Providing an easy‐to‐use and non‐invasive heart rate estimation system for home healthcare is imperative for detecting health deterioration in the early stages and for providing preventive treatments. To this end, we propose a ballistocardiographic method that estimates heart rate by analysing face videos and exploiting cyclic head motions caused by blood pumping through the carotid arteries during heartbeats. The proposed method addresses hand motion artefacts that occur while tracking head motions on a mobile platform. This is achieved by training an auto‐encoder that retrieves a heart rate‐related signal by learning consistent and statistically dominant relationships between facial landmarks. Moreover, unlike conventional methods, where a feature point tracker was used to track the facial landmarks, we propose using region‐based tracking to increase tracking robustness in challenging lighting conditions. Experimental results demonstrate that the proposed method performs best in challenging light conditions and can robustly manage hand motion artefacts on a mobile platform.

This study presents several new approaches to analyze the non-invasively recorded His-Purkinje system (HPS) signals fiom patients with myotonic muscular dystrophy.A high resolution electrocardiogram based on signal averaging to improve the signal-to-noise ratio(SNR) is a well established means to record HPS potentials. These new approaches used methods to temporally and spectrally separate the HPS potentials from the P wave potentials. These included both physiologically based and signal processing based schemes. Separating or shifting the P wave fiom the HPS potentials using heart rate dependent averaging and the addition of several highpass filtering methods proved somewhat, but not totally successful. In the group of patients with sequential recordings over a period of two years the progression of their muscular dystrophy may also be seen in the heart as well. This may then produce noticeable progressive trends or changes in their HPS waveforms over time. The most noticeable changes found in this study were temporal changes and morphological changes of the HPS activity of these patients over time. therapy. Others have looked at H P S signals in patients with Chagas disease, another progressive disease in the HPS, in an attempt to similarly quantify progression to heart block[4].The new approaches for H P S enhancement proposed in this study were primarily developed to temporally and spectrally separate the HPS from the P wave potentials.These included both physiologically and signal processing based schemes. Separating the P wave from the HPS potentials was done by using heart rate dependent averaging [5]. The highpass filtering methods used varying filter orders, types, and comer frequencies.One aim of the signal processing techniques was to apply filters and preserve the PR segment authenticity for precise signal measurement in the time domain. This approach required accurate identification of the QRS onset which served as a temporal pivot point to fold or create the mirror image of the PR segment. Hence both forward time and reverse time highpass filters were used enhance the HPS signals. 2.Methods