Phenotype-Based and Self-Learning Inter-Individual Sleep Apnea Screening With a Level IV-Like Monitoring System

Pattern recognition algorithm to identify detrusor overactivity on urodynamics

AIMS

Detrusor overactivity (DO) of the bladder is a finding on urodynamic studies (UDS) that often correlates with lower urinary tract symptoms and drives management. However, UDS interpretation remains nonstandardized. We sought to develop a mathematical model to reliably identify DO in UDS.

METHODS

We utilized UDS archive files for studies performed at our institution between 2013 and 2019. Raw tracings of vesical pressure, abdominal pressure, detrusor pressure, infused volume, and all annotations during UDS were obtained. Patients less than 1 year old, studies with calibration issues, or those with significant artifacts were excluded. In the training set, five representative DO patterns were identified. Candidate Pdet signal segments were matched to representative DO patterns. Manifold learning and dynamic time warping algorithms were used. Five-fold cross validation (CV) was used to evaluate the performance.

RESULTS

A total of 799 UDS studies were included. The median age was 9 years (range, 1-33). There were 1,742 DO events that did not overlap with annotated artifacts (cough, cry, valsalva, movements). The AUC of the training sets from the five-fold CV was 0.84 ± 0.01. The five-fold CV leads to an overall accuracy 81.35%, and sensitivity and specificity of detecting DO events are 76.92% and 81.41%, respectively, in the testing set.

CONCLUSIONS

Our predictive model using machine learning algorithms provides promising performance to facilitate automated identification of DO in UDS. This would allow for standardization and potentially more reliable UDS interpretation. Signal processing and machine learning interpretation of the other components of UDS are forthcoming.

Portable Sleep Apnea Syndrome Screening and Event Detection Using Long Short-Term Memory Recurrent Neural Network

Obstructive sleep apnea/hypopnea syndrome (OSAHS) is characterized by repeated airflow partial reduction or complete cessation due to upper airway collapse during sleep. OSAHS can induce frequent awake and intermittent hypoxia that is associated with hypertension and cardiovascular events. Full-channel Polysomnography (PSG) is the gold standard for diagnosing OSAHS; however, this PSG evaluation process is unsuitable for home screening. To solve this problem, a measuring module integrating abdominal and thoracic triaxial accelerometers, a pulsed oximeter (SpO₂) and an electrocardiogram sensor was devised in this study. Moreover, a long short-term memory recurrent neural network model is proposed to classify four types of sleep breathing patterns, namely obstructive sleep apnea (OSA), central sleep apnea (CSA), hypopnea (HYP) events and normal breathing (NOR). The proposed algorithm not only reports the apnea-hypopnea index (AHI) through the acquired overnight signals but also identifies the occurrences of OSA, CSA, HYP and NOR, which assists in OSAHS diagnosis. In the clinical experiment with 115 participants, the performances of the proposed system and algorithm were compared with those of traditional expert interpretation based on PSG signals. The accuracy of AHI severity group classification was 89.3%, and the AHI difference for PSG expert interpretation was 5.0±4.5. The overall accuracy of detecting abnormal OSA, CSA and HYP events was 92.3%.