Task 1: PLAX Points

Aim

This dataset (and associated paper) is for training a neural network to accurately measure in the PLAX view the:

• Interventricular septum thickness
• Posterior wall thickness
• Left ventricular internal dimension (diastole and systole)

To do this we obtained expert labels for 4 points:

• Top of the interventricular septum
• Bottom of the interventricular septum
• Top of the posterior wall
• Bottom of the posterior wall

A neural network was trained to localise these 4 points from which the 3 measurements could be made.

Paper

Under review

Dataset for model development

This is a snapshot of the data and code used for this paper. You should use the "latest release" if you are training your own neural network. These snapshots are provided for reproducibility.

The dataset for model development is divided into train and tune (referred to as progress-monitoring in the paper) sets. There are 2056 images in this dataset. 1894 have all 4 points marked. In total there are 5724 labelled images in this dataset (as we train for all tasks simultaneously).

Dataset for model validation

The dataset for model validation, which comprises of 100 echocardiograms is kept private for competition use.

Models

The model used for the paper was training run 147, epoch 300.

Code

A snapshot of the exact code used for the paper is provided for reproducibility. The latest version of the code is available on GitHub, where some additional code has been provided to make it easier to do inference on your own data.

Download

• Unity Imaging Echocardiography Model Development Dataset Images: Download
• Unity Imaging Echocardiography Model Development Dataset Labels: Download
• Unity Imaging Echocardiography Model [Version 147, Epoch 300] Download
• Unity Imaging Echocardiography Code: Download

Funding, support, and approvals

We are grateful to the following institutions for funding and support

• National Institute of Health Research (NIHR), who support the Clincial Lecturerships of Dr. Matthew Shun-Shin and Dr. James Howard at Imperial College.
• NIHR Imperial Biomedial Research Centre (NIHR Imperial BRC) for providing start-up funding to collect pilot data to enable project / programme grant applications.

This research and open-access release of the has been conducted under:

• The Imperial AI Echocardiography Dataset [IRAS: 279328, REC:20/SC/0386]