Wearable Sensors and MET Features: The importance of Preprocessing and Understanding the Data

Advanced Human Activity Recognition Using Wearable Sensors

This study reviews pre-2018 developments and proposes an improved HAR framework incorporating advanced feature engineering, sensor fusion, and ensemble learning techniques, which shows strong potential in healthcare, fitness, and smart environments.

Generalizable Physical Therapy Exercise Classification from Wearable IMU Data Using LOSO Evaluation

In this work, we propose a system to recognize human activities using the accelerometer and gyroscope of a smartphone, which has the potential to be used as a tool to monitor health conditions in real time. We use the UCI HAR dataset to assess the performance of different classifiers, such as Decision Tree, Naive Bayes, SVM, Random Forest, ANN, and XGBoost, as well as the effect of window sizes, as well as the generalization ability of the proposed system using the Leave-One-Subject-Out method. The ensemble classifiers, such as the XGBoost model, have the best performance among the classifiers used, with a maximum F1 score of 98.1%, as shown by the feature importance plot, which indicates the effectiveness of the combined time- and frequency-domain features used in the proposed

Modeling Human Activity Recognition with Deep Convolutional Neural Networks and Wearable Sensor Data

A deep convolutional neural network (DCNN) based method for recognizing human activities using body-worn sensors’ time-series data after an enormous data analysis on the data.

Machine Learning-Based Mental Health Classification Using Physiological and Physical Activity Data from Wearable Sensors

Mental health profoundly impacts individuals’ quality of life, productivity, and holistic well-being. The early identification of mental health disorders remains problematic, owing largely to the dependence on subjective evaluation methods. This study addresses these limitations by establishing a machine-learning-based mental health classification framework that leverages physiological and physical activity metrics from wearable sensors. Key physiological features, including heart rate, heart rate variability (HRV), sleep quality, and stress levels, along with physical activity data, were systematically collected and preprocessed via cleaning, normalization, and encoding. The performance of three distinct machine learning algorithms, namely Random Forest, Support Vector Machine (SVM), and K-Nearest Neighbor (KNN), was evaluated using accuracy, precision, recall, and F1 score. Empirical results revealed that the Random Forest model attained the highest classification performance, with accuracy, precision, recall, and F1-score values of 96.34%, 96.37%, 96.34%, and 96.34%, respectively, outperforming both SVM and KNN models. The findings underscore the utility of multimodal physiological and behavioral data from wearable devices as objective markers for mental health status. By integrating physiological indicators, activity patterns, and psychological assessments into a cohesive machine learning architecture, this research advance’s objective, continuous mental health monitoring.

Non-Contact Vital Signs Estimation Ml and Dl Techniques: A Comprehensive Survey

In the past decade,health care monitoring has emerged as the most promising area of research in the field of medicine.Vital sign signals have rapidly become a dominant factor with the recent advancement of healthcare applications. The Human Vital Signs (HVS), which are utilized for predicting medical and physical health issues earlier, are the essential ones that reveals the actual health status of the patients.Recent research works utilizes techniques like Electroencephalogram (EEG), Photoplethysmography (PPG), wireless sensors, Internet of Things (IoT), Electrocardiogram (ECG), and Remote-PPG (RPPG) However, a prominent role is played by the RPPG signal in HVS estimation. This paper aims to explore contactless HVS monitoring utilizing RPPG signals and also investigate the numerous Machine Learning (ML) and Deep Learning (DL) approaches that are employed for HVS prediction. Likewise, by evaluating certain quality metrics, the performance of the different techniques is validated. Further the survey examines the related models and  the best techniques for non-contact HVS measurement. The key objective of this study is to provide a  detailed insight into the purpose of HVS prediction.

for Human Activity Recognition: Overview, Advances, and Challenges