Automatic and standardized surgical reporting for central nervous system tumors

2025-08-12

AI Summary

Standardized, automated reporting tools for postoperative neuroimaging analysis of central nervous system (CNS) tumors are currently lacking. Method: We developed the first fully automated, RANO 2.0–compliant imaging report generation pipeline trained on multicenter data. It integrates Attention U-Net for simultaneous segmentation of contrast-enhancing residual tumor and resection cavity, and DenseNet for MR sequence classification and tumor-type identification of enhancing lesions. The pipeline was validated via five-fold cross-validation and multi-scale evaluation, and deployed on the open-source platform Raidionics. Results: Segmentation achieved Dice scores of 87% (tumor core) and 77% (resection cavity); sequence classification and tumor-type classification attained balanced accuracies of 99.5% and 80%, respectively—matching state-of-the-art BraTS benchmarks. This work represents the first systematic implementation of multi-task joint analysis and standardized reporting for postoperative CNS tumor assessment, advancing automation and standardization in clinical postoperative evaluation.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text

Abstract

Magnetic resonance (MR) imaging is essential for evaluating central nervous system (CNS) tumors, guiding surgical planning, treatment decisions, and assessing postoperative outcomes and complication risks. While recent work has advanced automated tumor segmentation and report generation, most efforts have focused on preoperative data, with limited attention to postoperative imaging analysis. This study introduces a comprehensive pipeline for standardized postsurtical reporting in CNS tumors. Using the Attention U-Net architecture, segmentation models were trained for the preoperative (non-enhancing) tumor core, postoperative contrast-enhancing residual tumor, and resection cavity. Additionally, MR sequence classification and tumor type identification for contrast-enhancing lesions were explored using the DenseNet architecture. The models were integrated into a reporting pipeline, following the RANO 2.0 guidelines. Training was conducted on multicentric datasets comprising 2000 to 7000 patients, using a 5-fold cross-validation. Evaluation included patient-, voxel-, and object-wise metrics, with benchmarking against the latest BraTS challenge results. The segmentation models achieved average voxel-wise Dice scores of 87%, 66%, 70%, and 77% for the tumor core, non-enhancing tumor core, contrast-enhancing residual tumor, and resection cavity, respectively. Classification models reached 99.5% balanced accuracy in MR sequence classification and 80% in tumor type classification. The pipeline presented in this study enables robust, automated segmentation, MR sequence classification, and standardized report generation aligned with RANO 2.0 guidelines, enhancing postoperative evaluation and clinical decision-making. The proposed models and methods were integrated into Raidionics, open-source software platform for CNS tumor analysis, now including a dedicated module for postsurgical analysis.

Problem

• Automated postoperative CNS tumor segmentation and reporting.
• Standardized analysis following RANO 2.0 guidelines.
• Integration of models into open-source clinical software.

Innovation

• Attention U-Net for tumor and cavity segmentation
• DenseNet for MR sequence and tumor classification
• RANO 2.0-aligned automated reporting pipeline