An Explainable Multi-Modal Neural Network Architecture for Predicting Epilepsy Comorbidities Based on Administrative Claims Data

Abstract

1AbstractEpilepsy is a complex brain disorder characterized by repetitive seizure events. Epilepsy patients often suffer from various and severe physical and psychological co-morbidities (e.g. anxiety, migraine, stroke, etc.). While general comorbidity prevalences and incidences can be estimated from epidemiological data, such an approach does not take into account that actual patient specific risks can depend on various individual factors, including medication. This motivates to develop a machine learning approach for predicting risks of future comorbidities for the individual epilepsy patient.In this work we use inpatient and outpatient administrative health claims data of around 19,500 US epilepsy patients. We suggest a dedicated multi-modal neural network architecture (Deep personalized LOngitudinal convolutional RIsk model - DeepLORI) to predict the time dependent risk of six common comorbidities of epilepsy patients. We demonstrate superior performance of DeepLORI in a comparison with several existing methods Moreover, we show that DeepLORI based predictions can be interpreted on the level of individual patients. Using a game theoretic approach, we identify relevant features in DeepLORI models and demonstrate that model predictions are explainable in the light of existing knowledge about the disease. Finally, we validate the model on independent data from around 97,000 patients, showing good generalization and stable prediction performance over time.

Multimodal explainable AI predicts upcoming speech behavior in adults who stutter

“…In multimodal deep learning, neural networks are used to integrate, fuse, and learn complementary representations from multiple input domains (Ngiam et al, 2011). Recent work has successfully fused images and text (Abavisani et al, 2020), detected adverse weather by combining different types of sensor information (Bijelic et al, 2020), estimated the 3-D surface of faces (Abrevaya et al, 2020), and combined information from multiple drug and diagnosis domains (Linden et al, 2021).…”