Research
AI-based Neurological disease diagnosis
1. Video-Based Quantification of Gait Impairments in Parkinson ’s Disease Using Skeleton-Silhouette Fusion Convolution Network
We developed an automated video-based Parkinsonian gait assessment method using a novel skeleton-silhouette fusion convolution network. In addition, seven network-derived supplementary features, including critical aspects of gait impairment (gait velocity, arm swing, etc.), are extracted to provide continuous measures enhancing low-resolution clinical rating scales.
We are interviewed by CCTV（Click here to watch the video in CCTV）
Q. Zeng, P. Liu, N. Yu, J. Wu*, W. Huo*, J. Han*, “Video-based quantification of gait impairments in Parkinson’s disease using skeleton-silhouette fusion convolution network”, IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol.31, pp. 2912-2922, 2023.
2. Quantification of Motor Function Post-stroke using Novel Fusion of Wearable Inertial and Mechanomyographic Sensors
Subjective clinical rating scales represent the gold-standard for diagnosis of motor function following stroke. In practice however, they suffer from well-recognized limitations including assessor variance, low inter-rater reliability and low resolution. Automated systems have been proposed for empirical quantification but have not significantly impacted clinical practice. We address translational challenges in this arena through: (1) implementation of a novel sensor suite combining inertial measurement and mechanomyography (MMG) to quantify hand and wrist motor function; and (2) introduction of a new range of signal features extracted from the suite to supplement predicted clinical scores.
L. Formstone, W. Huo*, S. Wilson, A. McGregor, P. Bentley, and R.Vaidyanathan, “Quantification of Motor Function Post-stroke using Novel Fusion of Wearable Inertial and Mechanomyographic Sensors”, IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 29, pp. 1158 - 1167, 2021.
3. A Heterogeneous Sensing Suite for Multisymptom Quantification of Parkinson's Disease
In this study, we address all through a novel wearable sensor system and machine learning algorithms. The sensor system is composed of a forcesensor, three inertial measurement units (IMUs) and four custom mechanomyography (MMG) sensors. The system was tested in its capacity to predict Unified Parkinson’s Disease Rating Scale (UPDRS) scores based on quantitative assessment of bradykinesia, rigidity and tremor in PD patients. 23 PD patients were tested with the sensor system in parallel with exams conducted by treating clinicians and 10 healthy subjects were recruited as a comparison control group. Results prove the system accurately predicts UPDRS scores for all symptoms (85.4% match on average with physician assessment) and discriminates between healthy subjects and PD patients (96.6% on average). MMG features can also be used for remote monitoring of severity and fluctuations in PD symptoms out-of-clinic. This closed-loop feedback system enables individually tailored and regularly updated treatment, facilitating better outcomes for a very large patient population.
W. Huo*, P. Angeles, Y. Tai, N. Pavese, S.Wilson, M. T. Hu and R.Vaidyanathan, “A Heterogeneous Sensing Suite for Multisymptom Quantification of Parkinson's Disease”, IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 28, no.6, pp. 1397 - 1406, 2020.