Real-time cardiac monitoring through the development of a smart holter to detect pathologies through an expert algorithm (Preprint)

BACKGROUND Constant monitoring of the heart’s state is essential for the early detection of pathologies. In the past, this analysis could only be carried out in hospitals, using sophisticated equipment handled by qualified staff. Today there is a wide range of portable monitoring devices (Holters) on the market but with a set of drawbacks (low number of leads supported and their low signal quality among others) that make it difficult to consider them as a viable replacement for the equipment used in medical practice OBJECTIVE This article describes the process of designing and implementing a Smart Holter able to record up to six leads at the same time, providing a signal quality comparable to the equipment used in medical practice, but with the dimensions, consumption and ease of use of portable devices. We also describe the workings of the expert algorithm for detecting cardiac anomalies in real time monitoring, which is embedded in the device itself, and which is capable of detecting tachycardia, bradycardia, ischemia and atrial fibrillation episodes with a high success rate. METHODS The hardware developed, performs the acquisition of 6 leads simultaneously, with a signal quality comparable to that of equipment used in medical practice. Each of the signals are processed by the algorithm described in this paper. This algorithm decomposes each lead into each heartbeat and extracts each of the segments that compose it (QRS complex) as well as a series of additional parameters. Based on the duration, amplitude and different thresholds, the system is able to detect with a high success rate, the existence of a certain cardiac pathology. RESULTS Performance evaluation shows the capacity of the Smart Holter devised to offer a high quality signal that combined with an embedded expert algorithm is capable of detecting tachycardia, bradycardia, ischemia and atrial fibrillation episodes in real time with a high success rate. CONCLUSIONS Development presented in this paper offers better characteristics, because it resolves a wide range of drawbacks inherent in that type of portable medical equipment, mainly in terms of signal quality. One aspect to highlight is the improvement in noise immunity of the equipment. Although it is not possible to compensate large artefacts produced while playing sports, which still remains a challenge for current monitoring systems, an important step has been taken in the right direction to achieve even greater attenuation in the future, through the use of dynamic filtering systems controlled digitally, unlike the systems currently present on the market.