Application of Surface-Enhanced Raman Spectroscopy and Machine Learning Omics Techniques in the Progression Assessment of Autosomal Dominant Polycystic Kidney Disease

表面增强拉曼光谱和机器学习组学技术在常染色体显性多囊肾病进展评估中的应用

阅读:1

Abstract

INTRODUCTION: The progression evaluation of autosomal dominant polycystic kidney disease (ADPKD) is critical to the treatment strategy and prognosis prediction of the disease, but the current evaluation methods are time-consuming and costly. METHODS: ADPKD patients were recruited in Shanghai Changzheng Hospital from February 2023 to October 2023. Based on the Mayo imaging classification (MIC) model, patients with different disease progression rates were classified into slowly progressive (SP) group (MIC 1A, 1B) or rapidly progressive (RP) group (MIC 1C, 1D, and 1E). The Raman spectra of urine samples from ADPKD patients were obtained by Surface-enhanced Raman spectroscopy (SERS) protocol and the characteristics of urine Raman spectra from SP and RP groups were analyzed. The ADPKD progression prediction model was constructed based on principal component analysis (PCA) and support vector machine (SVM). RESULTS: After adjusting for age, gender, and renal function, there were differences in Raman intensity of the six major Raman peaks (758/794/1,184/1,288/1,346/1,385 cm(-1)) between patients with SP and RP ADPKD (p < 0.05). The substances of the above Raman peaks include tryptophan, uracil, adenine, phosphodiester bond, glucose, lipids, etc. The accuracy, sensitivity, and specificity of the ADPKD progression prediction model were 82.93%, 77.55%, and 86.11%, and after leaving-one-out cross validation (LOOCV), the accuracy, sensitivity, and specificity of were 70.25%, 65.31%, and 73.61%, respectively. CONCLUSIONS: Urinary SERS detection can distinguish RP ADPKD patients from SP ones noninvasively, conveniently and quickly, with reasonable accuracy, sensitivity, and specificity. SERS combined with machine learning omics techniques might be a novel method to evaluate the progression of ADPKD.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。