Citations

Abstract Objective The goal is to provide an overview of artificial intelligence (AI) and machine learning (ML) methodology and appraisal tailored to clinicians and researchers in the headache field to facilitate interdisciplinary communications and research. Background The application of AI to the study of headache and other healthcare challenges is growing rapidly. It is critical that these findings be accurately interpreted by headache specialists, but this can be difficult for non‐AI specialists. Methods This paper is a narrative review of the fundamentals required to understand ML/AI headache research. Using guidance from key leaders in the field of headache medicine and AI, important references were reviewed and cited to provide a comprehensive overview of the terminology, methodology, applications, pitfalls, and bias of AI. Results We review how AI models are created, common model types, methods for evaluation, and examples of their application to headache medicine. We also highlight potential pitfalls relevant when consuming AI research, and discuss ethical issues of bias, privacy and abuse generated by AI. Additionally, we highlight recent related research from across headache‐related applications. Conclusion Many promising current and future applications of ML and AI exist in the field of headache medicine. Understanding the fundamentals of AI will allow readers to understand and critically appraise AI‐related research findings in their proper context. This paper will increase the reader's comfort in consuming AI/ML‐based research and will prepare them to think critically about related research developments.

AbstractBackgroundIn treatment planning for intensity‐modulated proton therapy (IMPT), we aim to deliver the prescribed dose to the target yet minimize the dose to adjacent healthy tissue. Mixed‐integer programming (MIP) has been applied in radiation therapy to generate treatment plans. However, MIP has not been used effectively for IMPT treatment planning with dose‐volume constraints. In this study, we incorporated dose‐volume constraints in an MIP model to generate treatment plans for IMPT.MethodsWe created a new MIP model for IMPT with dose volume constraints. Two groups of IMPT treatment plans were generated for each of three patients by using MIP models for a total of six plans: one plan was derived with the Limited‐memory Broyden–Fletcher–Goldfarb–Shanno (L‐BFGS) method while the other plan was derived with our MIP model with dose‐volume constraints. We then compared these two plans by dose‐volume histogram (DVH) indices to evaluate the performance of the new MIP model with dose‐volume constraints. In addition, we developed a model to more efficiently find the best balance between tumor coverage and normal tissue protection.ResultsThe MIP model with dose‐volume constraints generates IMPT treatment plans with comparable target dose coverage, target dose homogeneity, and the maximum dose to organs at risk (OARs) compared to treatment plans from the conventional quadratic programming method without any tedious trial‐and‐error process. Some notable reduction in the mean doses of OARs is observed.ConclusionsThe treatment plans from our MIP model with dose‐volume constraints can meet all dose‐volume constraints for OARs and targets without any tedious trial‐and‐error process. This model has the potential to automatically generate IMPT plans with consistent plan quality among different treatment planners and across institutions and better protection for important parallel OARs in an effective way.