Modification of a Conventional Deep Learning Model to Classify Simulated Breathing Patterns: A Step toward Real-Time Monitoring of Patients with Respiratory Infectious Diseases.

The emergence of the global coronavirus pandemic in 2019 (COVID-19 disease) created a need for remote methods to detect and continuously monitor patients with infectious respiratory diseases. Many different devices, including thermometers, pulse oximeters, smartwatches, and rings, were proposed to monitor the symptoms of infected individuals at home. However, these consumer-grade devices are typically not capable of automated monitoring during both day and night. This study aims to develop a method to classify and monitor breathing patterns in real-time using tissue hemodynamic responses and a deep convolutional neural network (CNN)-based classification algorithm. Tissue hemodynamic responses at the sternal manubrium were collected in 21 healthy volunteers using a wearable near-infrared spectroscopy (NIRS) device during three different breathing conditions. We developed a deep CNN-based classification algorithm to classify and monitor breathing patterns in real time. The classification method was designed by improving and modifying the pre-activation residual network (Pre-ResNet) previously developed to classify two-dimensional (2D) images. Three different one-dimensional CNN (1D-CNN) classification models based on Pre-ResNet were developed. By using these models, we were able to obtain an average classification accuracy of 88.79% (without Stage 1 (data size reducing convolutional layer)), 90.58% (with 1 × 3 Stage 1), and 91.77% (with 1 × 5 Stage 1).

Optical Monitoring of Breathing Patterns and Tissue Oxygenation: A Potential Application in COVID-19 Screening and Monitoring.

The worldwide outbreak of the novel Coronavirus (COVID-19) has highlighted the need for a screening and monitoring system for infectious respiratory diseases in the acute and chronic phase. The purpose of this study was to examine the feasibility of using a wearable near-infrared spectroscopy (NIRS) sensor to collect respiratory signals and distinguish between normal and simulated pathological breathing. Twenty-one healthy adults participated in an experiment that examined five separate breathing conditions. Respiratory signals were collected with a continuous-wave NIRS sensor (PortaLite, Artinis Medical Systems) affixed over the sternal manubrium. Following a three-minute baseline, participants began five minutes of imposed difficult breathing using a respiratory trainer. After a five minute recovery period, participants began five minutes of imposed rapid and shallow breathing. The study concluded with five additional minutes of regular breathing. NIRS signals were analyzed using a machine learning model to distinguish between normal and simulated pathological breathing. Three features: breathing interval, breathing depth, and O2Hb signal amplitude were extracted from the NIRS data and, when used together, resulted in a weighted average accuracy of 0.87. This study demonstrated that a wearable NIRS sensor can monitor respiratory patterns continuously and non-invasively and we identified three respiratory features that can distinguish between normal and simulated pathological breathing.

Studying the Accuracy and Function of Different Thermometry Techniques for Measuring Body Temperature.

The purpose of this study was to determine which thermometry technique is the most accurate for regular measurement of body temperature. We compared seven different commercially available thermometers with a gold standard medical-grade thermometer (Welch-Allyn): four digital infrared thermometers (Wellworks, Braun, Withings, MOBI), one digital sublingual thermometer (Braun), one zero heat flux thermometer (3M), and one infrared thermal imaging camera (FLIR One). Thirty young healthy adults participated in an experiment that altered core body temperature. After baseline measurements, participants placed their feet in a cold-water bath while consuming cold water for 30 min. Subsequently, feet were removed and covered with a blanket for 30 min. Throughout the session, temperature was recorded every 10 min with all devices. The Braun tympanic thermometer (left ear) had the best agreement with the gold standard (mean error: 0.044 °C). The FLIR One thermal imaging camera was the least accurate device (mean error: -0.522 °C). A sign test demonstrated that all thermometry devices were significantly different than the gold standard except for the Braun tympanic thermometer (left ear). Our study showed that not all temperature monitoring techniques are equal, and suggested that tympanic thermometers are the most accurate commercially available system for the regular measurement of body temperature.