HMC-QU Heart Medical Image Dataset

The HMC-QU dataset is an important medical imaging dataset jointly created by Hamad Medical Corporation (HMC), University of Tampere and Qatar University.Left Ventricular Wall Motion Estimation by Active Polynomials for Acute Myocardial Infarction Detection". The dataset contains two-dimensional echocardiographic (echo) recordings of the apical four-chamber (A4C) and apical two-chamber (A2C) views acquired during 2018 and 2019. These recordings were acquired with equipment from different manufacturers (such as Phillips and GE Vivid ultrasound machines) with a temporal resolution of 25 frames/s and spatial resolutions ranging from 422×636 to 768×1024 pixels.

To access the dataset, please send an email to mkiranyaz@qu.edu.qa to request access.

Dataset Contents

The HMC-QU dataset is mainly used for myocardial infarction (MI) detection and left ventricular wall segmentation research. The dataset contains 162 A4C views and 160 A2C views of 2D echocardiographic records. These records are from more than 10,000 ultrasound examinations, including more than 800 cases of admission for acute ST-segment elevation myocardial infarction (MI). There are 93 MI patients (all first-time and acute MI) and 69 normal (non-MI) subjects in the dataset.

Data Annotation

For the myocardial infarction detection task, the dataset provides annotations for each myocardial segment, divided into two categories: MI and non-MI. MI indicates any signs of regional wall motion abnormalities, while subjects without regional wall motion abnormalities are classified as non-MI. In addition, the dataset also contains 109 A4C view echocardiogram records, each frame of each cardiac cycle of these records has the corresponding semantic segmentation annotation of the left ventricular wall, and the annotation format is a segmentation mask of 224×224 pixels.

Dataset characteristics

The HMC-QU dataset is the first dataset shared with the research community for detecting myocardial infarction on the left ventricular wall of the heart. It not only covers data collected by multiple devices, but also contains high-quality annotation information, making it an ideal resource for studying myocardial infarction detection and left ventricular wall segmentation. In addition, the use of this dataset has been approved by the local ethics committee of HMC Hospital.