Comparison Between a Single-Lead ECG Garment Device and a Holter Monitor: A Signal Quality Assessment.

Wearable electronics are increasingly common and useful as health monitoring devices, many of which feature the ability to record a single-lead electrocardiogram (ECG). However, recording the ECG commonly requires the user to touch the device to complete the lead circuit, which prevents continuous data acquisition. An alternative approach to enable continuous monitoring without user initiation is to embed the leads in a garment. This study assessed ECG data obtained from the YouCare device (a novel sensorized garment) via comparison with a conventional Holter monitor. A cohort of thirty patients (age range: 20-82 years; 16 females and 14 males) were enrolled and monitored for twenty-four hours with both the YouCare device and a Holter monitor. ECG data from both devices were qualitatively assessed by a panel of three expert cardiologists and quantitatively analyzed using specialized software. Patients also responded to a survey about the comfort of the YouCare device as compared to the Holter monitor. The YouCare device was assessed to have 70% of its ECG signals as "Good", 12% as "Acceptable", and 18% as "Not Readable". The R-wave, independently recorded by the YouCare device and Holter monitor, were synchronized within measurement error during 99.4% of cardiac cycles. In addition, patients found the YouCare device more comfortable than the Holter monitor (comfortable 22 vs. 5 and uncomfortable 1 vs. 18, respectively). Therefore, the quality of ECG data collected from the garment-based device was comparable to a Holter monitor when the signal was sufficiently acquired, and the garment was also comfortable.

Machine learning-enhanced noninvasive prenatal testing of monogenic disorders.

OBJECTIVE: Single-nucleotide variants (SNVs) are of great significance in prenatal diagnosis as they are the leading cause of inherited single-gene disorders (SGDs). Identifying SNVs in a non-invasive prenatal screening (NIPS) scenario is particularly challenging for maternally inherited SNVs. We present an improved method to predict inherited SNVs from maternal or paternal origin in a genome-wide manner. METHODS: We performed SNV-NIPS based on the combination of fragments of cell free DNA (cfDNA) features, Bayesian inference and a machine-learning (ML) prediction refinement step using random forest (RF) classifiers trained on millions of non-pathogenic variants. We next evaluate the real-world performance of our refined method in a clinical setting by testing our models on 16 families with singleton pregnancies and varying fetal fraction (FF) levels, and validate the results over millions of inherited variants in each fetus. RESULTS: The average area under the ROC curve (AUC) values are 0.996 over all families for paternally inherited variants, 0.81 for the challenging maternally inherited variants, 0.86 for homozygous biallelic variants and 0.95 for compound heterozygous variants. Discriminative AUCs were achieved even in families with a low FF. We further investigate the performance of our method in correctly predicting SNVs in coding regions of clinically relevant genes and demonstrate significantly improved AUCs in these regions. Finally, we focus on the pathogenic variants in our cohort and show that our method correctly predicts if the fetus is unaffected or affected in all (10/10, 100%) of the families containing a pathogenic SNV. CONCLUSIONS: Overall, we demonstrate our ability to perform genome-wide NIPS for maternal and homozygous biallelic variants and showcase the utility of our method in a clinical setting.

Electrocardiogram Monitoring Wearable Devices and Artificial-Intelligence-Enabled Diagnostic Capabilities: A Review.

Worldwide, population aging and unhealthy lifestyles have increased the incidence of high-risk health conditions such as cardiovascular diseases, sleep apnea, and other conditions. Recently, to facilitate early identification and diagnosis, efforts have been made in the research and development of new wearable devices to make them smaller, more comfortable, more accurate, and increasingly compatible with artificial intelligence technologies. These efforts can pave the way to the longer and continuous health monitoring of different biosignals, including the real-time detection of diseases, thus providing more timely and accurate predictions of health events that can drastically improve the healthcare management of patients. Most recent reviews focus on a specific category of disease, the use of artificial intelligence in 12-lead electrocardiograms, or on wearable technology. However, we present recent advances in the use of electrocardiogram signals acquired with wearable devices or from publicly available databases and the analysis of such signals with artificial intelligence methods to detect and predict diseases. As expected, most of the available research focuses on heart diseases, sleep apnea, and other emerging areas, such as mental stress. From a methodological point of view, although traditional statistical methods and machine learning are still widely used, we observe an increasing use of more advanced deep learning methods, specifically architectures that can handle the complexity of biosignal data. These deep learning methods typically include convolutional and recurrent neural networks. Moreover, when proposing new artificial intelligence methods, we observe that the prevalent choice is to use publicly available databases rather than collecting new data.

Automated detection of atrial fibrillation based on vocal features analysis.

INTRODUCTION: Early detection of atrial fibrillation (AF) is desirable but challenging due to the often-asymptomatic nature of AF. Known screening methods are limited and most of them depend of electrocardiography or other techniques with direct contact with the skin. Analysis of voice signals from natural speech has been reported for several applications in medicine. The study goal was to evaluate the usefulness of vocal features analysis for the detection of AF. METHODS: This prospective study was performed in two medical centers. Patients with persistent AF admitted for cardioversion were enrolled. The patients pronounced the vowels "Ahh" and "Ohh" were recorded synchronously with an ECG tracing. An algorithm was developed to provide an "AF indicator" for detection of AF from the speech signal. RESULTS: A total of 158 patients were recruited. The final analysis of "Ahh" and "Ohh" syllables was performed on 143 and 142 patients, respectively. The mean age was 71.4 ± 9.3 and 43% of patients were females. The developed AF indicator was reliable. Its numerical value decreased significantly in sinus rhythm (SR) after the cardioversion ("Ahh": from 13.98 ± 3.10 to 7.49 ± 1.58; "Ohh": from 11.39 ± 2.99 to 2.99 ± 1.61). The values at SR were significantly more homogenous compared to AF as indicated by a lower standard deviation. The area under the receiver operating characteristic curve was >0.98 and >0.89 ("Ahh" and "Ohh," respectively, p < .001). The AF indicator sensitivity is 95% with 82% specificity. CONCLUSION: This study is the first report to demonstrate feasibility and reliability of the identification of AF episodes using voice analysis with acceptable accuracy, within the identified limitations of our study methods. The developed AF indicator has higher accuracy using the "Ahh" syllable versus "Ohh."

