D133. Development of a Mobile Application for Patient Screening in Headache Surgery

Abstract

PURPOSE: Patient selection is crucial for satisfactory outcomes in headache surgery. Pain drawings are an important tool for presurgical patient screening, but their interpretation requires extensive training. We aimed to A.) apply machine learning (ML) algorithms to pain drawing analysis and B.) integrate this tool into a mobile application to facilitate screening for providers. METHODS: 115 pain drawings created by patients were analyzed and categorized for presence of nerve pain, anatomic distribution of pain, as well as candidacy for surgery by two headache surgery experts (WGA and LG). A random forest ML algorithm was developed based on this analysis and trained using 3208 copies of the initial patient drawings. The algorithm was asked to evaluate 200 randomly selected drawings. Android Studio 2.0 (Google, California, USA) was used to create a mobile application. RESULTS: The algorithm correctly determined nerve pain, the anatomic pain distribution of pain, and candidacy for surgery in 97.5% of test drawings (195/200). Across the different anatomic pain distributions, the mean algorithm accuracy was 98.4± 1.3%. The mobile application surface enabled non-surgical providers to successfully use the algorithm. CONCLUSION: ML can be used in the interpretation of patient pain drawings. The integration of this screening tool into a mobile application will enable non-surgical providers such as neurologists and primary care physicians, as well patients to screen for nerve related headache increasing the likelihood of timely referral to a qualified headache surgeon.