Development of a pediatric cardiac computer aided auscultation decision support system

Deep Learning Algorithms to Detect Murmurs Associated With Structural Heart Disease

Background The success of cardiac auscultation varies widely among medical professionals, which can lead to missed treatments for structural heart disease. Applying machine learning to cardiac auscultation could address this problem, but despite recent interest, few algorithms have been brought to clinical practice. We evaluated a novel suite of Food and Drug Administration-cleared algorithms trained via deep learning on >15 000 heart sound recordings. Methods and Results We validated the algorithms on a data set of 2375 recordings from 615 unique subjects. This data set was collected in real clinical environments using commercially available digital stethoscopes, annotated by board-certified cardiologists, and paired with echocardiograms as the gold standard. To model the algorithm in clinical practice, we compared its performance against 10 clinicians on a subset of the validation database. Our algorithm reliably detected structural murmurs with a sensitivity of 85.6% and specificity of 84.4%. When limiting the analysis to clearly audible murmurs in adults, performance improved to a sensitivity of 97.9% and specificity of 90.6%. The algorithm also reported timing within the cardiac cycle, differentiating between systolic and diastolic murmurs. Despite optimizing acoustics for the clinicians, the algorithm substantially outperformed the clinicians (average clinician accuracy, 77.9%; algorithm accuracy, 84.7%.) Conclusions The algorithms accurately identified murmurs associated with structural heart disease. Our results illustrate a marked contrast between the consistency of the algorithm and the substantial interobserver variability of clinicians. Our results suggest that adopting machine learning algorithms into clinical practice could improve the detection of structural heart disease to facilitate patient care.

Automated identification of innocent Still's murmur using a convolutional neural network

Background

Still's murmur is the most prevalent innocent heart murmur of childhood. Auscultation is the primary clinical tool to identify this murmur as innocent. Whereas pediatric cardiologists routinely perform this task, primary care providers are less successful in distinguishing Still's murmur from the murmurs of true heart disease. This results in a large number of children with a Still's murmur being referred to pediatric cardiologists.

Objectives

To develop a computer algorithm that can aid primary care providers to identify the innocent Still's murmur at the point of care, to substantially decrease over-referral.

Methods

The study included Still's murmurs, pathological murmurs, other innocent murmurs, and normal (i.e., non-murmur) heart sounds of 1,473 pediatric patients recorded using a commercial electronic stethoscope. The recordings with accompanying clinical diagnoses provided by a pediatric cardiologist were used to train and test the convolutional neural network-based algorithm.

Results

A comparative analysis showed that the algorithm using only the murmur sounds recorded at the lower left sternal border achieved the highest accuracy. The developed algorithm identified Still's murmur with 90.0% sensitivity and 98.3% specificity for the default decision threshold. The area under the receiver operating characteristic curve was 0.943.

Conclusions

Still's murmur can be identified with high accuracy with the algorithm we developed. Using this approach, the algorithm could help to reduce the rate of unnecessary pediatric cardiologist referrals and use of echocardiography for a common benign finding.

Embedded platform based heart murmur classification using deep learning approach

Ubiquitous Perturbations in cardiac auscultation properties, cardiovascular diseases (CVDs) are widely recognized. In the auscultation procedure, the appearance of pathological cardiac murmurs is linked to heart disorders. A noble automated detection system using 1-D Convolutional Neural Network (CNN) for the detection of pathological heart murmurs is proposed in this study, which removes the difficult task of extracting and selecting features. It directly acts on the phonocardiogram (PCG) signals. The fundamental purpose of this research is to develop a classification model for consistent recognition of cardiac murmurs when the data-set is imbalanced. In view of this, the proposed study for the imbalanced data-set incorporates the Adaptive Synthetic (ADASYN) approach to generate synthetic data for the minority class. The outcome analysis illustrates the positive result in the identification of heart murmurs on both balanced and imbalanced data-sets. Therefore, the developed deep learning model will learn better from the minority class and classify heart murmurs accurately.

Nonauscultatory clinical criteria are sensitive for cardiac pathology in low-risk paediatric heart murmurs

Background

Healthy children with likely innocent heart murmurs are frequently referred to cardiologists for reassurance. Existing guidelines that advise against these referrals are not consistently followed partly because they involve subjective auscultatory judgements with which many care providers are uncomfortable. Here, we investigate whether clinical criteria with no subjective auscultatory component are sensitive for cardiac pathology.

Methods

A retrospective chart review was performed of all new patients seen in our paediatric cardiology clinic for assessment of a murmur from January 1, 2016 through June 30, 2018. Patients were characterized as "low-risk" if they met all of the following criteria: asymptomatic; normal physical examination other than the murmur; no risk factors for congenital heart disease; and age over 12 months. The primary outcomes were the sensitivity for ruling out pathology and the negative predictive value of the proposed criteria.

