Clinically Informed Automated Assessment of Finger Tapping Videos in Parkinson's Disease

Abstract

The utilization of Artificial Intelligence (AI) for assessing motor performance in Parkinson's Disease (PD) offers substantial potential, particularly if the results can be integrated into clinical decision-making processes. However, the precise quantification of PD symptoms remains a persistent challenge. The current standard Unified Parkinson's Disease Rating Scale (UPDRS) and its variations serve as the primary clinical tools for evaluating motor symptoms in PD, but are time-intensive and prone to inter-rater variability. Recent work has applied data-driven machine learning techniques to analyze videos of PD patients performing motor tasks, such as finger tapping, a UPDRS task to assess bradykinesia. However, these methods often use abstract features that are not closely related to clinical experience. In this paper, we introduce a customized machine learning approach for the automated scoring of UPDRS bradykinesia using single-view RGB videos of finger tapping, based on the extraction of detailed features that rigorously conform to the established UPDRS guidelines. We applied the method to 75 videos from 50 PD patients collected in both a laboratory and a realistic clinic environment. The classification performance agreed well with expert assessors, and the features selected by the Decision Tree aligned with clinical knowledge. Our proposed framework was designed to remain relevant amid ongoing patient recruitment and technological progress. The proposed approach incorporates features that closely resonate with clinical reasoning and shows promise for clinical implementation in the foreseeable future.

Keywords: Parkinson’s disease; UDPRS quantification; data-driven; finger tapping; machine learning.

Figure 1

The main structure of the proposed framework. The pipeline consists of three distinct cycles: data cycle, model learning cycle, and algorithm deployment cycle.

Figure 2

Data preprocessing flowchart.

Figure 3

Illustration of the hand keypoints.

Figure 4

Depiction of the time series data preprocessing pipeline. The input time series, denoted as $\mathcal{I}$, corresponds to the distance between $K_4$ and $K_8$ during the finger tapping task. The X−axis indicates the frame index within an image sequence, reflecting the duration of the finger tapping action. The Y−axis represents the normalized amplitude of finger movement, adjusted according to the palm size of each subject.

Figure 5

Extraction of $F_{amp-\alpha 1}$, $F_{amp-\alpha 2}$, and $F_{amp-bp}$.

Figure 6

Illustration of halt and hesitation. The red row indicates the precise moment when the halt and hesitation occur.

Figure 7

Correlations between the 15 extracted features and UPDRS scores.

Figure 8

Structure of the proposed Decision Tree.

Figure 9

Confusion matrix of the predicted score labels and the true labels.