Overcoming Uncertain Incompleteness for Robust Multimodal Sequential Diagnosis Prediction via Knowledge Distillation and Random Data Erasing

In this paper, we present NECHO v2, a novel framework designed to enhance the predictive accuracy of multimodal sequential patient diagnoses under uncertain missing visit sequences, a common challenge in clinical settings. Firstly, we modify NECHO to handle uncertain modality representation dominance under the imperfect data. Next, we develop a systematic knowledge distillation by employing the modified NECHO as both teacher and student. It encompasses a modality-wise contrastive and hierarchical distillation, transformer representation random distillation, along with other distillations to align representations tightly and effectively. We also utilise random erasing on individual data points within sequences during both training and distillation of teacher to lightly simulate scenario with missing visit information to foster effective knowledge transfer. As a result, NECHO v2 verifies itself by showing superiority in multimodal sequential diagnosis prediction on both balanced and imbalanced incomplete settings on multimodal healthcare data.