NSCLC Recurrence Prediction via Radiomic-Pathomic Feature Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a lack of a biomarker to accurately stratify and predict disease risk in early stage non-small cell lung cancer (NSCLC) patients, leading to sub-optimal decision-making for adjuvant chemotherapy and varying incidence of local recurrence, which complicates treatment planning.
Innovation Solution
A combined set of radiomic and pathomic features extracted from computed tomography (CT) imagery and digitized tissue slides is used to train a machine learning classifier, predicting recurrence in early stage NSCLC by integrating features at different scales, including nuclear shape, texture, and spatial arrangement, to improve predictive accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single type of biomarker (clinical or radiographic) is used for prediction, then the decision-making process is simple, but the predictive accuracy is insufficient
Solution Approach 1:
The patent combines multiple types of biomarkers (clinical factors, radiographic features, and pathologic characteristics) into an integrated prediction system. This merging of diverse data sources enables accurate risk stratification by capturing multiple aspects of tumor behavior that single-marker systems miss, directly resolving the contradiction between simplicity and accuracy.
Solution Approach 2:
The prediction system functions as a composite biomarker panel, integrating heterogeneous information from different modalities (imaging, pathology, clinical data) into a unified predictive framework. This composite approach leverages the complementary strengths of each biomarker type to achieve superior predictive performance compared to individual markers.
2Reliability
If adjuvant chemotherapy is offered to all early stage NSCLC patients, then treatment coverage is comprehensive, but unnecessary treatment and resource waste increase
Solution Approach 1:
The patent applies local quality by tailoring treatment recommendations to individual patient risk profiles rather than applying uniform treatment. High-risk patients receive aggressive adjuvant chemotherapy, while low-risk patients receive surveillance, optimizing treatment effectiveness for each subgroup and avoiding unnecessary treatment in low-risk populations.
Solution Approach 2:
The system changes the decision parameter from binary (treat all or treat none) to continuous risk stratification. By quantifying recurrence risk as a spectrum, the system enables nuanced treatment decisions that match therapy intensity to patient-specific risk levels, improving overall treatment effectiveness while reducing resource waste on low-risk patients.
3Measurement precision
If comprehensive biomarker analysis is performed, then predictive accuracy improves, but the time and computational resources required increase
Solution Approach 1:
The system performs preliminary action by extracting and analyzing all relevant features from pathology slides and imaging data during the initial diagnostic workup, before treatment decisions are made. This advance processing ensures that comprehensive biomarker analysis is completed as part of routine evaluation, minimizing additional time requirements and enabling immediate risk stratification.
Data Source
AI summary
Embodiments predict early stage NSCLC recurrence, and include processors configured to access a pathology image of a region of tissue demonstrating early stage NSCLC; extract a set of pathomic features from the pathology image; access a radiological image of the region of tissue; extract a set of radiomic features from the radiological image; generate a combined feature set that includes at least one member of the set of pathomic features, and at least one member of the set of radiomic features; compute a probability that the region of tissue will experience NSCLC recurrence based, at least in part, on the combined feature set; and classify the region of tissue as recurrent or non-recurrent based, at least in part, on the probability. Embodiments may display the classification, or generate a personalized treatment plan based on the classification.