UNITY EchoTDI — Automated Multibeat Tissue Doppler Echocardiography Analysis

Download Dataset References

---

Overview

Tissue Doppler Imaging (TDI) is an echocardiographic technique used to assess myocardial motion velocities.
Manual interpretation is time-consuming and introduces variability in clinical workflow.

The UNITY EchoTDI dataset provides Doppler strips annotated with S’, E’, and A’ peaks, enabling benchmarking of automated algorithms for beat detection and keypoint localisation.

---

Dataset Description

• Dataset size: 280 Doppler strips (5,327 beats) from 48 patients (mean age 64±11 years).
• Views: Septal and lateral annuli.
• Annotations:
  • 3 expert cardiologists provided independent annotations.
  • Consensus of experts used as ground truth.
• Resolution: Cropped beats resized to 192×192 pixels.
• Beats: Multi-beat Doppler strips of varying lengths included.
• Ethics: Ethical approval obtained; imaging data anonymised.

Metadata provided per Doppler strip includes:

• Filename (Strip ID)
• Patient ID (anonymised)
• Beat count
• Peak velocity coordinates (S’, E’, A’)
• Expert annotations (per operator)

---

Request Dataset Access

The UNITY EchoTDI dataset will soon be made available for research.

To request access, please complete the form below:

Submit

---

License

This dataset is licensed under a
Creative Commons BY-NC-ND 4.0 International License.

---

Project Team

• Elisabeth S. Lane
• Jevgeni Jevsikov
• Massoud Zolgharni

---

References

• TDI Beat and Keypoint Detection (Springer)
• Automated Multi-beat TDI Analysis using Deep Neural Networks (Springer)