Evaluation of Neural Machine translation for conversion of International Classification of disease codes to the Abbreviated injury Scale

The Abbreviated Injury Scale (AIS) is an essential tool for injury research since it allows for comparisons of injury severity among patients, however, the International Classification of Diseases (ICD) is more widely used to capture medical information. The problem of conversion between these two medical coding systems has similarities to the challenges encountered in language translation. We therefore hypothesize that neural machine translation (NMT), a deep learning technique which is commonly used for human language translation, could be used to convert ICD codes to AIS. The objective of this study was to compare the accuracy of a NMT model for determining injury severity compared to two established methods of conversion. The injury severity classifications used for this study were Injury Severity Score (ISS) ≥ 16, Maximum AIS severity (MAIS) ≥ 3, and MAIS ≥ 2. Data from a US national trauma registry, which has patient injuries coded in both AIS and ICD, was used to train a NMT model. Testing data from a separate year was used to determine the accuracy of the NMT model predictions against the actual ISS recorded in the registry. The prediction accuracy of the NMT model was compared to that of the official Association for the Advancement of Automotive Medicine (AAAM) ICD-AIS map and the R package 'ICD Program for Injury Categorization in R' (ICDPIC-R). The results show that the NMT model was the most accurate across all injury severity classifications, followed by the ICD-AIS map and then ICDPIC-R package. The NMT model also showed the highest correlation between the predicted and observe ISS scores. Overall, NMT appears to be a promising method for predicting injury severity from ICD codes, however, validation in external databases is needed.

Peer-Developed Modules on Basic Biostatistics and Evidence-Based Medicine Principles for Undergraduate Medical Education

Introduction

Evidence-based medicine (EBM) is pivotal in shaping patient care, yet it is challenging to incorporate into undergraduate medical education (UME) due to a lack of dedicated resources within the preclinical curriculum. To address this challenge, we used a peer-led approach to explain difficult concepts through language that students can understand at their shared level of understanding.

Methods

Four second-year medical students trained in EBM over 18 months by facilitating monthly journal clubs, ultimately leading to their involvement as peer-instructors. With input from a faculty expert, peer-instructors designed integrative PowerPoint modules and interactive problem sets on basic biostatistics and EBM principles. Assessment included formative quizzes with multiple attempts to achieve at least 80% to demonstrate mastery of core learning objectives. Afterwards, students were invited to provide feedback using a 5-point Likert scale survey.

Results

Of second-year students who participated, all 151 demonstrated 80% competency on each quiz. Eighty-seven (58%) students completed the survey on which, 77% agreed/strongly agreed that their level of understanding of EBM improved after the peer-led sessions, 76% agreed/strongly agreed that the sessions were more conducive to learning compared to traditional lectures, and 94% agreed/strongly agreed that the material covered was relevant to the USMLE Step 1.

Discussion

This peer-led approach has been rated as effective by learners, improving their ability to critically appraise and apply clinical evidence. To promote integration of EBM into UME, we have prepared modules, problem sets, quizzes, and an outline of the problem-solving sessions for universal adoption.