Medical Image Processing Device for Personalized Disease Visualization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current medical image processing technologies cannot effectively present personalized images to patients, as they typically rely on images of other subjects, making it difficult for doctors to explain disease states, progression, and recovery to patients in a visually relatable manner.
Innovation Solution
A medical image processing device and program that acquires, modifies, and displays medical images of a subject, allowing doctors to designate and visualize lesions, enabling the generation and display of predicted disease images based on the subject's own medical images, including modifications such as lesion addition, enlargement, reduction, or deletion, using machine learning algorithms for accurate representation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If images of other subjects are used for explanation, then general disease information can be provided, but personalization and patient relatability are lost
Solution Approach 1:
The patent creates a copy of the patient's own medical image and modifies it to show predicted disease states. This copying approach maintains personalization while avoiding the complexity of generating entirely new images, as the base image already captures the patient's unique anatomical characteristics.
Solution Approach 2:
The system performs preliminary processing of the patient's medical image to create a modified version showing predicted disease progression. This preliminary action prepares the personalized visual aid in advance, reducing the complexity during the actual explanation process.
2Measurement precision
If predicted disease images are generated using machine learning, then accuracy of disease state representation is improved, but processing time and computational resources increase
Solution Approach 1:
The machine learning model is trained in advance on large datasets of medical images and disease outcomes. This preliminary training allows the model to quickly generate accurate predictions during actual clinical use, reducing processing time while maintaining high accuracy.
Solution Approach 2:
The system uses the patient's existing medical image as a base and applies learned transformations to predict disease states. This copying and transformation approach is computationally more efficient than generating images from scratch, reducing processing time while maintaining accuracy.
Data Source
AI summary
A controller of a medical image processing device acquires a medical image of a subject. The controller causes a display to display a pre-modification image in which at least the position or range of a lesion to be modified on the medical image is displayed. The controller receives an instruction to designate at least the position or range of the lesion to be modified in a state in which the pre-modification image is displayed on the display. When at least the position or range of the lesion is designated, the controller acquires a predicted disease image in which the lesion is modified according to the designated information on the basis of the medical image. The controller causes the display to display the predicted disease image and the pre-modification image simultaneously or in a switching manner.