Mtliquid enables efficient multi task learning for healthcare monitoring
MTLiquid: Enabling Efficient Multi-Task Learning using Liquid Neural Networks for Lightweight Healthcare Monitoring Systems
Machine LearningNeural and Evolutionary Computing
Summary
Healthcare devices often collect data like heart rate or temperature irregularly and need to predict multiple health issues at once. The authors developed MTLiquid, a new type of neural network that can handle several prediction tasks together while using less memory. It shares most of its calculations across tasks but has separate parts for each specific issue, making it both accurate and efficient. Their tests show MTLiquid predicts death risk and sepsis in ICU patients almost as well as specialized single-task models but uses much less computer memory.
What this means in practice
- •For hospital data teams: Improve ICU monitoring systems by running multi-task models that detect patient risks with less memory and maintain accuracy on mixed-interval data.
- •For wearable device developers: Build lightweight continuous monitoring apps that predict multiple health conditions simultaneously using fewer computational resources.$Commercial implications: Enables compact healthcare wearables offering multi-condition monitoring with efficient continuous-time prediction, attractive to consumer health markets.
Authors
Rachmad Vidya Wicaksana Putra, Fahad Abdul Rauf, Muhammad Shafique
Abstract
Continuous-time sensing and monitoring with timely and accurate decision-making are critical for many real-world applications. In healthcare monitoring systems, physiological signals are often available or sampled at irregular time intervals, hence requiring continuous-time processing to provide accurate prediction. Moreover, such systems often need to solve multiple detection/prediction tasks to provide a comprehensive patient review from different physiological aspects for more accurate decision-making. To solve this, continuous-time neural networks (CTNNs) can be employed. However, state-of-the-art works typically solve only one task at each network, thereby limiting their efficiency gains. To address this limitation, we propose MTLiquid, a novel methodology to enable efficient multi-task learning in continuous-time processing for healthcare monitoring systems through effective network design and training strategy. MTLiquid employs: (1) multiple input and output heads to accommodate different tasks, while sharing the same backbone network across tasks; as well as (2) an effective training strategy that leverages a loss-weighting technique to balance learning updates across different tasks and a proportional data presentation technique to address imbalanced dataset sizes. Experimental results for mortality prediction (P12) and sepsis early detection (P19) tasks for ICU patients show that, MTLiquid achieves strong performance (AUROC: 0.84 for P12 and 0.94 for P19) comparable to the state-of-the-art single-task learning in both continuous-time networks (AUROC: 0.84 for P12 and 0.95 for P19) and discrete-time networks (AUROC: 0.79-0.82 for P12 and 0.92-0.94 for P19), while incurring significantly smaller memory cost by 44%-94%. These results highlight the potential of our MTLiquid methodology to enable lightweight continuous-time healthcare monitoring systems for better decision-making.