Machine Learning Image Classification for Chest X-Ray Quality Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The high rejection rate of chest X-rays due to technical quality issues such as over- or under-exposure, wrong positioning, and limited field of view leads to resource wastage, patient delays, and potential medical errors, as these flaws are difficult to identify and correct in real-time during image acquisition.
Innovation Solution
A system utilizing machine learning algorithms to classify medical images as diagnostically acceptable or unacceptable at the point of acquisition, prompting technicians to adjust parameters and recapture images to ensure diagnostic quality, thereby reducing rejections and improving operational efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If machine learning classification is performed at the point of image acquisition, then the repeat rate of chest radiographs is reduced, but the device complexity increases
Solution Approach 1:
A machine learning-based classification system is introduced as an intermediary between the X-ray imaging device and the radiologist. The system automatically evaluates image quality and provides feedback to the technologist, reducing the need for repeat scans while maintaining operational simplicity at the point of care.
Solution Approach 2:
The system provides real-time feedback to the technologist by classifying images as diagnostic or non-diagnostic and highlighting specific quality issues. This feedback loop enables immediate correction of technical flaws, reducing repeat rates without requiring complex manual review processes.
2Reliability
If real-time classification is performed during image acquisition, then diagnostic quality is improved, but the speed of image processing decreases
Solution Approach 1:
The machine learning model performs preliminary classification of image quality during the acquisition process itself, before the image is sent for radiological interpretation. This preliminary action identifies technical flaws immediately, allowing for real-time correction while maintaining overall processing speed through automated evaluation.
Data Source
AI summary
Methods and systems for determining a diagnostically unacceptable medical image. One system includes at least one electronic processor configured to receive a new medical image captured via a medical imaging device. The at least one electronic processor is also configured to determine a classification of the new medical image using a model developed with machine learning using training information that includes a plurality of medical images and an associated classification for each medical image, each associated classification identifying whether the associated medical image is diagnostically unacceptable, wherein the classification of the new medical image indicates whether the new medical image is diagnostically unacceptable. The at least one electronic processor is also configured to, when the classification indicates that the new medical image is diagnostically unacceptable, prompt a user of the medical imaging device to adjust a parameter associated with the new medical image and recapture the new medical image.