. Medical applications highlighted liquid biopsy methods for non-invasive diagnostics, routine laboratory tests to identify overlooked illnesses, and AI's role in oral and maxillofacial imaging. Explainable AI and diverse image processing tools improved diagnostics, while text classification detected anorexic behavior in blog posts. The workshop fostered knowledge sharing, discussions, and emphasized the need for further AI development in radioprotection research in support of emerging public health issues. The organizers plan to continue the initiative as an annual event, promoting collaboration and addressing issues and perspectives in AI applications with a focus on low-dose radioprotection research. Researchers involved in radioprotection research and experts in relevant public policy domains are invited to explore the utility of AI in low-dose radiation research at the next workshop.

Artificial Intelligence Applications for Biomedical Cancer Research: A Review

Artificial intelligence (AI) has rapidly evolved and demonstrated its potential in transforming biomedical cancer research, offering innovative solutions for cancer diagnosis, treatment, and overall patient care. Over the past two decades, AI has played a pivotal role in revolutionizing various facets of cancer clinical research. In this comprehensive review, we delve into the diverse applications of AI across the cancer care continuum, encompassing radiodiagnosis, radiotherapy, chemotherapy, immunotherapy, targeted therapy, surgery, and nanotechnology. AI has revolutionized cancer diagnosis, enabling early detection and precise characterization through advanced image analysis techniques. In radiodiagnosis, AI-driven algorithms enhance the accuracy of medical imaging, making it an invaluable tool for clinicians in the detection and assessment of cancer. AI has also revolutionized radiotherapy, facilitating precise tumor boundary delineation, optimizing treatment planning, and enabling real-time adjustments to improve therapeutic outcomes while minimizing collateral damage to healthy tissues. In chemotherapy, AI models have emerged as powerful tools for predicting patient responses to different treatment regimens, allowing for more personalized and effective strategies. In immunotherapy, AI analyzes genetic and imaging data to select ideal candidates for treatment and predict responses. Targeted therapy has seen great advancements with AI, aiding in the identification of specific molecular targets for tailored treatments. AI plays a vital role in surgery by offering real-time navigation and support, enhancing surgical precision. Moreover, the synergy between AI and nanotechnology promises the development of personalized nanomedicines, offering more efficient and targeted cancer treatments. While challenges related to data quality, interpretability, and ethical considerations persist, the future of AI in cancer research holds tremendous promise for improving patient outcomes through advanced and individualized care.

Application of simultaneous uncertainty quantification for image segmentation with probabilistic deep learning: Performance benchmarking of oropharyngeal cancer target delineation as a use-case

Background

Oropharyngeal cancer (OPC) is a widespread disease, with radiotherapy being a core treatment modality. Manual segmentation of the primary gross tumor volume (GTVp) is currently employed for OPC radiotherapy planning, but is subject to significant interobserver variability. Deep learning (DL) approaches have shown promise in automating GTVp segmentation, but comparative (auto)confidence metrics of these models predictions has not been well-explored. Quantifying instance-specific DL model uncertainty is crucial to improving clinician trust and facilitating broad clinical implementation. Therefore, in this study, probabilistic DL models for GTVp auto-segmentation were developed using large-scale PET/CT datasets, and various uncertainty auto-estimation methods were systematically investigated and benchmarked.

Methods

We utilized the publicly available 2021 HECKTOR Challenge training dataset with 224 co-registered PET/CT scans of OPC patients with corresponding GTVp segmentations as a development set. A separate set of 67 co-registered PET/CT scans of OPC patients with corresponding GTVp segmentations was used for external validation. Two approximate Bayesian deep learning methods, the MC Dropout Ensemble and Deep Ensemble, both with five submodels, were evaluated for GTVp segmentation and uncertainty performance. The segmentation performance was evaluated using the volumetric Dice similarity coefficient (DSC), mean surface distance (MSD), and Hausdorff distance at 95% (95HD). The uncertainty was evaluated using four measures from literature: coefficient of variation (CV), structure expected entropy, structure predictive entropy, and structure mutual information, and additionally with our novel Dice-risk measure. The utility of uncertainty information was evaluated with the accuracy of uncertainty-based segmentation performance prediction using the Accuracy vs Uncertainty (AvU) metric, and by examining the linear correlation between uncertainty estimates and DSC. In addition, batch-based and instance-based referral processes were examined, where the patients with high uncertainty were rejected from the set. In the batch referral process, the area under the referral curve with DSC (R-DSC AUC) was used for evaluation, whereas in the instance referral process, the DSC at various uncertainty thresholds were examined.

