OligoM-Cancer: A multidimensional information platform for deep phenotyping of heterogenous oligometastatic cancer.
OligoM-Cancer:一個多維度信息平台,用於深度表型分析異質性寡轉移癌。
Comput Struct Biotechnol J 2024-09-11
A prospective pragmatic evaluation of automatic trial matching tools in a molecular tumor board.
分子腫瘤委員會中自動試驗匹配工具的前瞻性實用評估。
NPJ Precis Oncol 2025-01-27
這項研究評估了四種臨床試驗匹配工具(Klineo、ScreenAct、Trialing 和 DigitalECMT)在癌症患者獲取治療創新方面的效果,研究地點為 Centre Leon Berard 的分子腫瘤委員會。分析157名患者後發現,平均每位患者獲得2.19個臨床試驗建議,但38%的患者沒有任何建議。這些工具的精確度和召回率都偏低,分別約為0.33和0.32。使用所有工具可增加26%的試驗選擇,且結合大型語言模型可提升性能5%。研究建議需有專家監督並推動匹配技術進步。
PubMedDOI
Learning and actioning general principles of cancer cell drug sensitivity.
癌細胞藥物敏感性的一般原則學習與實踐。
Nat Commun 2025-02-14
Cancer gene identification through integrating causal prompting large language model with omics data-driven causal inference.
透過整合因果提示的大型語言模型與基因組數據驅動的因果推斷進行癌症基因識別。
Brief Bioinform 2025-03-12
Bridging Data Gaps in Oncology: Large Language Models and Collaborative Filtering for Cancer Treatment Recommendations.
彌補腫瘤學資料缺口:運用大型語言模型與協同過濾於癌症治療建議
medRxiv 2025-04-29
A Knowledge-Enhanced Platform (MetaSepsisKnowHub) for Retrieval Augmented Generation-Based Sepsis Heterogeneity and Personalized Management: Development Study.
基於檢索增強生成的敗血症異質性與個人化管理之知識強化平台(MetaSepsisKnowHub):開發性研究
J Med Internet Res 2025-06-06