To develop an automatic, efficient, and interpretable system for assessing Parkinson's Tremor (PT) to assist in the pre-diagnosis of Parkinson's disease (PD).
Key Findings:
Achieved 91.3% accuracy and 80.0% F1-score in PT classification.
Achieved 76.4% accuracy and 76.7% F1-score in tremor rating classification.
Improved pose extraction performance by 25% using AlphaPose compared to previous methods.
Interpretation:
The proposed system demonstrates significant improvements in PT classification and tremor severity estimation, indicating its potential as a reliable tool for early PD diagnosis.
Limitations:
The system's performance may vary with different video quality and patient demographics.
The reliance on upper body poses may overlook other relevant tremor manifestations.
Conclusion:
The developed video-based deep learning system offers a promising non-intrusive method for assessing Parkinson's Tremor, potentially aiding in the early diagnosis of Parkinson's disease.
Baptist Health Foundation announced that it has received a $2 million donation from Anthony and Joyce Esernia to establish a new endowed chair at Baptist Health Miami Neuroscience Institute.
