Medical Image Processor Bone Segmentation Accuracy

Solution Overview

Problem

Current medical imaging technologies face challenges in accurately performing multi-organ segmentation due to unknown organ identification in input images, leading to inappropriate input conditions.

Innovation Solution

A medical information processing device that includes a processor for bone segmentation and organ segmentation, using a dictionary-based approach to identify and extract organ regions from CT images by referencing bone information, allowing for robust segmentation across different sizes and physiques.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional multi-organ segmentation methods are used without bone segmentation, then the processing speed is faster, but the segmentation accuracy for patients of various sizes and physiques deteriorates

Engineering Contradiction:

Improvesegmentation accuracyVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs bone segmentation as a preliminary step before organ segmentation. The processor first segments bones in the medical image to obtain bone segmentation results, then uses these results as input for subsequent organ segmentation. This preliminary action establishes accurate anatomical reference points that improve the precision of subsequent organ segmentation across patients of various sizes and physiques.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces bone segmentation results as an intermediary between the raw medical image and the final organ segmentation output. The bone segmentation results serve as a mediator that provides anatomical context and spatial relationships, enabling more accurate organ segmentation. This intermediary step transforms the input image into a form that is more suitable for accurate organ segmentation.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If bone segmentation is performed before organ segmentation, then the segmentation accuracy for various patient sizes is improved, but the device complexity increases

Engineering Contradiction:

Improvesegmentation accuracyVSAvoidprocessing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the overall segmentation task into two distinct segments: bone segmentation and organ segmentation. By segmenting the processing into separate stages, each stage can be optimized independently. The bone segmentation module focuses on identifying skeletal structures, while the organ segmentation module leverages these results to identify organs, reducing the overall complexity compared to attempting to segment all structures simultaneously.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent performs bone segmentation as a preliminary step before organ segmentation. The processor first segments bones in the medical image to obtain bone segmentation results, then uses these results as input for subsequent organ segmentation. This preliminary action establishes accurate anatomical reference points that improve the precision of subsequent organ segmentation across patients of various sizes and physiques.

Inventive Principle:

Principle #10Preliminary action

3Adaptability or versatility

If organ segmentation is performed without bone segmentation reference, then the processing workflow is simpler, but the adaptability to different patient physiques deteriorates

Engineering Contradiction:

Improveadaptability to patient physiquesVSAvoidworkflow complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent changes the input parameters for organ segmentation by using bone segmentation results instead of raw medical images. This parameter transformation adapts the segmentation process to different patient physiques because bone structures provide consistent anatomical landmarks across various body types. The bone segmentation results normalize the input data, enabling the organ segmentation model to generalize better across different patient populations.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent introduces bone segmentation results as an intermediary between the raw medical image and the final organ segmentation output. The bone segmentation results serve as a mediator that provides anatomical context and spatial relationships, enabling more accurate organ segmentation. This intermediary step transforms the input image into a form that is more suitable for accurate organ segmentation.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS20240265527A1Medical information processing device, medical information processing method, and non-transitory computer readable medium

Publication Date: 2024.08.08 CANON KK

AI summary

A medical information processing device includes a processor. The processor is configured to execute bone segmentation of an image and to acquire first data, execute, based on the first data, organ segmentation of the image, and acquire second data.