Integrating Expert ODEs into Neural ODEs: Pharmacology and Disease Progression

Modeling a system's temporal behaviour in reaction to external stimuli is a fundamental problem in many areas. Pure Machine Learning (ML) approaches often fail in the small sample regime and cannot provide actionable insights beyond predictions. A promising modification has been to incorporate expert domain knowledge into ML models. The application we consider is predicting the patient health status and disease progression over time, where a wealth of domain knowledge is available from pharmacology. Pharmacological models describe the dynamics of carefully-chosen medically meaningful variables in terms of systems of Ordinary Differential Equations (ODEs). However, these models only describe a limited collection of variables, and these variables are often not observable in clinical environments. To close this gap, we propose the latent hybridisation model (LHM) that integrates a system of expert-designed ODEs with machine-learned Neural ODEs to fully describe the dynamics of the system and to link the expert and latent variables to observable quantities. We evaluated LHM on synthetic data as well as real-world intensive care data of COVID-19 patients. LHM consistently outperforms previous works, especially when few training samples are available such as at the beginning of the pandemic. © 2021 Neural information processing systems foundation. All rights reserved.

Wireless Sensor Networks for Telerehabilitation of Parkinson’s Disease Using Rhythmic Auditory Stimulation

Parkinson’s Disease (PD) is a neurodegenerative disease affecting mainly the elderly. Patients affected by PD may experience slowness of movements, loss of automatic movements, and impaired posture and balance. Physical therapy is highly recommended to improve their walking where therapists instruct patients to perform big and loud exercises. Rhythmic Auditory Stimulation (RAS) is a method used in therapy where external stimuli are used to facilitate movement initiation and continuation. Aside from face-to-face therapy sessions, home rehabilitation programs are used by PD patients with mobility issues and who live in remote areas. Telerehabilitation is a growing practice amid the COVID-19 pandemic. This work describes the design and implementation of a wireless sensor network to remotely and objectively monitor the rehabilitation progress of patients at their own homes. The system, designed in consultation with a physical therapist, includes insole sensors which measure step parameters, a base station as a phone application which facilitates RAS training sessions and communication interface between the therapist and patients, and an online server storing all training results for viewing. Step data from the system’s real-time analysis were validated against post-processed and reconstructed signals from the raw sensor data gathered across different beats. The system has an accuracy of at least 80% and 72% for the total steps and correct steps respectively. © 2022, ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering.