Apnoea-Pi: Sleep disorder monitoring with open-source electronics and acoustics

Abstract

Apnoea is a sleep disorder that affects an increasing number of adults causing harm from fatigue to a growing chance of heart problems. Apnoea disorders can be treated but advanced monitoring and diagnosing tools are needed to identify its strand and offer adequate treatment. Therefore, Apnoea tracking is vital to help keep patients healthy. Sleep Apnoea can cause a number of conditions such as fatigue, high blood pressure, liver functionality and an increased risk of type 2 diabetes. These complications make it necessary to monitor as many potential patients as possible by designing an instrument that is accurate, comfortable to use, fit for purpose, cost effective and with embedded computation capabilities to store, process and transmit time series data. In this work we present Apnoea-Pi, an adaptation of our Acousto-Pi open source surface acoustic wave platform to monitor Apnoea in patients using ultrasonic humidity sensing.

Ubiquitous WiFi and Acoustic Sensing: Principles, Technologies, and Applications

“…[5] Phasebeat [7] RT-Fall [10] Liu  . [6] FarSense [32] MoBreath [51] WiPhone [50] SiFall [15] FILA [67] FIMD [66] Widar [57] mD-Track [59] Widar2.0 [58] DAFI [62] WiDE [68] Fidora [69] WiKey [18] WiFi-ID [54] WiWho [53] FingerPass [72] Nicolaou and Efstratiou [78] Amaging [87] SpiroSmart [88] SpiroSonic [90] Resptracker [91] BatMapper [94] Pradhan   [95] CARACAL [96] Dong  . [97] Kafle  .…”

“…Health-related applications based on acoustic mainly rely on the detection of physiological conditions, such as the capture of human breathing, heartbeat, and so on, or some event monitoring, such as the fall detection, abnormal sleep behaviors, and so on, as well as medical-related assistance control to assist body recovery, etc. Larson et al [88][89][90][91]142] captured physiological signals, such as human respiration, heartbeat, vital capacity, chest wall motion, and some other non-voice sounds, such as swallowing during eating. Based on these physiological signals, Nahdakumar et al [142] turned a cell phone into an active sonar system that emits modulated sound signals and detects breathing-induced minute movements of the chest and abdomen from reflexes, and developed the identification of various types of sleeps from sonar reflexes algorithms for apnea events, including obstructive apneas, central apneas, and hypopneas.…”

“…Based on this, many subsequent studies also emphasized on capturing heartbeat features by using machine learning techniques to classify the required research. The work [91] , continuous multi-person respiration tracking using an acoustic-based COTS device, employing a two-stage algorithm to separate and recombine respiration signals from multiple paths in a short period of time, can distinguish the respiration of at least four subjects within a distance of three meters. SpiroSmart [88] uses a built-in microphone for spirometry.…”

“…After obtaining the WiFi or acoustic signal, the next step is to characterize it using various sensing techniques. Similar to WiFi sensing, typical applications based on acoustic sensing include daily actions monitoring [73][74][75][76][77][78] , gesture and hand movements recognition [79][80][81][82][83][84][85][86][87] , health caring [88][89][90][91][92] , localization and navigation [93][94][95][96][97][98] and privacy and security [99][100][101][102][103][104][105][106][107][108][109][110][111] . In the following, we will also introduce the basic content of signals and other characterization methods.…”