AI Slide and Block Quality Verification for Tumor Sequencing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current genomic sequencing workflows for tumors are inefficient due to the manual and time-consuming process of selecting the best slides or sections for sequencing and verifying the adequacy of tumor tissue in samples.
Innovation Solution
A computer-implemented method using machine learning models to verify slide and block quality by assessing tumor content, purity, and necrosis, providing a quality score for selecting the most suitable FFPE blocks and slides for molecular testing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pathologists manually review materials to select the best slides or sections for sequencing, then they can identify suitable tissue blocks, but the process becomes time-consuming and challenging
Solution Approach 1:
The patent replaces the manual mechanical review process with an automated machine learning system. The ML model automatically analyzes digital images of tissue blocks and slides, evaluating tumor content, purity, and necrosis without human intervention, thereby eliminating time loss while maintaining or improving selection accuracy.
Solution Approach 2:
The system enables self-service by allowing the ML model to autonomously assess and rank tissue blocks based on predefined quality criteria. The model independently identifies the best blocks for sequencing without requiring pathologist review, making the system self-sufficient in the selection task.
2Reliability
If pathologists manually verify tumor tissue adequacy in chosen blocks, then they can ensure sufficient tumor content, but the process is challenging and prone to errors
Solution Approach 1:
The patent replaces manual verification with automated machine learning analysis. The ML model consistently evaluates tumor content, purity, and necrosis using standardized algorithms, eliminating human error and variability while simplifying the verification process into an automated computational task.
Solution Approach 2:
The system provides quantitative feedback through quality scores and tumor content measurements for each evaluated block. This objective feedback mechanism allows for reliable verification by comparing measured values against predefined thresholds, reducing complexity through automated decision rules.
3Productivity
If traditional workflow processes are used for genomic sequencing, then standard procedures are followed, but inefficiencies and time delays occur
Solution Approach 1:
The patent performs preliminary assessment of all tissue blocks before the sequencing decision is made. The ML model pre-evaluates tumor content, purity, and necrosis in advance, allowing the workflow to proceed directly to sequencing with the best blocks already identified, eliminating time delays associated with manual review during the sequencing process.
Solution Approach 2:
The automated ML-based selection system replaces manual pathologist review, dramatically increasing productivity by processing and evaluating blocks much faster than human reviewers can, thereby reducing overall workflow time while maintaining quality standards.
Data Source
Figure 1A~1B
Figure 1C
Figure 2
AI summary
Systems and methods are disclosed for verifying slide and block quality for testing. The method may comprise receiving a collection of one or more digital images at a digital storage device. The collection may be associated with a tissue block and corresponding to an instance. The method may comprise applying a machine learning model to the collection to identify a presence or an absence of an attribute, determining an amount or a percentage of tissue with the attribute from a digital image in the collection that indicates the presence of the attribute, and outputting a quality score corresponding to the determined amount or percentage.