A practical guide for nephrologist peer reviewers: evaluating artificial intelligence and machine learning research in nephrology.
腎臟科醫師同儕審查實用指南:評估腎臟科人工智慧與機器學習研究
Ren Fail 2025-07-07
AI 正在翻轉腎臟科,協助提升腎臟病的診斷、預測和治療,但臨床應用還有資料品質、偏誤和透明度等挑戰。文章提出用 TRIPOD-AI 檢查表來審查 AI 研究,強調資料完整、模型驗證和倫理。建議多元資料和多中心驗證來減少偏誤。未來會結合多種資料、深度學習影像、穿戴裝置和更智慧的決策系統。
相關文章PubMedDOI推理
Development and validation of the predictive scoring system for 90-day mortality in critical ill patients with exertional heatstroke.
重症勞力型熱射病患者 90 天死亡率預測評分系統的開發與驗證
Ren Fail 2025-07-07
Integrating estimated glomerular filtration rate and kidney replacement therapy criteria within the definition of kidney failure.
將估算腎絲球過濾率與腎臟替代治療標準納入腎衰竭定義
Kidney Int 2025-07-06
SEN, SLANH, and SEEDO Consensus report on Obesity related kidney disease. Proposal for a new classification.
SEN、SLANH 與 SEEDO 關於肥胖相關腎臟疾病的共識報告:新分類的提案
Kidney Int 2025-07-05
Pulmonary-renal syndrome under Dostarlimab, a recent PD-1 inhibitor, in a patient with Endometrial Cancer.
Endometrial Cancer 患者於使用新型 PD-1 抑制劑 Dostarlimab 後出現肺腎症候群
J Nephrol 2025-07-05