Skip to content

Incentivizing Reasoning in Vision-Language Models for Interpretable View Classification in Echocardiography

Notifications You must be signed in to change notification settings

matindaghyani/gmm-for-view-classification

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

59 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Vision-Language Models for Interpretation and Classification of Echocardiographic Views

This project implements a framework for echocardiographic view classification that goes beyond simple labeling by providing clinically grounded reasoning for every decision.


The Problem: The "Black Box" of Medical AI

Standard deep learning models can classify cardiac views with high accuracy, but they lack the ability to explain why a specific view was chosen. In clinical settings, a diagnosis without a justification is of limited use. Clinicians need to know if a model identified specific landmarks—like the left atrium or the mitral valve—to trust the final output.

The Solution: Aligning VLMs with Clinical Guidelines

We leverage a Vision-Language Model (VLM) fine-tuned via Reinforcement Learning to act as a reasoning-driven assistant.

Key Features

Explainable AI (XAI): Generates step-by-step anatomical justifications for each predicted view (e.g., identifying chamber locations and orientations).

GRPO Optimization: Utilizes Group Relative Policy Optimization to enhance reasoning capabilities without the need for an explicit value function, ensuring stable and efficient training.

Clinically-Informed Rewards: A rule-based reward system evaluates the model's text against official clinical guidelines, rewarding the mention of correct anatomical landmarks and penalizing incorrect ones.

Structural Prompting: Incorporates "anatomical clues" into the model's input to guide its reasoning path without revealing the ground-truth label.

About

Incentivizing Reasoning in Vision-Language Models for Interpretable View Classification in Echocardiography

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published