This Patient Looks Like That Patient: Prototypical Networks for Interpretable Diagnosis Prediction from Clinical Text
This addresses the need for interpretable AI in clinical practice to support doctors' decision-making, though it is an incremental improvement by combining existing techniques like prototypical networks and attention.
The authors tackled the problem of making deep neural models for diagnosis prediction from clinical text both accurate and interpretable for doctors, resulting in a model that outperforms existing baselines on two clinical datasets and provides valuable explanations as confirmed by medical evaluations.
The use of deep neural models for diagnosis prediction from clinical text has shown promising results. However, in clinical practice such models must not only be accurate, but provide doctors with interpretable and helpful results. We introduce ProtoPatient, a novel method based on prototypical networks and label-wise attention with both of these abilities. ProtoPatient makes predictions based on parts of the text that are similar to prototypical patients - providing justifications that doctors understand. We evaluate the model on two publicly available clinical datasets and show that it outperforms existing baselines. Quantitative and qualitative evaluations with medical doctors further demonstrate that the model provides valuable explanations for clinical decision support.