Dual-Mode Sweat Urea Sensor based on Ti3C2Tx MXene/CuO nanocomposite: Colorimetric-Electrochemical Detection Optimized by Machine Learning and Genetic Algorithms

IF 3.7 1区化学 Q1 CHEMISTRY, ANALYTICAL

Sensors and Actuators B: Chemical Pub Date : 2025-09-29 DOI:10.1016/j.snb.2025.138882

Thidarut Laochai, Hirotomo Nishihara, Jiaqian Qin, Manunya Okhawilai, Joseph Wang, Nadnudda Rodthongkum, Wiwittawin Sukmas

{"title":"Dual-Mode Sweat Urea Sensor based on Ti3C2Tx MXene/CuO nanocomposite: Colorimetric-Electrochemical Detection Optimized by Machine Learning and Genetic Algorithms","authors":"Thidarut Laochai, Hirotomo Nishihara, Jiaqian Qin, Manunya Okhawilai, Joseph Wang, Nadnudda Rodthongkum, Wiwittawin Sukmas","doi":"10.1016/j.snb.2025.138882","DOIUrl":null,"url":null,"abstract":"This study presents a high-performance dual-mode wearable sensor for real-time, non-invasive sweat urea monitoring, integrating Ti3C2Tx MXene/CuO nanocomposites with machine learning and genetic algorithm (ML-GA) optimization. The sensor combines two complementary modalities: (1) a urease-based colorimetric assay on nanocomposite-modified cotton for semi-quantitative detection (10–80 mM, R² = 0.9855), and (2) a non-enzymatic electrochemical system using Ti3C2Tx MXene/CuO/PEDOT:PSS-modified electrodes for precise quantification (0.5–60 mM, R² = 0.9706). The nanocomposite’s high surface area, conductivity, and catalytic activity enhance both sensing mechanisms, supported by DFT calculations showing strong interfacial interactions and urea adsorption. An ML framework evaluated >20 regression models, identifying Random Forest Regression as most accurate (R² = 0.9650 for colorimetric; 0.8900 for electrochemical). GA optimization fine-tuned material ratios and parameters. On-body validation showed strong sweat-to-blood urea correlation (p < 0.01), confirming clinical relevance. This ML-GA-enhanced dual-mode sensor offers a reliable, user-friendly tool for personalized health monitoring and early renal disorder detection.","PeriodicalId":425,"journal":{"name":"Sensors and Actuators B: Chemical","volume":"4 1","pages":""},"PeriodicalIF":3.7000,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Sensors and Actuators B: Chemical","FirstCategoryId":"92","ListUrlMain":"https://doi.org/10.1016/j.snb.2025.138882","RegionNum":1,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CHEMISTRY, ANALYTICAL","Score":null,"Total":0}

引用次数: 0

Abstract

This study presents a high-performance dual-mode wearable sensor for real-time, non-invasive sweat urea monitoring, integrating Ti₃C₂T_x MXene/CuO nanocomposites with machine learning and genetic algorithm (ML-GA) optimization. The sensor combines two complementary modalities: (1) a urease-based colorimetric assay on nanocomposite-modified cotton for semi-quantitative detection (10–80 mM, R² = 0.9855), and (2) a non-enzymatic electrochemical system using Ti₃C₂T_x MXene/CuO/PEDOT:PSS-modified electrodes for precise quantification (0.5–60 mM, R² = 0.9706). The nanocomposite’s high surface area, conductivity, and catalytic activity enhance both sensing mechanisms, supported by DFT calculations showing strong interfacial interactions and urea adsorption. An ML framework evaluated >20 regression models, identifying Random Forest Regression as most accurate (R² = 0.9650 for colorimetric; 0.8900 for electrochemical). GA optimization fine-tuned material ratios and parameters. On-body validation showed strong sweat-to-blood urea correlation (p < 0.01), confirming clinical relevance. This ML-GA-enhanced dual-mode sensor offers a reliable, user-friendly tool for personalized health monitoring and early renal disorder detection.

Abstract Image

查看原文本刊更多论文

基于Ti3C2Tx MXene/CuO纳米复合材料的汗液尿素双模传感器：基于机器学习和遗传算法优化的比色-电化学检测

本研究将Ti3C2Tx MXene/CuO纳米复合材料与机器学习和遗传算法（ML-GA）优化相结合，提出了一种用于实时、无创汗液尿素监测的高性能双模可穿戴传感器。该传感器结合了两种互补模式：(1)基于脲酶的纳米复合改性棉花比色法进行半定量检测（10-80 mM, R²= 0.9855）；(2)使用Ti3C2Tx MXene/CuO/PEDOT: pss修饰电极的非酶电化学系统进行精确定量（0.5-60 mM, R²= 0.9706）。纳米复合材料的高表面积、导电性和催化活性增强了这两种传感机制，DFT计算显示出强的界面相互作用和尿素吸附。ML框架评估了20个回归模型，确定随机森林回归是最准确的（R²= 0.9650比色法；0.8900电化学）。遗传算法优化对材料配比和参数进行了微调。身体验证显示汗液与血尿素有很强的相关性（p < 0.01），证实了临床相关性。这种ml - ga增强的双模传感器为个性化健康监测和早期肾脏疾病检测提供了可靠、用户友好的工具。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Sensors and Actuators B: Chemical 工程技术-电化学

CiteScore

14.60

自引率

11.90%

发文量

1776

审稿时长

3.2 months

期刊介绍： Sensors & Actuators, B: Chemical is an international journal focused on the research and development of chemical transducers. It covers chemical sensors and biosensors, chemical actuators, and analytical microsystems. The journal is interdisciplinary, aiming to publish original works showcasing substantial advancements beyond the current state of the art in these fields, with practical applicability to solving meaningful analytical problems. Review articles are accepted by invitation from an Editor of the journal.