Nuclei Counting Workflow With Reviewable Image Annotations

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for counting cancerous and non-cancerous nuclei in cellular images rely heavily on manual counting, which is prone to human error and lacks clarity in audit trails, leading to potential miscounting and unclear image reviews.

Innovation Solution

An automatic counting algorithm that labels nuclei with colored dots for cancerous and non-cancerous cells, accompanied by a semi-automatic user interface for review and manual correction, with metadata storage to track counting data and ensure accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual counting of nuclei is performed by professionals, then the counting process can be completed, but human error increases and time consumption increases

Engineering Contradiction:
Improvecounting accuracyVSAvoidtime consumption
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces the manual mechanical counting process with an automated image processing system that uses algorithms to detect, label, and count nuclei in cellular images. The system automatically identifies and marks cancerous and non-cancerous nuclei, eliminating the need for manual inspection while improving both accuracy and efficiency.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent creates a digital representation of the counting process by generating annotated images that copy and highlight the nuclei positions. The system produces marked-up images with colored dots indicating each nucleus location and type, creating a permanent digital record that can be reviewed and verified without requiring repeated manual counting.

Inventive Principle:
Principle #26Copying

2Measurement precision

If manual counting is performed to ensure accuracy, then counting precision may improve, but the audit trail becomes unclear and images become difficult to review

Engineering Contradiction:
Improvecounting accuracyVSAvoidaudit trail clarity
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The patent uses color-coded markers to differentiate between cancerous and non-cancerous nuclei. Each nucleus is marked with a colored dot that visually indicates its classification, making it easy to review and verify counts. The color-coding system provides a clear visual audit trail that preserves information about each counted nucleus.

Inventive Principle:
Principle #32Color changes

Solution Approach 2:

The patent introduces an intermediary digital annotation layer that sits between the original image and the final count. This layer includes metadata storage and structured data records that track each nucleus identification, creating a clear audit trail that maintains all counting information without obscuring the original image.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Reliability

If reviewers manually recount nuclei to verify accuracy, then counting accuracy may be confirmed, but the process is time-consuming and mistakes cannot be tracked

Engineering Contradiction:
Improvecount verificationVSAvoidreview time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent implements a feedback mechanism where the automated counting system provides initial results that can be reviewed and verified. The system allows reviewers to interact with the annotated images, confirm or correct classifications, and the process maintains a record of all review actions. This feedback loop ensures reliability while reducing the time required compared to complete manual recounting.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent performs preliminary automated counting and annotation before human review, preparing the data in advance. The system pre-processes the images, identifies nuclei, and creates initial classifications, so that reviewers only need to verify rather than perform complete counting. This preliminary action significantly reduces review time while maintaining reliability.

Inventive Principle:
Principle #10Preliminary action

4Measurement precision

If multiple users manually count and review images, then consensus can be reached for accurate statistics, but the process becomes complex and error-prone

Engineering Contradiction:
Improvestatistical accuracyVSAvoidprocess complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent merges multiple review processes into a single integrated system. Multiple users can review and annotate the same image within the system, with all actions recorded and consolidated. The system automatically manages the review process, tracks changes, and combines results to reach consensus, reducing the complexity that would arise from separate manual processes.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent creates a universal platform that handles multiple functions: automated counting, annotation, review, verification, and statistical compilation. The system serves as a multi-functional tool that supports the entire nuclei counting workflow, from initial detection to final statistical analysis, reducing the need for multiple separate processes and tools.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentUS12494037B2Process for providing accurate counting of nuclei in cellular images
Publication Date: 2025.12.09 KOLPEKWAR RISHIK
  • US12494037B2 patent drawing
  • US12494037B2 patent drawing
  • US12494037B2 patent drawing

AI summary

This invention accomplishes an improvement to nuclei counting through an automatic counting algorithm that labels the nuclei in the images for the reviewers similar to how an individual would analyze cancerous and non-cancerous nuclei. The invention then provides a manual review step to reanalyze the image through a semi-automatic user interface. In addition, the counting data is stored in image metadata and a condensed version of the modified images is created.