The CirCor DigiScope Dataset: From Murmur Detection to Murmur Classification

Abstract

Cardiac auscultation is one of the most costeffective techniques used to detect and identify many heart conditions. Computer-assisted decision systems based on auscultation can support physicians in their decisions. Unfortunately, the application of such systems in clinical trials is still minimal since most of them only aim to detect the presence of extra or abnormal waves in the phonocardiogram signal, i.e., only a binary ground truth variable (normal vs abnormal) is provided. This is mainly due to the lack of large publicly available datasets, where a more detailed description of such abnormal waves (e.g., cardiac murmurs) exists.To pave the way to more effective research on healthcare recommendation systems based on auscultation, our team has prepared the currently largest pediatric heart sound dataset. A total of 5282 recordings have been collected from the four main auscultation locations of 1568 patients, in the process, 215780 heart sounds have been manually annotated. Furthermore, and for the first time, each cardiac murmur has been manually annotated by an expert annotator according to its timing, shape, pitch, grading, and quality. In addition, the auscultation locations where the murmur is present were identified as well as the auscultation location where the murmur is detected more intensively. Such detailed description for a relatively large number of heart sounds may pave the way for new machine learning algorithms with a real-world application for the detection and analysis of murmur waves for diagnostic purposes.

Heart murmur detection from phonocardiogram recordings: The George B. Moody PhysioNet Challenge 2022

“…The CirCor DigiScope dataset [7] was used for the 2022 George B. Moody PhysioNet Challenge. The dataset consists of one or more PCG recordings from different auscultation locations on each patient's body.…”

“…The study protocol was approved by the 5192-Complexo Hospitalar HUOC/PROCAPE Institutional Review Board, under the request of the Real Hospital Portugues de Beneficencia em Pernambuco. A detailed description of the dataset can be found in [7].…”

“…We designed ( 6) and ( 7) so that the mean cost of an expert screening was lowest when only 25% of the patient cohort received expert screenings but higher when screening below and above the capacity of the system. Figure 2 shows these costs across different patient cohort and screening sizes, and the Appendix provides a fuller derivation of ( 6) and (7).…”

Multiple Instance Learning Framework can Facilitate Explainability in Murmur Detection

“…Most participants are in the 1-11 year age group. For a more detailed description of the dataset, we refer to the original description in [25,26]. A segmentation into the phases of the heart cycle (S1, systole, S2, diastole) was created algorithmically and verified by experts.…”

How Do We Assess the Trustworthiness of AI? Introducing the Trustworthiness Assessment Model (TrAM)

“…For example, this could be to detect heart sound anomalies in medicine [108], picking the best applicant in personal selection [65], or detecting suspects in surveillance [56]. Their individual standards (hence their requirement list) for a trustworthy system might contain a bias-free stable performance [32,41,93], combined with specific requirements in the labeling process, a higher performance than the current state-of-the-art ADM system in a benchmark data set [88,103], robustness, and a specific methodological or algorithmic approach [4,52].…”