Categorical Timeline Allocation and Alignment for Diagnostic Head Movement Tracking Feature Analysis

Fig. 3. Illustration of the pattern alignment matrix and the categorical sequence grouping process. The entries in the alignment matrix is 1 when the sequential pattern (indices of the rows) is included in the source sequence (indices of the columns), and 0 otherwise. The occurrence number of a sequential pattern in a diagnostic category is calculated as the sum of entries within the group of columns of this category.

Categorical Timeline Allocation and Alignment for Diagnostic Head Movement…

Atypical head movement pattern characterization is a potentially important cue for identifying children with autism spectrum disorder. In this paper, we implemented a computational framework for extracting the temporal patterns of head movement and utilizing the imbalance of temporal pattern distribution between diagnostic categories (e.g., children with or without autism spectrum disorder) as potential diagnostic cues. The timeline analysis results show a large number of temporal patterns with significant imbalances between diagnostic categories. The temporal patterns show strong classification power on discriminative and predictive analysis metrics. The long-time-span temporal patterns (e.g., patterns spanning 15-30 sec.) exhibit stronger discriminative capabilities compared with the temporal patterns with relatively shorter time spans. Temporal patterns with high coverage ratios (existing in a large portion of the video durations) also show high discriminative capacity.

Read more . . .

 

Mitsunori Ogihara, Zakia Hammal, Katherine B. Martin, Jeffrey F. Cohn, Justine Cassell, Gang Ren and Daniel S. Messinger. Categorical timeline allocation and alignment for diagnostic head movement tracking feature analysis, Workshop Paper, CVPR ’19 Workshop on Face and Gesture Analysis for Health Informatics (FGAHI), pp. 43 – 51, June, 2019