Sleep Apnea Syndrome (SAS) is highly prevalent worldwide and severely affects quality of life. SAS screening with self-reported symptoms, upper-airway features, physical exam and questionnaires have been reported with sensitivity but poor specificity. Challenges with polysomnography (PSG) for SAS screening include high operating costs, inadequate availability and limited repeatability. We present a novel SAS screening tool using the Vital Connect HealthPatch™ sensor.

Unobtrusive continuous monitoring of important vital signs and activity metrics has the potential to provide remote health monitoring, at-home screening, and rapid notification of critical events such as heart attacks, falls, or respiratory distress. This paper contains validation results of a wireless Bluetooth Low Energy (BLE) patch sensor consisting of two electrocardiography (ECG) electrodes, a microcontroller, a tri-axial accelerometer, and a BLE transceiver. The sensor measures heart rate, heart rate variability (HRV), respiratory rate, posture, steps, and falls and was evaluated on a total of 25 adult participants who performed breathing exercises, activities of daily living (ADLs), various stretches, stationary cycling, walking/running, and simulated falls. Compared to reference devices, the heart rate measurement had a mean absolute error (MAE) of less than 2 bpm, time-domain HRV measurements had an RMS error of less than 15 ms, respiratory rate had an MAE of 1.1 breaths per minute during metronome breathing, posture detection had an accuracy of over 95% in two of the three patch locations, steps were counted with an absolute error of less than 5%, and falls were detected with a sensitivity of 95.2% and specificity of 100%...

There has been a growing interest in out-of-center sleep testing with portable devices for accurate diagnosis of sleep apnea syndrome. This paper presents a new algorithm that extracts features based on filtering and statistical dispersion of the nasal airflow respiration signal and detects apnea events on a per-second basis. The data records were randomly selected from the Sleep Heart Health Study (SHHS-2) database to represent 100 control subjects with Apnea-Hypopnea Index (AHI) less than 5, and 100 apnea subjects with AHI values from 30 to 75. The algorithm was optimized according to the product of sensitivity and positive predictive value of apnea events among a training dataset of 50 apnea subjects with a constraint on the false positive rate among a training dataset of 50 control subjects. From testing of the algorithm on separate datasets, the false positive rate among 50 control subjects was found to be 1.3 events per hour, which corresponds to 100% specificity of classifying apnea subjects. The sensitivity and positive predictive value among 50 apnea subjects were found to be 83.6% and 72.3%, respectively. Among the identified false positive events in the apnea subjects, 64.3% of the events were found to be hypopnea events. Thus, incorporation of hypopnea detection would enhance the performance of the apnea detection algorithm...

Continuous monitoring of respiratory rate in am- bulatory conditions has widespread applications for screening of respiratory diseases and remote patient monitoring. Unfortunately, minimally obtrusive techniques often suffer from low accuracy. In this paper, we describe an algorithm with low computational complexity for combining multiple respiratory measurements to estimate breathing rate from an unobtrusive chest patch sensor. Respiratory rates derived from the respi- ratory sinus arrhythmia (RSA) and modulation of the QRS amplitude of electrocardiography (ECG) are combined with a respiratory rate derived from tri-axial accelerometer data. The three respiration rates are combined by a weighted average using weights based on quality metrics for each signal. The algorithm was evaluated on 15 elderly subjects who performed spontaneous and metronome breathing as well as a variety of activities of daily living (ADLs). When compared to a reference device, the mean absolute error was 1.02 breaths per minute (BrPM) during metronome breathing, 1.67 BrPM during spontaneous breathing, and 2.03 BrPM during ADLs....

This paper describes an adhesive sensor system worn on the skin that automatically detects human falls. The sensor, which consists of a tri-axial accelerometer, a microcontroller and a Bluetooth Low Energy transceiver, can be worn anywhere on a subject’s torso and in any orientation. In order to distinguish easily between falls and activities of daily living (ADL), a possible fall is detected only if an impact is detected and if the subject is horizontal shortly afterwards. As an additional criterion to reduce false positives, a fall is confirmed if the user activity level several seconds after a possible fall is below a threshold. Intentional falls onto a gymnastics mat were performed by 10 volunteers (total of 297 falls); ADL were performed by 15 elderly volunteers (total of 315 ADL). The fall detection algorithm provided a sensitivity of 99% and a specificity of 100%..