Data-Driven Classifiers for Early Meal Detection Using ECG

Review Chemistry, Analytical

Unsupervised Anomaly Detection for IoT-Based Multivariate Time Series: Existing Solutions, Performance Analysis and Future Directions

Mohammed Ayalew Belay et al.

Summary: The recent wave of digitalization involves the widespread use of sensors in various environments, contributing to the advancement of full autonomy in industrial scenarios. The analysis of multivariate time series data produced by sensors plays a crucial role in identifying normal and abnormal operational conditions. However, the simultaneous analysis of temporal patterns within sensors and spatial dependencies between sensors poses a challenge in Multivariate Time Series Anomaly Detection (MTSAD). Unsupervised MTSAD techniques, including deep learning, have been developed to address this challenge. This article provides an extensive review of the current state of the art in MTSAD, along with a numerical evaluation of 13 promising algorithms on publicly available datasets, highlighting their advantages and shortcomings.

SENSORS (2023)

添加到收藏夹

Proceedings Paper Engineering, Electrical & Electronic

Comparison of Different Classifiers for Early Meal Detection Using Abdominal Sounds

Muhammad Asaad Cheema et al.

Summary: This paper compares different classifiers for early meal detection using abdominal sounds. The results show that neural networks and convolutional neural networks provide shorter detection time and no false positives. Early and reliable meal detection can alleviate the mental burden of diabetic patients and reduce the risk of hypoglycemia.

2022 IEEE 12TH SENSOR ARRAY AND MULTICHANNEL SIGNAL PROCESSING WORKSHOP (SAM) (2022)

添加到收藏夹

Article Engineering, Electrical & Electronic

Multiscale Deep Neural Network for Obstructive Sleep Apnea Detection Using RR Interval From Single-Lead ECG Signal

Qi Shen et al.

Summary: This study proposed an OSA detection method based on a multiscale dilation attention 1-D convolutional neural network and a weighted-loss time-dependent classification model. By introducing dilated convolution and attention mechanism, as well as combining weighted loss function and hidden Markov model, this method achieved high accuracy in identifying sleep apnea.

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT (2021)

添加到收藏夹

Article Computer Science, Information Systems