Results

Of 915 total patients, 214 met the low-risk criteria. The sensitivity of our criteria for ruling out pathology was 97.2% (95% confidence interval 94.1% to 99.0%) and the negative predictive value was also 97.2% (95% confidence interval 94.0% to 98.7%). Six of the 214 low-risk patients had pathology (2.8%; 95% confidence interval 1.3% to 6.0%), none of which has required intervention since diagnosis. Each of these six children had a murmur that sounded pathological to the auscultating cardiologist.

Conclusions

Basic clinical criteria that do not require auscultation are highly sensitive for ruling out significant cardiac pathology in children over 12 months of age.

Intelligent Diagnosis of Heart Murmurs in Children with Congenital Heart Disease

Heart auscultation is a convenient tool for early diagnosis of heart diseases and is being developed to be an intelligent tool used in online medicine. Currently, there are few studies on intelligent diagnosis of pediatric murmurs due to congenital heart disease (CHD). The purpose of the study was to develop a method of intelligent diagnosis of pediatric CHD murmurs. Phonocardiogram (PCG) signals of 86 children were recorded with 24 children having normal heart sounds and 62 children having CHD murmurs. A segmentation method based on the discrete wavelet transform combined with Hadamard product was implemented to locate the first and the second heart sounds from the PCG signal. Ten features specific to CHD murmurs were extracted as the input of classifier after segmentation. Eighty-six artificial neural network classifiers were composed into a classification system to identify CHD murmurs. The accuracy, sensitivity, and specificity of diagnosis for heart murmurs were 93%, 93.5%, and 91.7%, respectively. In conclusion, a method of intelligent diagnosis of pediatric CHD murmurs is developed successfully and can be used for online screening of CHD in children.

Artificial Intelligence-Assisted Auscultation of Heart Murmurs: Validation by Virtual Clinical Trial

Artificial intelligence (AI) has potential to improve the accuracy of screening for valvular and congenital heart disease by auscultation. However, despite recent advances in signal processing and classification algorithms focused on heart sounds, clinical acceptance of this technology has been limited, in part due to lack of objective performance data. We hypothesized that a heart murmur detection algorithm could be quantitatively and objectively evaluated by virtual clinical trial. All cases from the Johns Hopkins Cardiac Auscultatory Recording Database (CARD) with either a pathologic murmur, an innocent murmur or no murmur were selected. The test algorithm, developed independently of CARD, analyzed each recording using an automated batch processing protocol. 3180 heart sound recordings from 603 outpatient visits were selected from CARD. Algorithm estimation of heart rate was similar to gold standard. Sensitivity and specificity for detection of pathologic cases were 93% (CI 90-95%) and 81% (CI 75-85%), respectively, with accuracy 88% (CI 85-91%). Performance varied according to algorithm certainty measure, age of patient, heart rate, murmur intensity, location of recording on the chest and pathologic diagnosis. This is the first reported comprehensive and objective evaluation of an AI-based murmur detection algorithm to our knowledge. The test algorithm performed well in this virtual clinical trial. This strategy can be used to efficiently compare performance of other algorithms against the same dataset and improve understanding of the potential clinical usefulness of AI-assisted auscultation.

The Diagnostic Utility of Computer-Assisted Auscultation for the Early Detection of Cardiac Murmurs of Structural Origin in the Periodic Health Evaluation

BACKGROUND

Identification of the nature of cardiac murmurs during the periodic health evaluation (PHE) of athletes is challenging due to the difficulty in distinguishing between murmurs of physiological or structural origin. Previously, computer-assisted auscultation (CAA) has shown promise to support appropriate referrals in the nonathlete pediatric population.

HYPOTHESIS

CAA has the ability to accurately detect cardiac murmurs of structural origin during a PHE in collegiate athletes.

STUDY DESIGN

Cross-sectional, descriptive study.

LEVEL OF EVIDENCE

Level 3.

METHODS

A total of 131 collegiate athletes (104 men, 28 women; mean age, 20 ± 2 years) completed a sports physician (SP)-driven PHE consisting of a cardiac history questionnaire and a physical examination. An independent CAA assessment was performed by a technician who was blinded to the SP findings. Athletes with suspected structural murmurs or other clinical reasons for concern were referred to a cardiologist for confirmatory echocardiography (EC).

RESULTS

Twenty-five athletes were referred for further investigation (17 murmurs, 6 abnormal electrocardiographs, 1 displaced apex, and 1 possible case of Marfan syndrome). EC confirmed 3 structural and 22 physiological murmurs. The SP flagged 5 individuals with possible underlying structural pathology; 2 of these murmurs were confirmed as structural in nature. Fourteen murmurs were referred by CAA; 3 of these were confirmed as structural in origin by EC. One such murmur was not detected by the SP, however, and detected by CAA. The sensitivity of CAA was 100% compared with 66.7% shown by the SP, while specificity was 50% and 66.7%, respectively.

CONCLUSION

CAA shows potential to be a feasible adjunct for improving the identification of structural murmurs in the athlete population. Over-referral by CAA for EC requires further investigation and possible refinements to the current algorithm. Further studies are needed to determine the true sensitivity, specificity, and cost efficacy of the device among the athletic population.