Implementation of a Personalized Digital App for Pediatric Preanesthesia Evaluation and Education: Ongoing Usability Analysis and Dynamic Improvement Scheme.

BACKGROUND: Preanesthesia evaluation is a basic practice preceding any surgical procedure, aimed at tailoring individualized anesthetic plans for patients, improving safety, and providing patients with educational knowledge and tools in preparation for the surgery day. In the last 2 decades, eHealth and mobile health (mHealth) settings have gradually replaced part of the face-to-face encounters as the platform for preanesthesia communication between doctor and patient, yielding a range of benefits as demonstrated in recent publications. Nevertheless, there is a lack of studies examining the effectiveness of surgical mHealth apps focusing on the pediatric preanesthetic setting and addressing their usability among families. OBJECTIVE: This study describes a dynamic approach for the development process of GistMD's preanesthesia mHealth system, a mobile-based educational and management system designed for the pediatric setting. METHODS: The study was conducted in 4 departments at a 1500-bed quaternary, academic medical center in Tel Aviv, Israel. During the study period, the link to the preanesthesia system was sent via SMS text messages to families whose children were about to undergo surgery. The system included preanesthesia questionnaires, educational videos, downloadable instructions, and consent forms. Continuous collection and examination of usability data were conducted during the implementation term including responsiveness, effectiveness, and satisfaction indicators. The information collected in each stage was used to draw conclusions regarding potential usability gaps of the system and to plan product adjustments for the following period. RESULTS: During 141 days of implementation, the link to the GistMD preanesthesia management system was sent to 769 families, and product-fit actions were implemented during this term: (1) changing text message scheduling for addressing learnability and accessibility, resulting in a significant increase of 27% (χ21=12.65, P<.001) in view rates and 27.4% (χ21=30.01, P<.001) in satisfaction rates; (2) reducing the number of screens to increase efficiency and operability, leading to a significant decrease of 8.6% in cases where users did not perform any activity on the system after logging in (χ21=6.18, P=.02); (3) conducting a patient-focused campaign in 2 departments aimed at addressing memorability, leading to significant increases in 8 of the 12 usability indicators. CONCLUSIONS: Our results indicate that mHealth product-fit decisions originating from theory-based approaches and ongoing usability data analysis allow tailoring of the most appropriate responses for usability gaps, as reflected in increased use rates and satisfaction. In the case of the preanesthesia management system in the pediatric setting, increased usability conveyed important benefits for patients and families. This work suggests a framework and study methods that may also be applicable in other mHealth settings and domains.

Changes in High-Frequency Intracardiac Electrogram Indicate Cardiac Ischemia.

High-frequency QRS (HFQRS) analysis of surface ECG is a reliable marker of cardiac ischemia (CI). This study aimed to assess the response of HFQRS signals from standard intracardiac electrodes (iHFQRS) to CI in swine and compare them with conventional ST-segment deviations. Devices with three intracardiac leads were implanted in three swine in a controlled environment. CI was induced by inflating a balloon in epicardial coronary arteries. A designated signal-processing algorithm was applied to quantify the iHFQRS content before, during, and after each occlusion. iHFQRS time responses were compared to conventional ST-segment deviations. Thirty-three over thirty-nine (85%) of the occlusions presented significant reduction in the iHFQRS signal, preceding ST-segment change, being the only indicator of CI in brief occlusions. iHFQRS was found to be an early indicator for the onset of CI and demonstrated superior sensitivity to conventional ST-segment deviations during brief ischemic episodes.

High-frequency QRS electrocardiogram analysis during exercise stress testing for detecting ischemia.

INTRODUCTION: ECG stress testing is an inexpensive and non-invasive detector of myocardial ischemia; addition of high-frequency QRS analysis (HFQRS) may improve accuracy. This study compared HFQRS during exercise in patients with and without ischemia as defined by multiple criteria. MATERIAL AND METHODS: High-resolution ECGs were recorded for 139 patients undergoing T99-sestamibi/T201-thallium stress testing. Twenty-three were positive by at least two and 37 were negative for ischemia by all three of the following criteria: nuclear scan, ST-segment analysis and typical angina. Sixty-four not meeting criteria for positive or negative, six with adenosine test and nine patients with ECG recording artifacts were excluded. Mean age of the study group was 62+/-10 years, 83% were male. Ischemic patients had a higher incidence of previous myocardial infarction and coronary intervention than non-ischemic patients (74% vs. 46%; P=0.03 and 70% vs. 43%; P=0.05, respectively), but had a lower body mass index (28.7+/-5 vs. 33.0+/-8; P=0.015). HFQRS analysis consisting of signal averaging (150-250 Hz) and calculation of root mean squared values for each lead at different time points was performed and was similar between the groups. The relative change in HFQRS (RCQ) was calculated for each lead: {(maxHFQRS-minHFQRS)/maxHFQRS}. For each patient an RCQ index was calculated by averaging the two leads with the greatest RCQ value. The RCQ index was greater in ischemic vs. non-ischemic patients (45% vs. 34%; P=0.0069). CONCLUSION: Maximum decrease in HFQRS, as quantified by RCQ index, was greater in ischemic vs. non-ischemic patients. Use of the RCQ index may improve the diagnosis of ischemia during exercise stress testing.