Huang Xingyue, Hu Yugang, Zhang Yao, Zhou Qing
Department of Ultrasound Imaging, Renmin Hospital of Wuhan University, Wuhan, 430061, People's Republic of China.
Int J Gen Med. 2024 May 6;17:1877-1885. doi: 10.2147/IJGM.S462896. eCollection 2024.
To establish a radiomics nomogram based on two-dimensional ultrasound for risk assessment of diabetic kidney disease (DKD) in patients with type 2 diabetes mellitus (T2DM).
This study retrospectively collected two-dimensional ultrasound images and clinical data from 52 patients with T2DM who underwent renal biopsy in our hospital from January 2023 to August 2023. Based on the pathological results, all patients were categorized into two groups: DKD (n=33) and non-DKD (n=19). The radiomic features of the segmented kidney in ultrasound pictures were retrieved and selected to calculate each patient's rad-score. A predictive nomogram based on rad-score and clinical features was then constructed and validated based on the calibration curve.
The rad-score for all patients were computed based on five imaging characteristics extracted from the ultrasound images. The predictive nomogram was developed with the rad-score, diabetic retinopathy, duration of diabetes, and glycosylated hemoglobin. Moreover, This radiomics nomogram showed outstanding calibration capability, discrimination as well as therapeutic usefulness.
We constructed a nomogram based on two-dimensional ultrasound for DKD in T2DM patientsThe model has been proven to have good predictive performance, showing its potential in identifying DKD in T2DM patients and assisting in making appropriate early interventions.
建立基于二维超声的列线图，用于评估2型糖尿病（T2DM）患者糖尿病肾病（DKD）的风险。
本研究回顾性收集了2023年1月至2023年8月在我院接受肾活检的52例T2DM患者的二维超声图像和临床资料。根据病理结果，将所有患者分为两组：DKD组（n = 33）和非DKD组（n = 19）。提取并选择超声图像中分割肾脏的影像组学特征，计算每位患者的rad评分。然后基于rad评分和临床特征构建预测列线图，并根据校准曲线进行验证。
根据从超声图像中提取的五个成像特征计算所有患者的rad评分。利用rad评分、糖尿病视网膜病变、糖尿病病程和糖化血红蛋白建立了预测列线图。此外，该影像组学列线图显示出出色的校准能力、区分能力和治疗实用性。
我们构建了基于二维超声的T2DM患者DKD列线图。该模型已被证明具有良好的预测性能，显示出其在识别T2DM患者DKD及协助进行适当早期干预方面的潜力。
