Image Quality Assessment Framework - PLAX Foreshortening


Project Description


Foreshortening in PLAX images can obscure critical cardiac structures, leading to inaccurate assessments. This project aims to create a standardized, open-access dataset of PLAX images with varying degrees of foreshortening, annotated by experts.



model
Network Architecture

Project Goals

  • Develop a Reference Dataset: Compile a comprehensive dataset of PLAX images with different foreshortening levels, annotated by experienced echocardiographers.
  • Standardize foreshortening Assessment: Establish objective criteria for evaluating the degree of foreshortening in PLAX images.
  • Empower AI Development: Provide a high-quality dataset to train and validate AI algorithms for automatic foreshortening detection and correction.
  • Improve Image Acquisition: Develop training materials and guidelines to help sonographers optimize image acquisition and reduce foreshortening.
  • Enhance Diagnostic Accuracy: Improve the reliability of PLAX echocardiographic interpretations for better patient care.

Your Role

  • Review PLAX Echocardiographic Videos: You will be presented with a series of PLAX echocardiographic videos.
  • Assess Foreshortening: Use the provided guidelines and your expertise to assess the degree of foreshortening in each video.
  • Rank Videos: Rank the videos based on the severity of foreshortening, from least to most foreshortened.
  • Provide Feedback (Optional): Share your insights and suggestions for improving the annotation process or the assessment criteria.

How to Participate

  1. Create an Account: Register on the UnityImaging.net labelling platform.
  2. Access the Project: Join the "PLAX Foreshortening Assessment Project."
  3. Review Guidelines: Familiarize yourself with the detailed annotation instructions and criteria for assessing foreshortening.
  4. Start Annotating: Begin reviewing and ranking PLAX echocardiographic videos based on the degree of foreshortening.
  5. Provide Feedback (Optional): Share your insights to help refine the process and improve the dataset.

Ranking Criteria

  • Left Ventricle Visualization: Assess how well the entire left ventricle, from apex to base, is visible.
  • Overall Image Quality: Rate the overall clarity and diagnostic quality of the image, considering factors like contrast and resolution.

Who Should Participate?

  • Echocardiographers: Your expertise in image interpretation is crucial for accurate assessment.
  • Cardiologists: Your clinical knowledge can help understand the impact of foreshortening on diagnosis.
  • Sonographers: Your experience in image acquisition can provide valuable insights into optimizing image quality.