Daily activity patterns from wearable accelerometry predict physical frailty and concern about falling
Jingyi Zhang, Jingtao Zhang, Peter Shull, Catherine Park, Shaoxiong Sun, Bijan Najafi, Changhong Wang · 2026
Read Daily activity patterns from wearable accelerometry predict physical frailty and concern about falling by Jingyi Zhang, Jingtao Zhang, Peter Shull, Catherine Park, Shaoxiong Sun, Bijan Najafi, Changhong Wang (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/PMC13612427.txt.
Description
Jingyi Zhang leads a seven-author team reading frailty off a pendant sensor in 2026. Frailty and fear of falling feed each other in older adults, yet both are usually measured by questionnaires needing a professional in the room. The team recorded forty-eight hours of chest acceleration from 146 participants and converted the activity streams into barcode-like patterns. Complexity metrics plus a bidirectional LSTM network fed a multi-task model that predicts both conditions at once.
The joint model scored F1 values of 93.12 percent for physical frailty and 86.27 percent for concern about falling. Because the two outcomes were learned together, the model could exploit their known interplay. The authors claim the first joint prediction of frailty and fall concern from real-world wearable data. Their pitch is practical: a cheap pendant worn for two days could flag elders before the spiral of fear, inactivity, and decline takes hold.
Catalog record
| CREATOR | Jingyi Zhang, Jingtao Zhang, Peter Shull, Catherine Park, Shaoxiong Sun, Bijan Najafi, Changhong Wang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 74129 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |