Search

Pneumonia, asthma, sudden infant death syndrome (SIDS), and the most recent epidemic, COVID-19, are the most common lung diseases associated with respiratory difficulties. However, existing health monitoring systems use large and in-contact devices, which causes an uncomfortable experience. The difficulty in acquiring breathing signals for non-stationary individuals limits the use of ultra-wideband radar for breathing monitoring. This is due to ineffective signal clutter removal and body movement removal algorithms for collecting accurate breathing signals. This paper proposes a breathing signal analysis for non-contact physiological monitoring to improve quality of life. The radar-based sensors are used for collecting the breathing signal for each subject. The processed signal has been analyzed using continuous wavelet transform (CWT) and wavelet coherence with the Monte Carlo method. The finding shows that there is a significant difference between the three types of breathing patterns; normal, high, and slow. The findings may provide a comprehensive framework for processing and interpreting breathing signals, resulting in breakthroughs in respiratory healthcare, illness management, and overall well-being. [ABSTRACT FROM AUTHOR]

Breast cancer is one of the life-threatening cancer that leads to the most death due to cancer among women. Early diagnosis might help to reduce mortality. Thus, this research aims to study different approaches to the deep learning neural network model for breast cancer early detection for better prognosis. The performance of deep learning approaches such as Artificial Neural Network (ANN), Recurrent Neural Network (RNN) and Convolution Neural Network (CNN) is evaluated using the dataset from the University of Wisconsin. The findings show ANN achieved high accuracy of 99.9 % compared to others in detecting breast cancer. ANN can deliver better results with the provided dataset. However, more improvement is needed for better performance to ensure that the approach used is reliable enough for early breast cancer diagnosis. [ABSTRACT FROM AUTHOR]