Lucia Pitarch, Jordi Bernad, Lacramioara Dranca, Carlos Bobed Lisbona, Jorge Gracia. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.

BACKGROUND: has also been used to build these kinds of systems, but contradictory results have been reported: some works conclude that Kinect does not provide an accurate method of measuring gait kinematics variables, but others, on the contrary, report good accuracy results. METHODS: In this work, we have built a Kinect-based system that can distinguish between different PD stages, and have performed a clinical study with 30 patients suffering from PD belonging to three groups: early PD patients without axial impairment, more evolved PD patients with higher gait impairment but without Freezing of Gait (FoG), and patients with advanced PD and FoG. Those patients were recorded by two Kinect devices when they were walking in a hospital corridor. The datasets obtained from the Kinect were preprocessed, 115 features identified, some methods were applied to select the relevant features (correlation based feature selection, information gain, and consistency subset evaluation), and different classification methods (decision trees, Bayesian networks, neural networks and K-nearest neighbours classifiers) were evaluated with the goal of finding the most accurate method for PD stage classification. RESULTS: The classifier that provided the best results is a particular case of a Bayesian Network classifier (similar to a Naïve Bayesian classifier) built from a set of 7 relevant features selected by the correlation-based on feature selection method. The accuracy obtained for that classifier using 10-fold cross validation is 93.40%. The relevant features are related to left shin angles, left humerus angles, frontal and lateral bents, left forearm angles and the number of steps during spin. CONCLUSIONS: In this paper, it is shown that using Kinect is adequate to build a inexpensive and comfortable system that classifies PD into three different stages related to FoG. Compared to the results of previous works, the obtained accuracy (93.40%) can be considered high. The relevant features for the classifier are: a) movement and position of the left arm, b) trunk position for slightly displaced walking sequences, and c) left shin angle, for straight walking sequences. However, we have obtained a better accuracy (96.23%) for a classifier that only uses features extracted from slightly displaced walking steps and spin walking steps. Finally, the obtained set of relevant features may lead to new rehabilitation therapies for PD patients with gait problems.

In this paper we investigate if techniques based on decision trees are also useful to classify ischemia. In order to do that, we have based on a previous own algorithm that is able of detecting ST-segment episodes, can be executed in real time and is lightweight enough to be implemented on mobile devices such as PDAs. Three main steps are performed by that algorithm: 1) signal preprocessing that extracts important features from the ECG signal 2) detection of suspect ST segment events and 3) rejection of irrelevant ST segment events. We have found that techniques based on decision trees are interesting for the last step in order to obtain a set of rules that reject some of the suspect ST segment events detected in the second step. The freely available part of the LTST database has been used to develop the detector and the rest of the same database has been used for validation purposes. The sensitivity of the detector over all the records of the LTST database is 89.89% for episodes annotated according to C protocol and the positive predictivity is 70.03% for episodes annotated according to A protocol. Those results improve our previously obtained ones and we think that they are comparable to other results that appear in the literature. However, it has to be noticed that we have not found works that detect ischemia in real time and show validation results over the LTST database