Customized Textile Capacitive Insole Sensor for Center of Pressure Analysis.

Center of pressure refers to the centroid of the ground reaction force vector detected underneath the walking foot, which is a summary measure representing body segment movements during human locomotion. In this study, we developed a cost-effective, lightweight insole-type textile capacitive sensor (I-TCPs) to analyze plantar pressure (PP) distribution and center of pressure (COP) trajectory. To test the accuracy of I-TCPs, the measured pressure data was compared with that of F-scan. The sensor performance test was divided into a static baseline test and a dynamic gait experiment, both at two different gait speeds self-selected by the subjects. Static gait results showed that I-TCPs were capable of recognizing PP segments at different gait speeds. Dynamic gait results showed an average RMSE of 1.29 ± 0.47 mm in COPx (mediolateral shift) and 12.55 ± 5.08 mm in COPy (anteroposterior shift) at a comfortable gait speed. The COP correlation between I-TCPs and F-scan was 0.54 ± 0.09 in COPx and 0.92 ± 0.04 in COPy in comfortable gait speed conditions, in which COPy values presented a stronger correlation. RMSE and correlation in fast gait speed conditions also showed similar results. The findings of this study can be the basis for future research, including rehabilitation engineering, developing ICT devices, and creating smart wearable sensors to improve quality of life for patients and healthy individuals.

Deep Learning in Multi-Class Lung Diseases' Classification on Chest X-ray Images.

Chest X-ray radiographic (CXR) imagery enables earlier and easier lung disease diagnosis. Therefore, in this paper, we propose a deep learning method using a transfer learning technique to classify lung diseases on CXR images to improve the efficiency and accuracy of computer-aided diagnostic systems' (CADs') diagnostic performance. Our proposed method is a one-step, end-to-end learning, which means that raw CXR images are directly inputted into a deep learning model (EfficientNet v2-M) to extract their meaningful features in identifying disease categories. We experimented using our proposed method on three classes of normal, pneumonia, and pneumothorax of the U.S. National Institutes of Health (NIH) data set, and achieved validation performances of loss = 0.6933, accuracy = 82.15%, sensitivity = 81.40%, and specificity = 91.65%. We also experimented on the Cheonan Soonchunhyang University Hospital (SCH) data set on four classes of normal, pneumonia, pneumothorax, and tuberculosis, and achieved validation performances of loss = 0.7658, accuracy = 82.20%, sensitivity = 81.40%, and specificity = 94.48%; testing accuracy of normal, pneumonia, pneumothorax, and tuberculosis classes was 63.60%, 82.30%, 82.80%, and 89.90%, respectively.

Estimation of Knee Joint Angle Using Textile Capacitive Sensor and Artificial Neural Network Implementing with Three Shoe Types at Two Gait Speeds: A Preliminary Investigation.

The lower limb joints might be affected by different shoe types and gait speeds. Monitoring joint angles might require skill and proper technique to obtain accurate data for analysis. We aimed to estimate the knee joint angle using a textile capacitive sensor and artificial neural network (ANN) implementing with three shoe types at two gait speeds. We developed a textile capacitive sensor with a simple structure design and less costly placing in insole shoes to measure the foot plantar pressure for building the deep learning models. The smartphone was used to video during walking at each condition, and Kinovea was applied to calibrate the knee joint angle. Six ANN models were created; three shoe-based ANN models, two speed-based ANN models, and one ANN model that used datasets from all experiment conditions to build a model. All ANN models at comfortable and fast gait provided a high correlation efficiency (0.75 to 0.97) with a mean relative error lower than 15% implement for three testing shoes. And compare the ANN with A convolution neural network contributes a similar result in predict the knee joint angle. A textile capacitive sensor is reliable for measuring foot plantar pressure, which could be used with the ANN algorithm to predict the knee joint angle even using high heel shoes.

Deep Learning in Physiological Signal Data: A Survey.

Deep Learning (DL), a successful promising approach for discriminative and generative tasks, has recently proved its high potential in 2D medical imaging analysis; however, physiological data in the form of 1D signals have yet to be beneficially exploited from this novel approach to fulfil the desired medical tasks. Therefore, in this paper we survey the latest scientific research on deep learning in physiological signal data such as electromyogram (EMG), electrocardiogram (ECG), electroencephalogram (EEG), and electrooculogram (EOG). We found 147 papers published between January 2018 and October 2019 inclusive from various journals and publishers. The objective of this paper is to conduct a detailed study to comprehend, categorize, and compare the key parameters of the deep-learning approaches that have been used in physiological signal analysis for various medical applications. The key parameters of deep-learning approach that we review are the input data type, deep-learning task, deep-learning model, training architecture, and dataset sources. Those are the main key parameters that affect system performance. We taxonomize the research works using deep-learning method in physiological signal analysis based on: (1) physiological signal data perspective, such as data modality and medical application; and (2) deep-learning concept perspective such as training architecture and dataset sources.