arXiv:2606. 08538v1 Announce Type: new Abstract: Routine laboratory panels drawn during cancer treatment constitute longitudinal physiological recordings of organ function, yet their temporal structure is discarded by single-timepoint prognostic tools.
arXiv:2608.24688v1 Announce Type: new Abstract: Precision oncology necessitates a longitudinal model of patient state that captures cancer evolution and treatment over time, integrating multimodal ob...
arXiv:2610.33054v1 Announce Type: new Abstract: Simulating an individual's disease course requires connecting patient history to future events, clinical timing and disease morphology. Longitudinal pr...
arXiv:2606. 11144v1 Announce Type: new Abstract: Resistance to first-line osimertinib in EGFR-mutant non-small-cell lung cancer (NSCLC) is the canonical example of predictable clonal evolution under therapeutic pressure, yet no public benchmark exists for training or evaluating computational models on the corresponding longitudinal patient trajectories.
GraphBEAT is a temporal graph representation learning framework that bucketizes clinical events from longitudinal electronic health records into time‑ordered graphs, enabling early risk stratification for gastrointestinal cancers. It uses a time‑decay message‑passing mechanism and multi‑head attention pooling to generate patient‑level representations, and was trained on multi‑site diagnosis codes and laboratory tests to predict pancreatic, liver, and intestinal cancer at 12, 24, and 36‑month lead times. Across all experiments, GraphBEAT outperformed baseline graph neural networks and sequence models while requiring far fewer floating‑point operations and parameters.
arXiv:2610.19178v1 Announce Type: new Abstract: Before deploying clinical prediction models, their reliability must be assessed to ensure safe use. Errors may arise from distribution shifts between t...
arXiv:2607. 07725v1 Announce Type: cross Abstract: Genomic prediction models often fail to transfer across institutions because sequencing panels differ across sites, creating structural feature missingness at deployment.
arXiv:2609.26578v1 Announce Type: new Abstract: Accurate preoperative subtype classification of renal cell carcinoma (RCC) from contrast-enhanced CT remains clinically challenging because clear cell...
arXiv:2607. 09521v1 Announce Type: new Abstract: Does every cancer patient truly need a complete diagnostic workup for accurate survival prediction?
The paper presents a newly curated, multi-center, multi-modal, and longitudinal lung cancer dataset comprising 1,365 patients with whole-slide images, CT scans, PET scans, structured clinical data, transcriptomics, and follow-up information. The dataset features substantial, non-uniform missingness across modalities, making it ideal for evaluating robust multi-modal fusion strategies. Benchmarks on 12‑month overall survival, disease‑specific survival, and longitudinal hazard prediction demonstrate that integrating complementary modalities consistently outperforms uni-modal approaches, even under severe missing data.
OpenMTB‑Audit is an open‑source benchmark that tests large language models on 500 synthetic non‑small cell lung cancer cases, covering five adversarial error categories and four safety labels: Supported, Partially Supported, Unsupported, and Insufficient Information. The study found that all eight tested LLMs over‑refused Partially Supported recommendations, collapsing labels to achieve high safety scores. A deterministic seven‑module framework, MTB‑AuditAgent, was introduced to reduce over‑refusal to 6.7% and reach 91.2% accuracy, while an oncologist annotation study highlighted disagreement around the boundary between information sufficiency and treatment optimization.
The study introduces NPC‑RubricBench, a phase‑aware benchmark that evaluates large language models (LLMs) on multistage clinical reasoning throughout the nasopharyngeal carcinoma care continuum. Using longitudinal data from 405 patients, the benchmark contains 941 patient‑phase units and 17,688 patient‑specific criteria across five dimensions, focusing strictly on decision‑time evidence. Evaluation of 30 general‑purpose and medical LLMs shows that strong diagnostic accuracy does not translate into better downstream care, with failure modes shifting from explicit decision errors in staging to decision omissions in initial work‑up and follow‑up, and reveals a trade‑off between prompt granularity, decision quality, evidence faithfulness, and safety adherence.
arXiv:2605. 25050v2 Announce Type: replace-cross Abstract: Integrating multimodal datasets in clinical oncology is frequently hindered by high dimensionality and blockwise missingness, where entire data sources are unavailable for specific patient subsets.