一种全自动基于深度学习的方法，用于在儿科动态肾闪烁显像中分割感兴趣区域并预测肾功能。

A fully automatic deep learning-based method for segmenting regions of interest and predicting renal function in pediatric dynamic renal scintigraphy.

机构信息

Department of Nuclear Medicine, Xinhua Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, 200092, China.

Institute of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai, 200092, China.

出版信息

Ann Nucl Med. 2024 May;38(5):382-390. doi: 10.1007/s12149-024-01907-7. Epub 2024 Feb 20.

DOI:10.1007/s12149-024-01907-7

PMID:38376629

Abstract

OBJECTIVE

Accurate delineation of renal regions of interest (ROIs) is critical for the assessment of renal function in pediatric dynamic renal scintigraphy (DRS). The purpose of this study was to develop and evaluate a deep learning (DL) model that can fully automatically delineate renal ROIs and calculate renal function in pediatric Technetium-ethylenedicysteine (Tc-EC) DRS.

METHODS

This study retrospectively analyzed 1,283 pediatric DRS data at a single center from January to December 2018. These patients were divided into training set (n = 1027), validation set (n = 128), and testing set (n = 128). A fully automatic segmentation of ROIs (FASR) model was developed and evaluated. The pixel values of the automatically segmented ROIs were calculated to predict renal blood perfusion rate (BPR) and differential renal function (DRF). Precision, recall rate, intersection over union (IOU), and Dice similarity coefficient (DSC) were used to evaluate the performance of FASR model. Intraclass correlation (ICC) and Pearson correlation analysis were used to compare the consistency of automatic and manual method in assessing the renal function parameters in the testing set.

RESULTS

The FASR model achieved a precision of 0.88, recall rate of 0.94, IOU of 0.83, and DSC of 0.91. In the testing set, the r values of BPR and DRF calculated by the two methods were 0.94 (P < 0.01) and 0.97 (P < 0.01), and the ICCs (95% confidence interval CI) were 0.94 (0.90-0.96) and 0.94 (0.91-0.96).

CONCLUSION

We propose a reliable and stable DL model that can fully automatically segment ROIs and accurately predict renal function in pediatric Tc-EC DRS.

摘要

目的

准确勾画肾脏感兴趣区（ROI）对于儿童动态肾闪烁显像（DRS）中肾功能的评估至关重要。本研究旨在开发并评估一种深度学习（DL）模型，该模型可全自动勾画儿童锝-乙二胺五乙酸（Tc-EC）DRS 的 ROI 并计算肾功能。

方法

本研究回顾性分析了 2018 年 1 月至 12 月在单中心进行的 1283 例儿童 DRS 数据。这些患者被分为训练集（n=1027）、验证集（n=128）和测试集（n=128）。开发并评估了一个全自动 ROI 分割（FASR）模型。自动分割的 ROI 的像素值用于预测肾血流灌注率（BPR）和分肾功能（DRF）。精确率、召回率、交并比（IOU）和 Dice 相似系数（DSC）用于评估 FASR 模型的性能。组内相关系数（ICC）和 Pearson 相关分析用于比较测试集中自动和手动方法评估肾功能参数的一致性。

结果

FASR 模型的精确率为 0.88，召回率为 0.94，IOU 为 0.83，DSC 为 0.91。在测试集中，两种方法计算的 BPR 和 DRF 的 r 值分别为 0.94（P<0.01）和 0.97（P<0.01），ICC（95%置信区间 CI）分别为 0.94（0.90-0.96）和 0.94（0.91-0.96）。