Combining elemental analysis of toenails and machine learning techniques as a non-invasive diagnostic tool for the robust classification of type-2 diabetes

作者：

Highlights：

• A toenail-based non-invasive method for diagnosing type-2 diabetes was developed.

• Al, Cs, Ni, V, Zn in toenails were significantly different for diabetes patients.

• Toenail concentrations of 22 elements were used for machine learning modeling.

• A random forest model correctly classified 7 out of 9 samples, with AUC = 0.90.

摘要

•A toenail-based non-invasive method for diagnosing type-2 diabetes was developed.•Al, Cs, Ni, V, Zn in toenails were significantly different for diabetes patients.•Toenail concentrations of 22 elements were used for machine learning modeling.•A random forest model correctly classified 7 out of 9 samples, with AUC = 0.90.

论文关键词：Diabetes diagnosis,Machine learning,Trace Elemental analysis,Chemometrics,ICP-MS,MIP OES

论文评审过程：Received 18 May 2018, Revised 31 July 2018, Accepted 1 August 2018, Available online 2 August 2018, Version of Record 15 August 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.08.002