Automated Key Image Selection for Medical Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Physicians face inefficiencies and inaccuracies in selecting and comparing key medical images across studies, which can lead to misdiagnosis due to the manual and time-consuming nature of identifying and arranging relevant images for clinical reporting.
Innovation Solution
A system and method utilizing an electronic processor to automatically determine and display key images from current and comparison studies based on rules, including patient demographics, modality, anatomy, and image characteristics, to aid in the analysis and reporting of medical images, potentially incorporating machine learning for rule generation and image selection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If physicians manually select and compare key images across multiple studies, then diagnostic accuracy can be maintained through careful review, but clinical efficiency deteriorates due to time-consuming manual processes
Solution Approach 1:
The system performs preliminary automated selection of key images and identification of comparison studies before the physician reviews the case. By pre-processing the image selection and organization tasks using rules-based automation, the system prepares the diagnostic workflow in advance, allowing physicians to focus on interpretation rather than image hunting, thus resolving the contradiction between efficiency and time consumption
Solution Approach 2:
The system enables self-service by automatically performing image selection, study identification, and montage creation without requiring physician intervention for these routine tasks. The automated rules engine independently identifies relevant images and comparison studies, freeing physicians from manual workload while maintaining diagnostic quality, thereby improving productivity without increasing time loss
2Reliability
If physicians manually identify relevant comparison studies, then diagnostic accuracy can be maintained through careful selection, but device complexity increases due to the need for manual search and comparison tools
Solution Approach 1:
The system introduces an intermediary automated rules engine that mediates between the large database of medical images and the physician's diagnostic needs. This intermediary automatically identifies relevant comparison studies and key images based on predefined rules, reducing the complexity of manual search while ensuring reliable identification of diagnostically relevant images, thus maintaining accuracy without requiring complex manual tools
Solution Approach 2:
The system replaces the mechanical manual process of searching and comparing images with an automated electronic rules-based system. By substituting the manual mechanical search process with automated algorithmic processing, the system reduces the complexity of the image management interface while maintaining diagnostic reliability through systematic application of selection criteria
3Ease of operation
If multiple images are displayed separately in a virtual stack, then image detail can be preserved for individual review, but ease of operation deteriorates due to difficulty in locating and comparing specific images
Solution Approach 1:
The system segments the display into multiple functional components: a montage view showing key images from current and comparison studies arranged for easy comparison, and a virtual stack providing detailed individual image review. This segmentation allows physicians to efficiently locate and compare images in the montage while preserving the ability to examine individual images in detail through the stack, thus improving ease of operation without losing comparison capability
Solution Approach 2:
The system adds a spatial dimension to image presentation by creating a montage layout that displays multiple key images from different studies in a two-dimensional grid arrangement. This dimensional transformation allows simultaneous visualization of multiple images for easy comparison, while the virtual stack provides an additional temporal dimension for sequential detailed review, resolving the contradiction between ease of locating images and preservation of comparison capability
Data Source
AI summary
A method and system is provided for automatically determining a key image for display to a user as part of analyzing an image study generated as part of a medical imaging procedure. The system includes a memory storing a plurality of image studies, a display device for displaying images and an electronic processor interacting with the memory and the display device. The electronic processor is configured to determine a first key image within a plurality of images included in a first image study and to automatically determine, by executing one or more rules associated with one or more of the first key image, a user, a type of the first image study, a modality generating the first image study, an anatomy, a location of the modality, and patient demographics, a second key image included in at least one second image study. The system displays the second key image with the first key image to aid a user.