Deep Learning Multi-Modal Medical Image Mutation Detection

Solution Overview

Problem

Current gene mutation detection methods are invasive, costly, and time-consuming, requiring tissue samples and sequencing, which is challenging especially for inaccessible target parts, leading to long detection times and high costs.

Innovation Solution

A computer-implemented method that uses image segmentation and fusion of multi-modal medical images to detect multi-mutations non-invasively, combining image segmentation and mutation detection to provide comprehensive and accurate results without the need for tissue samples, leveraging deep learning models and multi-modal medical imaging techniques.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If invasive tissue sampling and sequencing methods are used for gene mutation detection, then detection accuracy can be achieved, but detection time increases and cost increases

Engineering Contradiction:

Improvemutation detection accuracyVSAvoiddetection time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent creates a virtual copy of the tissue sample through 3D reconstruction from medical images. The system segments the medical image to extract tissue characteristics and reconstructs a virtual three-dimensional tissue model, which can be analyzed for mutation detection without physically extracting tissue. This virtual copy preserves the essential information needed for detection while eliminating the time-consuming invasive sampling process.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent replaces the mechanical invasive tissue sampling process with non-invasive medical image-based segmentation and 3D reconstruction. Instead of physically extracting tissue samples through biopsy procedures, the system uses image processing algorithms to segment and reconstruct tissue models from existing medical images, substituting mechanical intervention with computational methods.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If invasive tissue sampling and sequencing methods are used for gene mutation detection, then detection accuracy can be achieved, but cost increases

Engineering Contradiction:

Improvemutation detection accuracyVSAvoiddetection cost

Core Design Contradiction:

Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent creates a virtual copy of the tissue sample through 3D reconstruction from medical images. The system segments the medical image to extract tissue characteristics and reconstructs a virtual three-dimensional tissue model, which can be analyzed for mutation detection without physically extracting tissue. This virtual copy preserves the essential information needed for detection while eliminating the time-consuming invasive sampling process.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent replaces the mechanical invasive tissue sampling process with non-invasive medical image-based segmentation and 3D reconstruction. Instead of physically extracting tissue samples through biopsy procedures, the system uses image processing algorithms to segment and reconstruct tissue models from existing medical images, substituting mechanical intervention with computational methods.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Measurement precision

If invasive tissue sampling is performed for inaccessible target parts, then mutation detection is possible, but the procedure becomes more difficult and time-consuming

Engineering Contradiction:

Improvemutation detection capabilityVSAvoidsampling accessibility

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The patent creates a virtual copy of the tissue sample through 3D reconstruction from medical images. The system segments the medical image to extract tissue characteristics and reconstructs a virtual three-dimensional tissue model, which can be analyzed for mutation detection without physically extracting tissue. This virtual copy preserves the essential information needed for detection while eliminating the time-consuming invasive sampling process.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent replaces the mechanical invasive tissue sampling process with non-invasive medical image-based segmentation and 3D reconstruction. Instead of physically extracting tissue samples through biopsy procedures, the system uses image processing algorithms to segment and reconstruct tissue models from existing medical images, substituting mechanical intervention with computational methods.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS20240394871A1Computer-implemented method, method of training deep learning model, electronic device, and medium

Publication Date: 2024.11.28 BOE TECHNOLOGY GROUP CO LTD

AI summary

The present disclosure provides Aa computer-implemented method, a method of training a deep learning model, an electronic device, and a medium are provided. The method includes: obtaining a target image segmentation result according to a target medical image of a target part, wherein the target medical image includes a medical image in at least one modality; obtaining target fusion data according to the target medical image segmentation result and a medical image in a predetermined modality in the target medical image; and obtaining a target multi-mutation detection result according to the target fusion data.