On Selection of Threshold Values for Symbolic Entropy Analysis of Human Gait.

Human gait is a complex system affected by many other processes of human physiology. It has multiple inputs and multiple outputs. Due to its complex nature, signals obtained from this system also exhibit complexity and variability. It has been analyzed in many ways to extract the information inhabited by these signals. Entropy based methods showed a significant impact on analysis of gait signals. Threshold based symbolic entropy analysis is one of the entropy based method applied to human gait signals. In this method Normalized Corrected Shannon Entropy (NCSE) is calculated to compare the spontaneous output of the human locomotors system during different walking conditions. Selection of the threshold values is an important task and sometimes it depends upon the type and size of the data. Results are dependent on the proper selection of the threshold. In this paper, different threshold selection methods are discussed and their impact on the results are presented. It was observed that, variation in stride interval has performed better as a threshold value as compare to the other methods. It provided maximum separation among different groups of gait data used in this study. We concluded with the recommendations for the proper selection of the threshold values to apply symbolic entropy methods on human gait signals. Clinical relevance Various gait related problems are common in older adults that increase with age and are associated with reduced gait speed increased fall risk and other impairments. Consequently objective gait assessment in the clinics depending upon the size of the available data has become increasingly important for the classification of gait. It was found that while applying symbolic entropy method proper selection of threshold result into improved classification of different types of gait data which will help the clinician for better decision-making regarding treatment.

CNN-Based Classification of Craniosynostosis Using 2D Distance Maps.

Craniosynostosis is a condition associated with the premature fusion of skull sutures affecting infants. 3D photogrammetric scans are a promising alternative to computed tomography scans in cases of single suture or nonsyndromic synostosis for diagnostic imaging, but oftentimes diagnosis is not automated and relies on additional cephalometric measure-ments and the experience of the surgeon. We propose an alternative representation of the infant's head shape created from 3D photogrammetric surface scans as 2D distance maps. Those 2D distance maps rely on ray casting to extract distances from a center point to the head surface, arranging them into a 2D image grid. We use the distance map for an original convolutional neural network (CNN)-based classification approach, which is evaluated on a publicly available synthetic dataset for benchmarking and also tested on clinical data. Qualitative differences of different head shapes can be ob-served in the distance maps. The CNN-based classifier achieves accuracies of 100 % on the publicly available synthetic dataset and 98.86 % on the clinical test set. Our distance map approach demonstrates the diagnostic value of 3D photogrammetry and the possibility of automatic, CNN-based diagnosis. Future steps include the improvement of the mapping method and testing the CNN on more pathologies.