Understanding cardiovascular diseases from ultrasound images

AI Quick Summary

Understanding cardiovascular diseases by segmenting the carotid artery lumen in ultrasound images Initial situation and problem definition:One of the leading causes of death worldwide is related to cardiovascular diseases. In order to better study heart diseases, it is relevant to search for cardiovascular biomarkers in the corresponding region. Aim of the work and expected results:Given an ultrasound image of a longitudinal carotid artery view, the goal of this project is to correctly segment the carotid artery lumen. A correct segmentation would allow to easily identify the plaques in the vessel and thus to diagnose the health of a given patient. Visually, this region of interest is represented by the middle black region delimited by artery walls.The following three contributions are expected from this work. First, the proposal of a new dataset dataset for carotid artery lumen segmentation from ultrasound images. It involves the manual annotation of ~2000 existing ultrasound images of the longitudinal carotid arteries to get more familiar with cardiovascular ultrasound images. A second contribution is to study, review, and to implement different state-of-the-art deep learning approaches for cardiovascular structure segmentation. Finally, a third contribution involves the proposal of a novel deep learning architecture for the segmentation of longitudinal carotid arteries.Desired methods and procedure:The first stage of the project involves reading existing literature/state-of-the-art and also running experiments from available github repositories. The goal is to understand the weaknesses of existing methods. The second stage will involve tackling the weaknesses of existing approaches by exploring different blocks/modules and also by using knowledge from the medical literature. Every week, there will be a meeting, where the student will provide insights of the achievements via presentations, jupyter notebooks, etc. Discussions about next steps and guidance will be provided by the mentor. The source code will be developed in python/pytorch and will be stored in github. The. code can be built from scratch or from existing repositories. The bachelor work will be written in parallel as the experiments are being conducted.Creativity, variants, innovation: This project opens possibilities for exploring novel segmentation techniques by proposing new architectures and/or segmentation modules. As a first step, the literature will be reviewed and existing limitations shall be overcomed. Besides that, it gives the chance to the student to learn more about the digital health domain.

Project Idea Metadata

• Project Idea Name: Understanding cardiovascular diseases from ultrasound images
• Date: 8/24/2022 12:22:03 PM
• Administrators:

Project Idea Description

Initial situation and problem definition:

One of the leading causes of death worldwide is related to cardiovascular diseases. In order to better study heart diseases, it is relevant to search for cardiovascular biomarkers in the corresponding region.

Aim of the work and expected results:

Given an ultrasound image of a longitudinal carotid artery view, the goal of this project is to correctly segment the carotid artery lumen. A correct segmentation would allow to easily identify the plaques in the vessel and thus to diagnose the health of a given patient. Visually, this region of interest is represented by the middle black region delimited by artery walls.

The following three contributions are expected from this work. First, the proposal of a new dataset dataset for carotid artery lumen segmentation from ultrasound images. It involves the manual annotation of ~2000 existing ultrasound images of the longitudinal carotid arteries to get more familiar with cardiovascular ultrasound images. A second contribution is to study, review, and to implement different state-of-the-art deep learning approaches for cardiovascular structure segmentation. Finally, a third contribution involves the proposal of a novel deep learning architecture for the segmentation of longitudinal carotid arteries.

Desired methods and procedure:

The first stage of the project involves reading existing literature/state-of-the-art and also running experiments from available github repositories. The goal is to understand the weaknesses of existing methods. The second stage will involve tackling the weaknesses of existing approaches by exploring different blocks/modules and also by using knowledge from the medical literature. Every week, there will be a meeting, where the student will provide insights of the achievements via presentations, jupyter notebooks, etc. Discussions about next steps and guidance will be provided by the mentor. The source code will be developed in python/pytorch and will be stored in github. The. code can be built from scratch or from existing repositories. The bachelor work will be written in parallel as the experiments are being conducted.

Creativity, variants, innovation:

This project opens possibilities for exploring novel segmentation techniques by proposing new architectures and/or segmentation modules. As a first step, the literature will be reviewed and existing limitations shall be overcomed. Besides that, it gives the chance to the student to learn more about the digital health domain.

Understanding cardiovascular diseases by segmenting the carotid artery lumen in ultrasound images