Multi-Modal Machine Learning Framework for Predicting Early Recurrence of Brain Tumors Using MRI and Clinical Biomarkers

2025/09/01 by Cheng, Cheng, Chen, Zeping, Xie, Rui +2
#FOS: Computer and information sciences #Machine Learning (cs.LG)

paper · doi:10.48550/arxiv.2509.01161

Abstract

Accurately predicting early recurrence in brain tumor patients following surgical resection remains a clinical challenge. This study proposes a multi-modal machine learning framework that integrates structural MRI features with clinical biomarkers to improve postoperative recurrence prediction. We employ four machine learning algorithms -- Gradient Boosting Machine (GBM), Random Survival Forest (RSF), CoxBoost, and XGBoost -- and validate model performance using concordance index (C-index), time-dependent AUC, calibration curves, and decision curve analysis. Our model demonstrates promising performance, offering a potential tool for risk stratification and personalized follow-up planning.

Citations

• Governance-Ready Small Language Models for Medical Imaging: Prompting, Abstention, and PACS Integration
• Reasoning in Computer Vision: Taxonomy, Models, Tasks, and Methodologies
• Lumina-mGPT 2.0: Stand-Alone AutoRegressive Image Modeling
• Resurrect Mask AutoRegressive Modeling for Efficient and Scalable Image Generation
• DUSE: A Data Expansion Framework for Low-resource Automatic Modulation Recognition based on Active Learning
• GLIMPSE: Do Large Vision-Language Models Truly Think With Videos or Just Glimpse at Them?
• From Physics to Foundation Models: A Review of AI-Driven Quantitative Remote Sensing Inversion
• OFFSET: Segmentation-based Focus Shift Revision for Composed Image Retrieval
• Towards AI Search Paradigm
• From LLM-anation to LLM-orchestrator: Coordinating Small Models for Data Labeling
• SoundMind: RL-Incentivized Logic Reasoning for Audio-Language Models
• SatelliteFormula: Multi-Modal Symbolic Regression from Remote Sensing Imagery for Physics Discovery
• Learning Sparsity for Effective and Efficient Music Performance Question Answering
• ReAgent-V: A Reward-Driven Multi-Agent Framework for Video Understanding
• FCOS: A Two-Stage Recoverable Model Pruning Framework for Automatic Modulation Recognition
• Image-to-Image Translation with Diffusion Transformers and CLIP-Based Image Conditioning
• SepPrune: Structured Pruning for Efficient Deep Speech Separation
• FreCT: Frequency-augmented Convolutional Transformer for Robust Time Series Anomaly Detection
• Anyprefer: An Agentic Framework for Preference Data Synthesis
• DC4CR: When Cloud Removal Meets Diffusion Control in Remote Sensing
• Twin Co-Adaptive Dialogue for Progressive Image Generation
• Dual-channel Heterophilic Message Passing for Graph Fraud Detection
• DConAD: A Differencing-based Contrastive Representation Learning Framework for Time Series Anomaly Detection
• SatelliteCalculator: A Multi-Task Vision Foundation Model for Quantitative Remote Sensing Inversion
• ForgetMe: Evaluating Selective Forgetting in Generative Models
• A Diffusion-Based Framework for Terrain-Aware Remote Sensing Image Reconstruction
• Rainy: Unlocking Satellite Calibration for Deep Learning in Precipitation
• SCORE: Story Coherence and Retrieval Enhancement for AI Narratives
• FineCIR: Explicit Parsing of Fine-Grained Modification Semantics for Composed Image Retrieval
• Lumina-Image 2.0: A Unified and Efficient Image Generative Framework
• A-MESS: Anchor based Multimodal Embedding with Semantic Synchronization for Multimodal Intent Recognition
• Learning Musical Representations for Music Performance Question Answering
• Temporal Working Memory: Query-Guided Segment Refinement for Enhanced Multimodal Understanding
• CT-PatchTST: Channel-Time Patch Time-Series Transformer for Long-Term Renewable Energy Forecasting
• Enhancing Low-Cost Video Editing with Lightweight Adaptors and Temporal-Aware Inversion
• Enhancing Code LLMs with Reinforcement Learning in Code Generation: A Survey
• DUET: Dual Clustering Enhanced Multivariate Time Series Forecasting
• Reassessing Layer Pruning in LLMs: New Insights and Methods
• RedTest: Towards Measuring Redundancy in Deep Neural Networks Effectively
• SGLP: A Similarity Guided Fast Layer Partition Pruning for Compressing Large Deep Models
• Self-evolving Agents with reflective and memory-augmented abilities
• A Generic Layer Pruning Method for Signal Modulation Recognition Deep Learning Models
• Calibrated Self-Rewarding Vision Language Models
• Comprehensive Multimodal Deep Learning Survival Prediction Enabled by a Transformer Architecture: A Multicenter Study in Glioblastoma
• Human-centric Reward Optimization for Reinforcement Learning-based Automated Driving using Large Language Models
• CMAT: A Multi-Agent Collaboration Tuning Framework for Enhancing Small Language Models
• TFB: Towards Comprehensive and Fair Benchmarking of Time Series Forecasting Methods
• VMT-Adapter: Parameter-Efficient Transfer Learning for Multi-Task Dense Scene Understanding
• MmAP : Multi-modal Alignment Prompt for Cross-domain Multi-task Learning
• FT2TF: First-Person Statement Text-To-Talking Face Generation
• Can pre-trained models assist in dataset distillation?
• Analyzing and Mitigating Object Hallucination in Large Vision-Language Models
• Evaluation and Analysis of Hallucination in Large Vision-Language Models
• mPLUG-Owl: Modularization Empowers Large Language Models with Multimodality
• SR-init: An interpretable layer pruning method
• Understanding the Dynamics of DNNs Using Graph Modularity
• Graph-Based Similarity of Neural Network Representations
• The 2021 WHO Classification of Tumors of the Central Nervous System: a summary
• Dynamic-DeepHit: A Deep Learning Approach for Dynamic Survival Analysis With Competing Risks Based on Longitudinal Data
• Identifying the Best Machine Learning Algorithms for Brain Tumor Segmentation, Progression Assessment, and Overall Survival Prediction in the BRATS Challenge
• The Image Biomarker Standardization Initiative: Standardized Quantitative Radiomics for High-Throughput Image-based Phenotyping
• Parameter-Efficient Fine-Tuning for Pre-Trained Vision Models: A Survey and Benchmark

Related

• Radiomics and Clinical Features in Predictive Modelling of Brain Metastases Recurrence
• Uncovering Neuroimaging Biomarkers of Brain Tumor Surgery with AI-Driven Methods
• Brain Tumor Survival Prediction using Radiomics Features
• Brain Tumor Recurrence vs. Radiation Necrosis Classification and Patient Survivability Prediction
• BrainMetDetect: Predicting Primary Tumor from Brain Metastasis MRI Data Using Radiomic Features and Machine Learning Algorithms