CirCor DigiScope Dataset

CirCor DigiScope数据集是由葡萄牙Universidade Portucalense Infante D. Henrique的研究团队创建，专注于儿童心脏杂音检测与分类。该数据集包含5282个录音，涵盖了1568名患者的四个主要听诊位置的心音数据，总计手动标注了215780个心音。数据集详细记录了每个心脏杂音的时间、形状、音高、分级和质量，以及杂音最明显的听诊位置，为机器学习算法在心脏杂音检测和分析中的实际应用提供了丰富的资源。此数据集的应用领域主要在于支持基于听诊的医疗决策系统，以解决心脏疾病的早期诊断问题。

CirCor DigiScope Dataset was developed by a research team from Universidade Portucalense Infante D. Henrique in Portugal, with a focus on pediatric cardiac murmur detection and classification. This dataset includes 5,282 audio recordings of heart sounds collected from four primary auscultation positions of 1,568 patients, with a total of 215,780 manually annotated heart sound instances. It comprehensively documents the timing, shape, pitch, severity grading, and acoustic quality of each cardiac murmur, as well as the auscultation position where the murmur is most audible. This dataset offers a valuable resource for the practical deployment of machine learning algorithms in cardiac murmur detection and analysis. Its primary application scenarios involve supporting stethoscope-based medical decision-making systems to facilitate the early diagnosis of cardiovascular diseases.