Automatic quantification of breast cancer biomarkers from multiple 18F-FDG PET image segmentation

2025-02-06

AI Summary

This study addresses the challenge of achieving consistent lesion segmentation and metabolic biomarker quantification in breast cancer 18F-FDG PET imaging following neoadjuvant chemotherapy (NAC). We propose the first fully automated segmentation framework integrating nnUNet, few-shot transfer fine-tuning, and a novel cross-temporal active learning strategy. The method achieves robust primary tumor segmentation in baseline and follow-up PET scans (Dice scores: 0.89 and 0.78, respectively) and automatically quantifies key metabolic biomarkers—SUVmax, metabolic tumor volume (MTV), and total lesion glycolysis (TLG). Biomarker measurements show excellent correlation with manual annotations (r > 0.95) and reliably detect significant NAC-induced reductions in all three metrics (p < 0.001). Key contributions include: (i) the first cross-temporal active learning strategy to mitigate annotation scarcity in follow-up scans, and (ii) integration of image quality control and standardized quantitative pipelines to ensure clinical reproducibility. This system provides a high-accuracy, highly consistent tool for dynamic assessment of NAC response.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Neoadjuvant chemotherapy (NAC) has become a standard clinical practice for tumor downsizing in breast cancer with 18F-FDG Positron Emission Tomography (PET). Our work aims to leverage PET imaging for the segmentation of breast lesions. The focus is on developing an automated system that accurately segments primary tumor regions and extracts key biomarkers from these areas to provide insights into the evolution of breast cancer following the first course of NAC. 243 baseline 18F-FDG PET scans (PET_Bl) and 180 follow-up 18F-FDG PET scans (PET_Fu) were acquired before and after the first course of NAC, respectively. Firstly, a deep learning-based breast tumor segmentation method was developed. The optimal baseline model (model trained on baseline exams) was fine-tuned on 15 follow-up exams and adapted using active learning to segment tumor areas in PET_Fu. The pipeline computes biomarkers such as maximum standardized uptake value (SUVmax), metabolic tumor volume (MTV), and total lesion glycolysis (TLG) to evaluate tumor evolution between PET_Fu and PET_Bl. Quality control measures were employed to exclude aberrant outliers. The nnUNet deep learning model outperformed in tumor segmentation on PET_Bl, achieved a Dice similarity coefficient (DSC) of 0.89 and a Hausdorff distance (HD) of 3.52 mm. After fine-tuning, the model demonstrated a DSC of 0.78 and a HD of 4.95 mm on PET_Fu exams. Biomarkers analysis revealed very strong correlations whatever the biomarker between manually segmented and automatically predicted regions. The significant average decrease of SUVmax, MTV and TLG were 5.22, 11.79 cm3 and 19.23 cm3, respectively. The presented approach demonstrates an automated system for breast tumor segmentation from 18F-FDG PET. Thanks to the extracted biomarkers, our method enables the automatic assessment of cancer progression.

Problem

• Automated segmentation of breast cancer lesions
• Extraction of biomarkers from PET images
• Assessment of tumor evolution post-chemotherapy

Innovation

• Deep learning-based tumor segmentation
• Automated biomarker extraction system
• Active learning for model adaptation