Echocardiogram(Echo) is one of the common modality that captures state of the heart in the form images and videos. Using Ultrasound technique, an echo study captures multiple cross sections of the heart, termed as Views. Cardiologist utilizes measurements on the basis of these Views to analyse heart functions. In order to automate this process to measure how heart is functioning, an initial step is to identify Echo View types, akin to image classification.
In this talk, we will present development of Auto View Classification. We will emphasize on dataset development for Echo Views, challenges of Inter-Observer-Agreement, choice of deep learning models and showcase early results comparable with state-of-the-art. Finally, we will provide general guidelines while building AI models for healthcare in general.
- Layman’s intro to Heart and Echo
- What is a View and why it is important for Echo diagnosis.
- Why automatically identification of Echo View is needed?
- Overview of Auto View Classification with brief on Dataset development, inter-obeserver agreement, choice of Deep learning and SOTA results
- Learnings from building vision models in cardiology.
{{ gettext('Login to leave a comment') }}
{{ gettext('Post a comment…') }}{{ errorMsg }}
{{ gettext('No comments posted yet') }}