The Beginning of the AI-Enabled Preventative PAP Therapy Era: A First-in-Human Proof of Concept Interventional Study.

Positive Airway Pressure (PAP) therapy is the most common and efficacious treatment for Obstructive Sleep Apnea (OSA). However, it suffers from poor patient adherence due to discomfort and may not fully alleviate all adverse consequences of OSA. Identifying abnormal respiratory events before they have occurred may allow for improved management of PAP levels, leading to improved adherence and better patient outcomes. Our previous work has resulted in the successful development of a Machine-Learning (ML) algorithm for the prediction of future apneic events using existing airflow and air pressure sensors available internally to PAP devices. Although researchers have studied the use of ML for the prediction of apneas, research to date has focused primarily on using external polysomnography sensors that add to patient discomfort and has not investigated the use of internal-to-PAP sensors such as air pressure and airflow to predict and prevent respiratory events. We hypothesized that by using our predictive software, OSA events could be proactively prevented while maintaining patients' sleep quality. An intervention protocol was developed and applied to all patients to prevent OSA events. Although the protocol's cool-down period limited the number of prevention attempts, analysis of 11 participants revealed that our system improved many sleep parameters, which included a statistically significant 31.6% reduction in Apnea-Hypopnea Index, while maintaining sleep quality. Most importantly, our findings indicate the feasibility of unobtrusive identification and unique prevention of each respiratory event as well as paving the path to future truly personalized PAP therapy by further training of ML models on individual patients.

Artificial Olfactory Signal Modulation for Detection in Changing Environments.

Animals have evolved to sense in complex environments through both modulation behavior including sniffing as well as sophisticated neural processing including memory and neuromodulation. Here, we explore thermal modulation of chemically diverse sensor arrays, where response patterns are based on partitioning of odorants across the array. The differential response patterns contain information about the chemical nature of the odorant for identification. By transitioning away from well-defined concentration modulation, traditionally used in the field, to thermal modulation, it is possible to capture both diagnostic patterns as well as intensity information in complex environments. This performance is demonstrated with carbon-black based, chemically diverse sensor arrays, that are thermally modulated with light at 25 mHz exposed to different analytes of varying concentrations.

Lung Mechanics Tracking with Forced Oscillation Technique (FOT) Based on CMOS Synchronous Demodulation Principle.

In this paper, an area-efficient CMOS integrated solution for lung impedance extraction is presented. The lock-in principle is leveraged for its high effective bandpass selectivity, to acquire information about the airways, through stimulation by FOT (Forced Oscillation Technique). The modulated pressure and flow signals are down-converted by a quadrature voltage commutating passive mixer-first receiver. In addition to its linearity, and unlike the Gilbert cell, it can be biased at zero dc current to alleviate flicker noise contributions. The proposed solution is designed and fabricated in 0.18µm TSMC technology. The chip occupies an active silicon area of 4.7 mm2 (including buffers and pads) and dissipates 429.63 µW. The proposed approach offers real time tracking of respiratory mechanics and is expected to be a promising solution for portable health monitoring and cost-effective biomedical devices.

High Frequency-Low Amplitude Oscillometry: Continuous Unobtrusive Monitoring of Respiratory Function on PAP Machines.

Oscillometry or Forced Oscillation Technique, traditionally used in intermittent clinical measurements, has recently gained substantial attention from its application as a continuous monitoring tool for large and small airways. However, low frequency (<8 Hz) continuous oscillometry faces high breathing noise, and hence requires high oscillation amplitudes to maintain an acceptable signal-to-noise ratio. Therefore, PAP machines that utilize low frequency oscillometry do so intermittently to distinguish airway patency several seconds after a breathing pause has occurred. We hypothesized that high frequency and low amplitude (HFLA) oscillometry may be as sensitive and applicable for monitoring upper airway patency to distinguish between central and obstructive apnea and hypopnea events, and for monitoring respiratory impedance. An inline oscillometry prototype device was developed and connected to commercial PAP machines to test whether oscillometry at 17, 43, and 79 Hz are as sensitive to airway patency as oscillometry at 4 Hz. Analysis of 11 patients with 171 apneas and hypopneas showed that all frequency oscillometry inputs were equally sensitive in distinguishing between central and obstructive apneas, while 17 Hz and 43 Hz oscillometry were most sensitive in distinguishing between central and obstructive hypopneas. Observations during normal breathing also showed the same periodicity and cross-correlation between impedance measurements from HFLA oscillometry compared to 4 Hz. Our findings provide an unobtrusive means of distinguishing airway patency during sleep and a means of continuous monitoring of respiratory function, with the potential for detection and prediction of developing respiratory diseases and significantly richer context for data analytics.

Practical Considerations for Accuracy Evaluation in Sensor-Based Machine Learning and Deep Learning.

Accuracy evaluation in machine learning is based on the split of data into a training set and a test set. This critical step is applied to develop machine learning models including models based on sensor data. For sensor-based problems, comparing the accuracy of machine learning models using the train/test split provides only a baseline comparison in ideal situations. Such comparisons won't consider practical production problems that can impact the inference accuracy such as the sensors' thermal noise, performance with lower inference quantization, and tolerance to sensor failure. Therefore, this paper proposes a set of practical tests that can be applied when comparing the accuracy of machine learning models for sensor-based problems. First, the impact of the sensors' thermal noise on the models' inference accuracy was simulated. Machine learning algorithms have different levels of error resilience to thermal noise, as will be presented. Second, the models' accuracy using lower inference quantization was compared. Lowering inference quantization leads to lowering the analog-to-digital converter (ADC) resolution which is cost-effective in embedded designs. Moreover, in custom designs, analog-to-digital converters' (ADCs) effective number of bits (ENOB) is usually lower than the ideal number of bits due to various design factors. Therefore, it is practical to compare models' accuracy using lower inference quantization. Third, the models' accuracy tolerance to sensor failure was evaluated and compared. For this study, University of California Irvine (UCI) 'Daily and Sports Activities' dataset was used to present these practical tests and their impact on model selection.

Time-Varying Respiratory Mechanics as a Novel Mechanism Behind Frequency Dependence of Impedance: A Modeling Approach.

Frequency dependence of respiratory mechanics is a well-established behavior of the respiratory system and is known to be an indicator of severity of obstructive disease, attributed to both tissue viscoelasticity and heterogeneity of airflow in the lung. Despite the fact that respiratory parameters are known to vary in time, often amplified in disease, all analysis methods assume stationarity or short-time stationarity in the parameters used to describe the respiratory system, and the effects of this assumption have not yet been examined in any detail. Here, using a generalized approach, we developed a theory for time-varying respiratory mechanics in time-frequency domain for analysis of linear time-varying systems, then, we analyzed the same respiratory system model with time-varying parameters in the time domain. Both time-frequency domain and time-domain derivations revealed a striking correlation between time-varying behavior of the respiratory system and frequency dependence of resistance. Remarkably, this phenomenon arose from the amplitude of time variations of the elastance. This links two mechanisms that are known to increase in obstructive disease: apparent low frequency increases in resistance and the time variations of reactance.