Histologic Tumor Margin Analysis Using Whole-Slide Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for analyzing tumor margins, such as en face and Mohs Micrographic Surgery, face challenges in ensuring complete tissue sectioning without missing peripheral or deep margins, particularly as tissue size increases, leading to potential recurrence and increased surgical time under anesthesia.
Innovation Solution
A system and method utilizing machine learning algorithms to rapidly and accurately assess tumor margins by analyzing whole slide images, determining completeness and presence of tumors, and providing real-time mapping and reporting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If en face or Mohs Micrographic Surgery is used to analyze 100% of tissue margins, then local recurrence rate is reduced, but surgical time and complexity increase
Solution Approach 1:
The patent segments the tissue margin analysis into multiple discrete steps: (1) performing the surgical excision, (2) preparing serial cross-sections of the margin, (3) staining and mounting sections on slides, (4) imaging each slide, and (5) analyzing images sequentially. This segmentation allows the complex task of 100% margin analysis to be broken down into manageable, automated steps that reduce surgical time while maintaining completeness.
Solution Approach 2:
The patent applies preliminary action by preparing and analyzing tissue sections immediately during the surgical procedure rather than after. The system performs rapid imaging and analysis of serial sections in the operating room, providing real-time feedback on margin status before the patient leaves the surgical setting. This eliminates delays and allows for immediate resection if positive margins are detected.
2Loss of time
If standard breadloafing methodology is used for tissue analysis, then surgical time is reduced, but margin analysis completeness decreases to approximately 1%
Solution Approach 1:
The patent creates multiple copies of the tissue margin through serial sectioning. Instead of analyzing a single section, the system generates and analyzes multiple sequential cross-sections (typically 5-10 sections) of the excised margin. Each section is imaged and analyzed, providing redundant sampling that increases the likelihood of detecting tumor cells at the margin while maintaining rapid throughput suitable for intraoperative use.
3Reliability
If the number of tissue pieces generated and sectioned is increased to analyze more margin, then margin analysis percentage increases, but time consumption increases at both generation and reading steps
Solution Approach 1:
The patent replaces manual mechanical sectioning and reading processes with an automated imaging and analysis system. A microtome or cryostat generates serial sections, which are automatically stained, mounted on slides, and imaged using a automated slide scanner. The imaging system captures high-resolution images of each section, and computer algorithms automatically analyze the images for tumor presence. This automation eliminates time-consuming manual reading while maintaining the ability to analyze multiple sections.
Solution Approach 2:
The patent ensures continuity of useful action by implementing a streamlined workflow where serial sections are prepared and analyzed in continuous sequence without interruption. The automated system prepares sections, stains them, images them, and analyzes them in an unbroken workflow, eliminating idle time between steps. This continuous processing allows multiple sections to be analyzed rapidly, increasing margin coverage without proportionally increasing total time.
Data Source
AI summary
This invention provides a histologic system and method for rapidly and accurately assessing tumor margins for the presence or absence of tumor using machine learning algorithms. This affords a rapid and accurate histologic tumor readout and increase process efficiency and decreases the chance for human error. Advantageously and uniquely, the system and method allows for analyzing the tissue section as complete or incomplete as the first criteria to determine whether a tissue section is clear of tumor. A machine learning process receives whole slide images (WSI) of tissue and determines (a) if each image of the WSI contains complete/incomplete tissue samples and (b) if each image of the WSI contains tumorous tissue or an absence thereof. A reconstruction process generates a model of the tissue that maps types of tissue therein, and a display process provides results of the model or report for use and manipulation by a user.