PhysioNet Challenge 2016

简介 2016 年 PhysioNet/CinC 挑战赛旨在鼓励开发算法，对从各种临床或非临床（如家访）环境中收集的心音记录进行分类。目的是从单个胸前位置的单个短记录（10-60 秒）中确定是否应将记录的主题转诊以进行专家诊断。在心动周期中，心脏首先产生电活动，然后电活动引起心房和心室收缩。这反过来又迫使心脏腔室和身体周围的血液。心脏瓣膜的打开和关闭与血液的加速-减速有关，从而引起整个心脏结构的振动（心音和杂音）[1]。这些振动可以在胸壁上听到，听特定的心音可以指示心脏的健康状况。心音图 (PCG) 是心音记录的图形表示。图 1 说明了 PCG 记录的一小部分。

Introduction The 2016 PhysioNet/CinC Challenge aimed to encourage the development of algorithms for classifying heart sound recordings collected from various clinical or non-clinical settings (e.g., home visits). The goal was to determine whether the recorded subject should be referred for expert diagnosis based on a single short recording (10–60 seconds) obtained from a single precordial position. During the cardiac cycle, the heart first generates electrical activity, which subsequently triggers atrial and ventricular contractions. This in turn forces blood into the cardiac chambers and throughout the body. The opening and closing of heart valves are associated with blood acceleration-deceleration, which causes vibrations of the entire cardiac structure (heart sounds and murmurs) [1]. These vibrations can be heard on the chest wall, and auscultation of specific heart sounds can indicate the cardiac health status. A phonocardiogram (PCG) is a graphical representation of heart sound recordings. Figure 1 illustrates a small segment of a PCG recording.