Abstract

Abstract
Background and Objectives: Early detection of cognitive impairment is crucial for timely clinical interventions aimed at delaying progression to dementia. However, existing screening tools are not ideal for wide population screening. This study explores the potential of combining machine learning, specifically, one-class classification, with simpler and quicker motor-cognitive tasks to improve the early detection of cognitive impairment. Methods: We gathered data on gait, fingertapping, cognitive, and dual tasks from older adults with mild cognitive impairment and healthy controls. Using one-class classification, we modeled the behavior of the majority group (healthy controls), identifying deviations from this behavior as abnormal. To account for confounding effects, we integrated confound regression into the classification pipeline. We evaluated the performance of individual tasks, as well as the combination of features (early fusion) and models (late fusion). Additionally, we compared the results with those from two-class classification and a standard cognitive screening test. Results: We analyzed data from 37 healthy controls and 16 individuals with mild cognitive impairment. Results revealed that one-class classification had higher predictive accuracy for mild cognitive impairment, whereas two-class classification performed better in identifying healthy controls. Gait features yielded the best results for one-class classification. Combining individual models led to better performance than combining features from the different tasks. Notably, the one-class majority voting approach exhibited a sensitivity of 87.5% and a specificity of 75.7%, suggesting it may serve as a potential alternative to the standard cognitive screening test. In contrast, the two-class majority voting failed to improve the low sensitivities achieved by the individual models due to the underrepresentation of the impaired group. Conclusion: Our preliminary results support the use of one-class classification with confound control to detect abnormal patterns of gait, fingertapping, cognitive, and dual tasks, to improve the early detection of cognitive impairment. Further research is necessary to substantiate the method's effectiveness in broader clinical settings.

Abstract
This study presents variability assessment of real time measurements from in-vivo internal joint loads with instrumented implant during post-operative (PO) recovery process from total hip arthroplasty on daily living gait activities. A total of 112 trials walking supported by crutches in both hands, contralateral and ipsilateral sides, walking on treadmill at constant velocities, accelerating, decelerating and free walking, were assessed from 9 different patients ranging 0.3 to 76-month PO. Variability was assessed based on standard deviation of the vertical joint load normalized to each subject body weight with this metric adequacy to monitor PO recover.

Abstract
In our previous work, we explored a new invisible ECG biometrics approach that uses signals collected at the thighs using polymeric dry electrodes and sensors integrated into a toilet seat. However, the performance of the biometric templates remains unexplored. In this paper we evaluate how the ECG templates evolve, and the impact that potential changes may have on performance, using one case-study subject monitored over 31 days. This work is organized into two main parts. The first explores the morphological and physical traits of the subject throughout the 31 days based on data collected daily, three times per day at 6-hour intervals; in more than 80% of the sessions, all the signals were successfully acquired without showing noise nor movement artefacts. The second part is focused on evaluating the performance of Support Vector Machine (SVM) and Binary Convolutional Neural Network (BCNN) classifiers in the identification of the case study subject within a population of 10 individuals, covering an age range of (24 to 35 years); the top performer was the BCNN, achieving a perfect accuracy rate of 100% when tested on a group of two individuals.

Abstract
The electrical stimulation of pedicle screws is a technique used to ensure its correct placement within the vertebrae pedicle. Several authors have studied these screws' electrical properties with the objective of understanding if they are a potential source of false negatives. As titanium screws are anodized with different thicknesses of a high electrical resistance oxide (TiO2), this study investigated, using analytical, numerical, and experimental methods, how its thickness may affect pedicle screw's resistance and conductivity. Analytical results have demonstrated that the thickness of the TiO2 layer does result in a significant radial resistance increase (44.21 m Omega/nm, for & Oslash; 4.5 mm), and a decrease of conductivity with layers thicker than 150 nm. The numerical approach denotes that the geometry of the screw further results in a decrease in the pedicle screw conductivity, especially after 125 nm. Additionally, the experimental results demonstrate that there is indeed an effective decrease in conductivity with an increase in the TiO2 layer thickness, which is also reflected in the screw's total resistance. While the magnitude of the resistance associated with each TiO2 layer thickness may not be enough to compromise the ability to use anodized pedicle screws with a high-voltage electrical stimulator, pedicle screws should be the subject of more frequent electrical characterisation studies.