Attentive Prototype Learning with Wearable Sensor Mutual Information for Fall Risk Stratification of Parkinson’s Patients
Meng Zhang, Xuliang Ren, Jing Xu, Zhifen Guo, Qiumin Qu, Dongzhen Chen, Hongmei Cao · 2026
Read Attentive Prototype Learning with Wearable Sensor Mutual Information for Fall Risk Stratification of Parkinson’s Patients by Meng Zhang, Xuliang Ren, Jing Xu, Zhifen Guo, Qiumin Qu, Dongzhen Chen, Hongmei Cao (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13295412.txt.
Description
Meng Zhang and co-authors turn wearable motion sensors into a fall risk score for people with Parkinson's disease, in a 2026 exploratory study. Falls injure many patients, yet clinics still judge the risk largely by eye. Ninety-two patients wore inertial and photoelectric sensors on limbs and trunk while performing standard motor examination tasks. The team used mutual information, a measure of how much a signal reveals about a label, to link movement features to disease stage, which itself tracks fall risk.
The weighted combination of the most informative features became their fall risk stratification score. Adding that score to a classifier lifted test accuracy from one half to over four fifths, and a standard discrimination measure rose sharply as well. The authors caution that the work is exploratory, but they argue sensor-based scoring could give clinicians objective evidence for targeting prevention at patients with the shakiest balance.
Catalog record
| CREATOR | Meng Zhang, Xuliang Ren, Jing Xu, Zhifen Guo, Qiumin Qu, Dongzhen Chen, Hongmei Cao |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 35 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |