Closed-Loop Image Quality Feedback System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In medical imaging, there is a disconnect between image acquisition and interpretation, leading to low-quality images that hinder diagnosis and increase healthcare costs, as technologists acquire images without immediate feedback on their quality from interpreters, resulting in non-diagnostic results and the need for repeat imaging.
Innovation Solution
A closed-loop system using a medical workstation with a graphical user interface that allows interpreters to rate image quality, which is used to train a machine learning classifier to assess image quality in real-time during acquisition, alerting technologists to reacquire poor-quality images, thereby ensuring high-quality images are captured during the patient's visit.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If image acquisition and interpretation are performed by different specialists at different times, then workflow specialization is improved, but image quality and diagnostic accuracy deteriorate due to lack of immediate feedback
Solution Approach 1:
The system implements a feedback mechanism where image interpreters evaluate acquired images and provide quality assessments. This feedback is then used to train a machine learning classifier that provides real-time quality predictions to technologists during image acquisition, creating a closed-loop system that maintains quality without requiring direct interpreter involvement in every acquisition step.
Solution Approach 2:
The machine learning classifier is trained in advance using historical image data and interpreter evaluations. This pre-trained model then provides preliminary quality assessments during image acquisition, allowing technologists to make immediate adjustments before finalizing the image set, rather than waiting for interpreter feedback after acquisition is complete.
2Productivity
If technologists acquire images without immediate quality feedback, then acquisition speed is improved, but diagnostic value deteriorates due to non-diagnostic images
Solution Approach 1:
The system provides real-time feedback to technologists during image acquisition through the machine learning classifier's quality predictions. This allows technologists to maintain high acquisition speed while receiving immediate guidance on image quality, enabling them to adjust techniques on the spot rather than discovering quality issues after the fact.
Solution Approach 2:
The machine learning classifier empowers technologists to self-assess image quality during acquisition without requiring constant interpreter review. The system serves itself by using historical data to train the model, which then autonomously provides quality guidance to technologists, maintaining both speed and reliability.
3Reliability
If image quality is assessed and feedback is collected periodically, then quality control is improved, but responsiveness to prevent low-quality images deteriorates
Solution Approach 1:
The system transforms periodic feedback into continuous feedback by implementing real-time quality assessment during image acquisition. The machine learning classifier provides immediate quality predictions to technologists as images are acquired, enabling proactive quality control rather than reactive periodic review, thus maintaining both reliability and speed.
4Device complexity
If low-quality images are not identified immediately, then workflow simplicity is improved, but healthcare costs increase due to repeat imaging
Solution Approach 1:
The system introduces automated real-time feedback through the machine learning classifier, which identifies low-quality images immediately upon acquisition. This maintains workflow simplicity by automating the quality check process rather than requiring manual review of every image, while simultaneously preventing costly repeat imaging by alerting technologists to quality issues during the initial acquisition.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A medical imaging apparatus includes a radiology workstation (10) with a workstation display (14) and one or more workstation user input devices (16). A medical imaging device controller (26) includes a controller display (30) and one or more controller user input devices (32). The medical imaging device controller is connected to control a medical imaging device (40) to acquire medical images (44). One or more electronic processors (22, 38) are programmed to: operate the medical workstation to provide a graphical user interface (GUI) (24) that displays medical images stored in a radiology information system (RIS) (20), receives entry of medical examination reports, displays an image rating user dialog (70), and receives, via the image rating user dialog, image quality ratings for medical images displayed at the medical workstation; operate the medical imaging device controller to perform an imaging examination session including operating the medical imaging device controller to control the medical imaging device to acquire session medical images; while performing the imaging examination session, assign quality ratings to the session medical images based on image quality ratings received via the image quality rating user dialog displayed at the medical workstation; and while performing the imaging examination session, display quality ratings assigned to the session medical images.