Early Alzheimer's Detection Through Voice Analysis: Harnessing Locally Deployable LLMs via <i>ADetectoLocum</i>, a privacy-preserving diagnostic system.
透過語音分析進行早期阿茲海默症偵測:利用可在地端部署的大型語言模型(LLMs)與隱私保護診斷系統 <i>ADetectoLocum</i>
AMIA Jt Summits Transl Sci Proc 2025-06-12
ADetectoLocum 是一款可在醫院本地端運作的開源語言模型,能透過病患語音早期偵測阿茲海默症風險。它不需傳輸資料到外部,符合 HIPAA 法規,且準確率高於過去模型,是醫療院所安全又實用的 AD 診斷工具。
相關文章PubMed推理
Leveraging GPT-4o for Automated Extraction of Neural Projections from Scientific Literature.
運用 GPT-4o 自動化萃取科學文獻中的神經投射資訊
AMIA Jt Summits Transl Sci Proc 2025-06-12
Knowledge-Driven Feature Selection and Engineering for Genotype Data with Large Language Models.
運用大型語言模型於基因型資料的知識導向特徵選擇與特徵工程
AMIA Jt Summits Transl Sci Proc 2025-06-12
Biomni: A General-Purpose Biomedical AI Agent.
Biomni:通用型生物醫學 AI 智能代理
bioRxiv 2025-06-12
生物醫學研究常因流程繁瑣又重複,進展緩慢。Stanford 團隊推出 AI 助理 Biomni,能自動執行各種生醫研究任務,整合論文中的工具和資料庫,結合大型語言模型與程式碼執行,動態規劃複雜任務。測試證明 Biomni 泛用性高,無需特別調整就能處理多種生醫問題,幫助科學家加速研究。可至 https://biomni.stanford.edu 體驗。
相關文章PubMedDOI推理
sciLaMA: A Single-Cell Representation Learning Framework to Leverage Prior Knowledge from Large Language Models.
sciLaMA:利用大型語言模型先驗知識的單細胞表現學習框架
bioRxiv 2025-06-12
scATD: a high-throughput and interpretable framework for single-cell cancer drug resistance prediction and biomarker identification.
scATD:用於單細胞癌症藥物抗性預測與生物標誌物鑑定的高通量且可解釋性分析框架
Brief Bioinform 2025-06-12
Large language models outperform humans in identifying neuromyths but show sycophantic behavior in applied contexts.
大型語言模型在辨識神經迷思方面優於人類,但在應用情境中表現出諂媚行為。
Trends Neurosci Educ 2025-06-11