文献速递肾癌与肾肿瘤

一种全自动人工智能方法预测根治性或部分性肾切除术后肾功能

A Fully Automated Artificial Intelligence-Based Approach to Predict Renal Function After Radical or Partial Nephrectomy.
PubMed 文献 · Urology · 2025-02-04

该研究纳入KiTS19挑战赛中300例因肾肿瘤行肾切除术的患者,比较全自动人工智能模型与经验证的临床模型预测术后肾小球滤过率(GFR)的准确性。结果显示,AI模型与临床模型预测术后GFR的相关系数分别为0.75和0.77,预测GFR<45的曲线下面积分别为0.89和0.9,AI模型无需临床细节、医生时间或测量即可达到与临床模型相似的准确性。

要点

  • AI模型与临床模型预测术后GFR的相关系数分别为0.75和0.77。
  • AI模型与临床模型预测GFR<45的曲线下面积分别为0.89和0.9。
  • 全自动AI预测无需临床细节、医生时间或测量,准确性不劣于经验证临床模型。

核心结论

  • 全自动预测新基线肾功能与经验证的临床模型一样准确,且无需临床细节、医生时间或测量。

研究设计与样本

  • 研究纳入KiTS19挑战赛中300例因肾肿瘤行肾切除术的患者,术前收集GFR,术后3-12个月测量新基线GFR。
  • 采用全自动方法从术前计算机断层扫描(CT)确定分肾功能(SRF),结合深度学习分割模型,并利用分割掩膜估算术后GFR。
  • AI模型估算公式:根治性肾切除术(RN)为1.24×术前GFR-RN×对侧SRF;部分性肾切除术(PN)为术前GFR的89%。
  • 临床模型估算公式:术后GFR=35+术前GFR×0.65-18(若RN)-年龄×0.25+3(若肿瘤>7 cm)-2(若糖尿病)。
  • 使用相关系数和逻辑回归比较AI与临床模型对术后GFR的估算及预测GFR<45的能力。

主要发现

  • 中位年龄60岁,女性占41%,62%行PN。
  • 中位肿瘤大小4.2 cm,92%为恶性。
  • 与实测术后GFR相比,AI模型和临床模型的相关系数分别为0.75和0.77。
  • AI模型和临床模型预测GFR<45的曲线下面积分别为0.89和0.9。

局限性

  • 原文未报告局限性

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

OBJECTIVE: To test if our artificial intelligence (AI)-postoperative glomerular filtration rate (GFR) prediction is as accurate as a validated clinical model. The American Urologic Association recommends estimating postoperative GFR in patients with renal masses and prioritizing partial nephrectomy (PN) when GFR would be <45 ml/minutes/1.73 m2 if radical nephrectomy (RN) was performed. Previously validated models have limited clinical uptake. METHODS: We included 300 patients undergoing nephrectomy for renal tumors from the KiTS19 challenge. Preoperative GFR was collected just before surgery, and new baseline GFR 3-12 months postoperatively. Split-renal function (SRF) was determined in a fully automated way from preoperative computed tomography, combining our deep learning segmentation model, then using those segmentation masks to estimate postoperative GFR = 1.24 × GFRPre-RN × SRFContralateral for RN and 89% of GFRpreoperative for PN. A clinical model estimated postoperative GFR = 35 + GFRpreoperative x 0.65-18 (if RN)-age x 0.25 + 3 (if tumor>7 cm)-2 (if diabetes). We compared the AI and clinical model GFR estimations to the measured postoperative GFR using correlation coefficients and their ability to predict GFR < 45 using logistic regression. RESULTS: Median age was 60 years, 41% were female, and 62% had PN. Median tumor size was 4.2 cm, and 92% were malignant. Compared to the measured postoperative GFR, correlation coefficients were 0.75 and 0.77 for the AI and clinical models, respectively. The AI and clinical models performed similarly for predicting GFR < 45 (areas under the curve 0.89 and 0.9, respectively). CONCLUSION: Our fully automated prediction of new baseline renal function is as accurate as a validated clinical model without needing clinical details, clinician time, or measurements.

来源:Urology,查看原文。版权归原作者所有。

引用

  1. Abdallah N, Rathi N, Heller N, et al. A Fully Automated Artificial Intelligence-Based Approach to Predict Renal Function After Radical or Partial Nephrectomy. Urology. 2025. PMID: 39914676. doi:10.1016/j.urology.2025.01.073

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。