Medical Image Processing Apparatus for Barrett's Esophagus Diagnosis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current medical image processing technologies face challenges in accurately diagnosing Barrett's esophagus due to limitations in visualizing the geometric distortions and depth variations in endoscopic images, making it difficult to distinguish between columnar and squamous epitheliums.

Innovation Solution

A medical image processing apparatus and method that geometrically converts endoscopic images into a developed view, allowing for easier visualization of the squamocolumnar junction by projecting images onto a cylindrical surface and interpolating pixel positions, enabling more accurate diagnosis of Barrett's esophagus.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If endoscopic images are captured using conventional methods, then the imaging process is simple, but the geometric distortion and depth variation make it difficult to distinguish between columnar and squamous epitheliums

Engineering Contradiction:
Improvediagnostic accuracyVSAvoidimage processing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies dimensionality change by projecting the three-dimensional tubular structure of the esophagus onto a two-dimensional developed view. This transformation unfolds the cylindrical surface along its longitudinal axis, converting depth variations into lateral separations on the display. The coordinate transformation system maps three-dimensional spatial coordinates (x, y, z) to two-dimensional display coordinates (X, Y), where the circumferential position is represented by angle θ and the longitudinal position by distance z from the reference point. This dimensional reduction eliminates geometric distortion while preserving diagnostic information, enabling clear distinction between columnar and squamous epitheliums.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Ease of operation

If the endoscopic image is displayed without geometric conversion, then the display is simple, but the squamocolumnar junction and depth variations are not clearly visualized

Engineering Contradiction:
Improveease of diagnosisVSAvoidprocessing complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The patent implements preliminary action by pre-calculating and storing the coordinate transformation parameters that define the relationship between the three-dimensional tubular coordinate system and the two-dimensional display coordinate system. The transformation equations, which involve trigonometric functions of the circumferential angle θ and linear relationships with the longitudinal distance z, are prepared in advance. This allows the developed view to be generated rapidly during actual diagnosis by simply applying the pre-established transformation relationships to the captured image data, rather than performing complex geometric calculations in real-time.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS7857752B2Medical image processing apparatus and medical image processing method
Publication Date: 2010.12.28 OLYMPUS CORPORATION(JP)
  • US7857752B2 patent drawing
  • US7857752B2 patent drawing
  • US7857752B2 patent drawing

AI summary

A medical image processing apparatus having an image conversion section including a coordinate-system converting section for converting an image in an orthogonal coordinate system to an image in a polar coordinate system, for geometrically converting a medical image of a tubular part in vivo picked up and obtained by the coordinate-system converting section; a squamocolumnar junction detecting section for detecting a squamocolumnar junction that is a junction between a squamous epithelium and a columnar epithelium on the image of the polar coordinate system; and an analyzing section for comparing an evaluation value for a feature value determined for the detected squamocolumnar junction and a predetermined reference value and calculating an analysis result.