SOTA-DL-EchoSegmentation

Evaluation of state-of-the-art (SOTA) deep learning (DL) models in the segmentation of left and right ventricles in parasternal short-axis echocardiograms (PSAX-echo)

• SIUE Biomedical Imaging Research Lab - BIRL ..:: https://siue-biomedicalimaginglab.com/ ::..

A deep learning based approach to segment medical echocardiography PSAX images into its 3 main heart structures:

• Left ventricle (LV)
• Left myocardium (LM)
• Right ventricle (RV)

This repo contains Sample testing data from the Journal paper:

Evaluation of state-of-the-art deep learning models in the segmentation of the left and right ventricles in parasternal short-axis echocardiograms

Purpose

Previous studies on echocardiogram segmentation are focused on the left ventricle in parasternal long-axis views. Deep-learning models were evaluated on the segmentation of the ventricles in parasternal short-axis echocardiograms (PSAX-echo). Segmentation of the ventricles in complementary echocardiogram views will allow the computation of important metrics with the potential to aid in diagnosing cardio-pulmonary diseases and other cardiomyopathies. Evaluating state-of-the-art models with small datasets can reveal if they improve performance on limited data.

Approach

PSAX-echo was performed on 33 volunteer women. An experienced cardiologist identified end-diastole and end-systole frames from 387 scans, and expert observers manually traced the contours of the cardiac structures. Traced frames were pre-processed and used to create labels to train two domain-specific (Unet-Resnet101 and Unet-ResNet50), and four general-domain [three segment anything (SAM) variants, and the Detectron2] deep-learning models. The performance of the models was evaluated using the Dice similarity coefficient (DSC), Hausdorff distance (HD), and difference in cross-sectional area (DCSA).

Cite as:

Julian R. Cuellar, Vu Dinh, Manjula Burri, Julie Roelandts, James Wendling, Jon D. Klingensmith, "Evaluation of state-of-the-art deep learning models in the segmentation of the left and right ventricles in parasternal short-axis echocardiograms," J. Med. Imag. 12(2) 024002 (26 March 2025) https://doi.org/10.1117/1.JMI.12.2.024002

Dependencies

This Python file was tested on:

• Python 3.7
• ImageIO 2.9.0
• imageio-ffmpeg 0.5.1
• Keras 2.3.1
• MatPlotLib 3.3.2
• NumPy 1.19.2
• SciKit-Image 0.17.2

How to run

1. Create a system environment using Anaconda 3.
2. Create a Fork from this repository to your GitHub.
3. Clone the repo with your IDE - we use the PyCharm tools in our Lab.

UnetResNet Model

4. Download a trained model weights from OneDrive/Training_Models to the folder model/ in the local directory.
5. Select the IQ file that you want to segment and create a video. IQ files are in OneDrive/UltrasoundData.
6. Run the Python script

TrainTestRESNET.py

5. Segmented images are saved in Data/SegmentOutput/RESNET/ folder.

Detectron2 Model

4. Download a trained model weights from OneDrive/Training_Models to the folder model/ in the local directory.
5. Select the IQ file that you want to segment and create a video. IQ files are in OneDrive/UltrasoundData.
6. Run the Python script

TrainTestDetectron2.py

5. Segmented images are saved in Data/SegmentOutput/Detec2Mdl/ folder.

medSAM Models

2. Download a trained model weights from OneDrive/Training_Models to the folder model/ in the local directory.
3. Select the IQ file that you want to segment and create a video. IQ files are in OneDrive/UltrasoundData.
4. Run the Python script

TrainTestBIRLmedSAM.py

5. Segmented images are saved in Data/SegmentOutput/BIRLmedSAM folder.

Sample results

Here are two examples of segmented videos from the RESNET model, evaluated on selected PSAX cine loops.

• red : left ventricle
• purple : left myocardium
• yellow : right ventricle

MF0308PRE_5 segmented	MF0519_10 segmented