Summary of Detecting Daily Living Gait Amid Huntington’s Disease Chorea Using a Foundation Deep Learning Model, by Dafna Schwartz et al.
Detecting Daily Living Gait Amid Huntington’s Disease Chorea using a Foundation Deep Learning Model
by Dafna Schwartz, Lori Quinn, Nora E. Fritz, Lisa M. Muratori, Jeffery M. Hausdorff, Ran Gilad Bachrach
First submitted to arxiv on: 15 Dec 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary Wearable sensors offer a non-invasive way to collect physical activity (PA) data, with walking as a key component. Existing models often struggle to detect gait bouts in individuals with neurodegenerative diseases (NDDs) involving involuntary movements. A deep learning model called J-Net was developed, inspired by U-Net and fine-tuned for Huntington’s disease (HD) in-lab data, which processes wrist-worn accelerometer data to detect gait during daily living. The model achieved a 10-percentage point improvement in ROC-AUC for HD over existing methods, reaching 0.97 for in-lab data and showed no significant differences in median daily walking time between HD and controls. J-Net also correlated with the UHDRS-TMS clinical severity score (r=-0.52; p=0.02) and fine-tuning on Parkinson’s disease (PD) data improved gait detection over current methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Wearable sensors can track how much people move around, which is important for understanding diseases that affect movement. Right now, there are some problems with using these sensors to study certain types of diseases where the person’s movements might not be voluntary. A new computer program called J-Net was developed to help solve this problem. It uses information from a specific type of disease (Huntington’s) and can track how much people move around in their daily lives. The program did better than others at predicting when someone with Huntington’s was moving or not, and it also worked well for people with Parkinson’s. |
Keywords
» Artificial intelligence » Auc » Deep learning » Fine tuning
