The Cancer Genome Atlas
PulseAugur coverage of The Cancer Genome Atlas — every cluster mentioning The Cancer Genome Atlas across labs, papers, and developer communities, ranked by signal.
9 day(s) with sentiment data
-
Deep learning predicts gene expression from tissue images
Researchers have developed and validated deep regression models capable of predicting gene expression directly from whole-slide images (WSIs) of cancer tissue. These models, utilizing attention-based multiple instance l…
-
Foundation model and imaging link cancer genomes to scans
Researchers have developed a novel method that combines a foundation model called Evo~2 with clinical imaging to identify associations between genes and cancer phenotypes. This approach analyzes somatic mutations across…
-
New benchmark PathReportEval standardizes pathology report generation evaluation
Researchers have introduced PathReportEval, a new benchmark and evaluation framework designed to standardize the assessment of pathology report generation from whole-slide images. This framework addresses the limitation…
-
New benchmarks and methods advance medical vision-language models
Researchers have developed new benchmarks and distillation techniques to improve the capabilities of vision-language models (VLMs) in the medical domain. PathAgentBench focuses on evaluating VLMs' ability to acquire and…
-
New AdaSurvMamba framework enhances cancer prognosis analysis
Researchers have developed AdaSurvMamba, a new framework designed to improve multimodal survival analysis for cancer prognosis. This framework addresses limitations in current methods by dynamically adjusting the intera…
-
AI framework identifies cancer gene regulators across networks
Researchers have developed RegNetAgents, a novel AI-powered multi-agent framework designed to identify potential regulatory drivers within cancer genomics. This system integrates data from both bulk tumor and single-cel…
-
New TTA framework balances multimodal data for improved cancer survival prediction
Researchers have developed a new framework called "Together Then Apart" (TTA) for multimodal survival analysis in cancer prognosis. This approach aims to improve predictions by first aligning representations across diff…
-
New Active Learning Framework Slashes Histopathology Annotation Costs
Researchers have developed SHAL (Slide-level Hybrid Active Learning), a novel framework designed to significantly reduce the annotation burden in deep learning models for histopathology image segmentation. This patient-…
-
Pathology-Aware Prototype Distillation Enhances WSI Classification
Researchers have introduced TVT-PAPD, a novel self-supervised learning framework designed to improve the classification of whole slide images (WSIs) in pathology. This framework integrates a Tiny Vision Transformer with…
-
Research paper reveals widespread data leakage in pathology AI benchmarks
A recent research paper published on arXiv has uncovered significant data leakage issues within multimodal benchmarks used for whole-slide image (WSI) analysis in computational pathology. The study found that patient-le…
-
New framework creates efficient pathology models for edge deployment
Researchers have developed a new pretraining framework called MuCoDi to create smaller, more efficient pathology foundation models (PFMs) suitable for edge deployment. This method distills knowledge from multiple large …
-
EvoXplain framework reveals inconsistent ML model explanations
Researchers have developed EvoXplain, a new framework designed to assess the consistency of explanations generated by machine learning models. The tool investigates whether different training runs and model selection pr…
-
New methods advance continual learning for pathology image analysis · 5 sources tracked
Researchers have developed two novel approaches for continual learning in computational pathology, focusing on survival analysis for Whole Slide Images (WSIs). The first, MergeSurv, utilizes a merging-based framework wh…
-
New AI Framework Predicts Therapeutic Response Using Gene Expression Data
Researchers have developed PREDIKTOR, a novel multi-view framework designed to predict patient-specific therapeutic response using gene expression data. This framework aligns a personalized gene regulatory network with …
-
Research paper analyzes image transformation effects on latent space embeddings
A new research paper explores how image transformations affect the latent space representations used in histopathology classification. The study found that while embeddings of transformed images are closer to original e…
-
AI system uses digital twins to optimize clinical treatment decisions
Researchers have developed an AI system that uses digital twin simulations and reinforcement learning to optimize clinical decision-making for treatment response. The system, trained on historical data, continuously lea…
-
LLM pathology performance boosted by input design optimization
A new research paper demonstrates that seemingly minor design choices significantly impact the performance of large language models (LLMs) in pathology image analysis. By systematically analyzing factors like patch size…
-
New method learns to sparsify image tokens for efficient AI reasoning
Researchers have developed a novel method for processing gigapixel whole slide images in vision language models by treating token reduction as a trainable sparsification problem. This approach, detailed in a new arXiv p…
-
New benchmark TRAPS evaluates AI for cancer therapy prediction
Researchers have developed TRAPS, a new benchmark for pathway-guided cancer therapy response modeling. The study evaluated three biologically informed deep learning architectures—BINN, GraphPath, and PATH—across five ca…
-
BBOmix benchmark launched for biological AI hyperparameter tuning
Researchers have introduced BBOmix, a new open-source tabular benchmark designed for hyperparameter optimization in unsupervised biological representation learning. This benchmark addresses the computational expense of …