Doctor XAvIer: Explainable Diagnosis on Physician-Patient Dialogues and XAI Evaluation
It addresses the need for explainable AI in medical diagnostics, though it is incremental with a focus on evaluation metrics.
The paper tackles the problem of extracting clinical data from patient-doctor dialogues and explaining diagnostic predictions, achieving an F1-score of 0.97 in named entity recognition and symptom classification and 0.91 in diagnosis classification.
We introduce Doctor XAvIer, a BERT-based diagnostic system that extracts relevant clinical data from transcribed patient-doctor dialogues and explains predictions using feature attribution methods. We present a novel performance plot and evaluation metric for feature attribution methods: Feature Attribution Dropping (FAD) curve and its Normalized Area Under the Curve (N-AUC). FAD curve analysis shows that integrated gradients outperforms Shapley values in explaining diagnosis classification. Doctor XAvIer outperforms the baseline with 0.97 F1-score in named entity recognition and symptom pertinence classification and 0.91 F1-score in diagnosis classification.