Results

Both models behaved similarly in terms of the segmentation performance and uncertainty estimation. Specifically, the MC Dropout Ensemble had 0.776 DSC, 1.703 mm MSD, and 5.385 mm 95HD. The Deep Ensemble had 0.767 DSC, 1.717 mm MSD, and 5.477 mm 95HD. The uncertainty measure with the highest DSC correlation was structure predictive entropy with correlation coefficients of 0.699 and 0.692 for the MC Dropout Ensemble and the Deep Ensemble, respectively. The highest AvU value was 0.866 for both models. The best performing uncertainty measure for both models was the CV which had R-DSC AUC of 0.783 and 0.782 for the MC Dropout Ensemble and Deep Ensemble, respectively. With referring patients based on uncertainty thresholds from 0.85 validation DSC for all uncertainty measures, on average the DSC improved from the full dataset by 4.7% and 5.0% while referring 21.8% and 22% patients for MC Dropout Ensemble and Deep Ensemble, respectively.

Conclusion

We found that many of the investigated methods provide overall similar but distinct utility in terms of predicting segmentation quality and referral performance. These findings are a critical first-step towards more widespread implementation of uncertainty quantification in OPC GTVp segmentation.

Segmentation stability of human head and neck cancer medical images for radiotherapy applications under de-identification conditions: Benchmarking data sharing and artificial intelligence use-cases

Background

Demand for head and neck cancer (HNC) radiotherapy data in algorithmic development has prompted increased image dataset sharing. Medical images must comply with data protection requirements so that re-use is enabled without disclosing patient identifiers. Defacing, i.e., the removal of facial features from images, is often considered a reasonable compromise between data protection and re-usability for neuroimaging data. While defacing tools have been developed by the neuroimaging community, their acceptability for radiotherapy applications have not been explored. Therefore, this study systematically investigated the impact of available defacing algorithms on HNC organs at risk (OARs).

Methods

A publicly available dataset of magnetic resonance imaging scans for 55 HNC patients with eight segmented OARs (bilateral submandibular glands, parotid glands, level II neck lymph nodes, level III neck lymph nodes) was utilized. Eight publicly available defacing algorithms were investigated: afni_refacer, DeepDefacer, defacer, fsl_deface, mask_face, mri_deface, pydeface, and quickshear. Using a subset of scans where defacing succeeded (N=29), a 5-fold cross-validation 3D U-net based OAR auto-segmentation model was utilized to perform two main experiments: 1.) comparing original and defaced data for training when evaluated on original data; 2.) using original data for training and comparing the model evaluation on original and defaced data. Models were primarily assessed using the Dice similarity coefficient (DSC).

Results

Most defacing methods were unable to produce any usable images for evaluation, while mask_face, fsl_deface, and pydeface were unable to remove the face for 29%, 18%, and 24% of subjects, respectively. When using the original data for evaluation, the composite OAR DSC was statistically higher (p ≤ 0.05) for the model trained with the original data with a DSC of 0.760 compared to the mask_face, fsl_deface, and pydeface models with DSCs of 0.742, 0.736, and 0.449, respectively. Moreover, the model trained with original data had decreased performance (p ≤ 0.05) when evaluated on the defaced data with DSCs of 0.673, 0.693, and 0.406 for mask_face, fsl_deface, and pydeface, respectively.

Conclusion

Defacing algorithms may have a significant impact on HNC OAR auto-segmentation model training and testing. This work highlights the need for further development of HNC-specific image anonymization methods.