CLINICAL RELEVANCE

CAA may be a useful cardiac screening adjunct during the PHE of athletes, particularly as it may guide appropriate referral of suspected structural murmurs for further investigation.

Automated Identification of Innocent Still's Murmur in Children

OBJECTIVE

Still's murmur is the most common innocent heart murmur in children. It is also the most commonly misdiagnosed murmur, resulting in a high number of unnecessary referrals to pediatric cardiologist. The purpose of this study was to develop a computer algorithm for automated identification of Still's murmur that may help reduce unnecessary referrals.

METHODS

We first developed an accurate segmentation algorithm to locate the first and the second heart sounds. Once these sounds were identified, we extracted signal features specific to Still's murmur. Subsequently, machine learning-based classifiers, artificial neural network and support vector machine, were used to identify Still's murmur.

RESULTS

We evaluated our classifiers using the jackknife method using 87 Still's murmurs and 170 non-Still's murmurs. Our algorithm identified Still's murmur accurately with 84-93% sensitivity and 91-99% specificity.

CONCLUSION

We have achieved accurate automated identification of Still's murmur while minimizing false positives. The performance of our algorithm is comparable to the rate of murmur identification by auscultation by pediatric cardiologists.

SIGNIFICANCE

To our knowledge, our solution is the first murmur classifier that focuses singularly on Still's murmur. Following further refinement and testing, the presented algorithm could reduce the number of children with Still's murmur referred unnecessarily to pediatric cardiologists.

Current trends and perspectives for automated screening of cardiac murmurs

Although in high income countries rheumatic heart disease is now rare, it remains a major burden in low and middle income countries. In these world areas, physicians and expert sonographers are rare, and screening campaigns are usually performed by nomadic caregivers who can only recognise patients in an advanced phase of heart failure with high economic and social costs. Therefore, great interest exists regarding the possibility of developing a simple, low-cost procedure for screening valvular heart disease. With the development of computer science, the cardiac sound signal can be analysed in an automatic way. More precisely, a panel of features characterising the acoustic signal are extracted and sent to a decision-making software able to provide the final diagnosis. Although no system is currently available in the market, the rapid evolution of these technologies recently led to the activation of clinical trials. The aim of this note is to review the state of advancement of this technology (trends in feature selection and automatic diagnostic strategies), data available regarding performance of the technology in the clinical setting and finally what obstacles still need to be overcome before automated systems can be clinically/commercially viable.

Use of health information technology to reduce diagnostic errors

BACKGROUND

Health information technology (HIT) systems have the potential to reduce delayed, missed or incorrect diagnoses. We describe and classify the current state of diagnostic HIT and identify future research directions.

METHODS

A multi-pronged literature search was conducted using PubMed, Web of Science, backwards and forwards reference searches and contributions from domain experts. We included HIT systems evaluated in clinical and experimental settings as well as previous reviews, and excluded radiology computer-aided diagnosis, monitor alerts and alarms, and studies focused on disease staging and prognosis. Articles were organised within a conceptual framework of the diagnostic process and areas requiring further investigation were identified.

RESULTS

HIT approaches, tools and algorithms were identified and organised into 10 categories related to those assisting: (1) information gathering; (2) information organisation and display; (3) differential diagnosis generation; (4) weighing of diagnoses; (5) generation of diagnostic plan; (6) access to diagnostic reference information; (7) facilitating follow-up; (8) screening for early detection in asymptomatic patients; (9) collaborative diagnosis; and (10) facilitating diagnostic feedback to clinicians. We found many studies characterising potential interventions, but relatively few evaluating the interventions in actual clinical settings and even fewer demonstrating clinical impact.

CONCLUSIONS

Diagnostic HIT research is still in its early stages with few demonstrations of measurable clinical impact. Future efforts need to focus on: (1) improving methods and criteria for measurement of the diagnostic process using electronic data; (2) better usability and interfaces in electronic health records; (3) more meaningful incorporation of evidence-based diagnostic protocols within clinical workflows; and (4) systematic feedback of diagnostic performance.

The promise of computer-assisted auscultation in screening for structural heart disease and clinical teaching

Abstract

Cardiac auscultation has been the central clinical tool for the diagnosis of valvular and other structural heart diseases for over a century. Physicians acquire competence in this technique through considerable training and experience. In Africa, however, we face a shortage of physicians and have the lowest health personnel-to-population ratio in the world. One of the proposed solutions for tackling this crisis is the adoption of health technologies and product innovations to support different cadres of health workers as part of task shifting.

Computer-assisted auscultation (CAA) uses a digital stethoscope combined with acoustic neural networking to provide a visual display of heart sounds and murmurs, and analyses the recordings to distinguish between innocent and pathological murmurs. In so doing, CAA may serve as an objective tool for the screening of structural heart disease and facilitate the teaching of cardiac auscultation. This article reviews potential clinical applications of CAA.