GeneMamba: Early Parkinson’s Detection via Wearable Device and Genetic Data
Keywords: Accelerometer, Parkinson's Disease, Genetic Variants, Mamba, Time-series data
Abstract: Parkinson's disease (PD) is a progressive neurodegenerative disorder affecting millions worldwide, with its prevalence expected to rise as the global population ages. Early diagnosis is crucial for effective management and improved quality of life for patients. However, current accelerometer-based studies focus more on detecting the symptoms of PD, while less research has been conducted on early detection of PD. This study presents a novel multi-modal deep learning model named GeneMamba for early PD diagnosis, using state space modelling approaches to effectively analyze sequences and combining accelerometer data from wearable devices with genetic variants data. Our model predicts early PD occurrence up to 7 years before clinical onset, outperforming existing methods. Furthermore, through knowledge transfer, we enable accurate PD prediction using only wearable device data, enhancing our model's real-world applicability. Additionally, our interpretation methods uncover both established and previously unidentified genes associated with PD, advancing our understanding of the disease's genetic architecture and potentially highlighting new therapeutic targets. Our approach not only advances early PD diagnosis but also offers insights into the disease's etiology, paving the way for improved risk assessment and personalized interventions.
Primary Area: learning on time series and dynamical systems
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 12995
Loading