Lesion Reading Display Interface for Lung CT Uncertainty Classification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Medical image reading, particularly for lung CT images, is challenging due to high difficulty in identifying nodules and the likelihood of ignoring fine abnormalities, leading to inaccurate readings even by experienced medical teams.

Innovation Solution

A computer-based method is developed to display lesion reading results, including a user interface that distinguishes between certain and uncertain lesions, providing a degree of uncertainty and using a diagnostic model with network functions to classify regions, thereby assisting medical professionals in accurate diagnosis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If medical professionals manually read lung CT images, then diagnostic experience and intuition can be applied, but reading accuracy decreases due to high difficulty and human error in identifying nodules and fine abnormalities

Engineering Contradiction:
Improvelesion reading accuracyVSAvoiddiagnosis reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent introduces a diagnostic model as an intermediary between the medical image data and the final diagnosis. This model processes the images and provides assistance information to medical professionals, thereby improving reading accuracy while maintaining the reliability of human judgment in the final decision-making process.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system provides feedback to medical professionals by displaying assistance information including uncertain lesions and their characteristics. This feedback loop allows doctors to review and verify the model's findings, improving overall diagnostic accuracy while maintaining professional oversight.

Inventive Principle:
Principle #23Feedback

2Ease of operation

If a simple display method is used for lesion information, then the interface is easy to understand, but important details such as uncertain lesions and their characteristics may be overlooked

Engineering Contradiction:
Improveinterface usabilityVSAvoidlesion detail information
Core Design Contradiction:
Ease of operationVSLoss of information

Solution Approach 1:

The patent segments the display information into different categories: certain lesions and uncertain lesions with their specific characteristics. This segmentation allows the interface to present complex information in an organized, easy-to-navigate manner while ensuring that no important details are lost or overlooked by the medical professional.

Inventive Principle:
Principle #1Segmentation

3Loss of information

If all lesion information is displayed simultaneously, then complete information is provided, but the interface becomes complex and difficult to interpret

Engineering Contradiction:
Improvecompleteness of lesion informationVSAvoiduser interface complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The display interface is designed to be dynamic, adapting its complexity based on the specific case. The system selectively displays certain and uncertain lesions with their characteristics, providing complete information when needed while maintaining simplicity for routine cases. This dynamic approach balances information completeness with interface usability.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS20210407637A1Method to display lesion readings result
Publication Date: 2021.12.30 VUNO INC
  • US20210407637A1 patent drawing
  • US20210407637A1 patent drawing
  • US20210407637A1 patent drawing

AI summary

Disclosed is a computer program stored in a computer readable storage medium. When the computer program is executed by one or more processors, the computer program provides a user interface for displaying a lesion readings result, and the user interface may include: lesion information comprised in medical data; and one or more findings related to the lesion information and displayed in response to a user interaction with the lesion information.