CIE-Lab* Coupled with AI/ML for Rapid Detection of Rhodamine-b in Food Matrix

IF 3 3区农林科学 Q2 FOOD SCIENCE & TECHNOLOGY

Food Analytical Methods Pub Date : 2026-09-29 DOI:10.1007/s12161-026-03291-6

Amrita Shaw, Abhirami S. J., Khusboo Chhetri, Chandrasekar S. N., Deepak L. N. Kallepalli, Jhinuk Gupta

{"title":"CIE-L*a*b* Coupled with AI/ML for Rapid Detection of Rhodamine-b in Food Matrix","authors":"Amrita Shaw, Abhirami S. J., Khusboo Chhetri, Chandrasekar S. N., Deepak L. N. Kallepalli, Jhinuk Gupta","doi":"10.1007/s12161-026-03291-6","DOIUrl":null,"url":null,"abstract":"<div><p>Rh-b, a synthetic dye banned by the Food Safety and Standards Authority of India (FSSAI), continues to be illegally used in processed, ready-to-eat foods to enhance visual appeal. Conventional detection methods such as HPLC and UV–visible spectroscopy require expensive instrumentation and elaborate sample processing. This proof-of-concept study presents a non-invasive AI/ML tool for rapid, qualitative detection of Rh-b in processed foods without isolating the dye from the food. Six structurally and compositionally diverse food matrices (chilli powder, chilli sauce, cotton candy, hard candy, gummy, and gobi Manchurian) were spiked with Rh-b across a known range of concentrations; colorimetric data (CIE-L*a*b*) was recorded using color reader and validated statistically through analysis of variance (with effect size) and Principal Component Analysis (PCA). Four machine learning algorithms (Random Forest (RF), Support Vector Machine (SVM-(RBF)), K-Nearest Neighbours (KNN), and Logistic Regression) were benchmarked for model development. RF and SVM (RBF) provided the most consistent performance (Accuracy = 97.2%), with RF showing marginal superiority in selective performance metrics. Feature importance and effect-size analyses revealed the matrix-specific nature of the discriminative color parameters. The tool reliably detected Rhodamine-b (Rh-b) at concentrations as low as 25 ppm in five out of six matrices. Independent method comparison with UV–visible spectroscopy confirmed superior performance of the developed tool at lower concentrations. </p><h3>Graphical Abstract</h3>\n<div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":561,"journal":{"name":"Food Analytical Methods","volume":"19 11","pages":""},"PeriodicalIF":3.0000,"publicationDate":"2026-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Food Analytical Methods","FirstCategoryId":"97","ListUrlMain":"https://link.springer.com/article/10.1007/s12161-026-03291-6","RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"FOOD SCIENCE & TECHNOLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

Rh-b, a synthetic dye banned by the Food Safety and Standards Authority of India (FSSAI), continues to be illegally used in processed, ready-to-eat foods to enhance visual appeal. Conventional detection methods such as HPLC and UV–visible spectroscopy require expensive instrumentation and elaborate sample processing. This proof-of-concept study presents a non-invasive AI/ML tool for rapid, qualitative detection of Rh-b in processed foods without isolating the dye from the food. Six structurally and compositionally diverse food matrices (chilli powder, chilli sauce, cotton candy, hard candy, gummy, and gobi Manchurian) were spiked with Rh-b across a known range of concentrations; colorimetric data (CIE-L*a*b*) was recorded using color reader and validated statistically through analysis of variance (with effect size) and Principal Component Analysis (PCA). Four machine learning algorithms (Random Forest (RF), Support Vector Machine (SVM-(RBF)), K-Nearest Neighbours (KNN), and Logistic Regression) were benchmarked for model development. RF and SVM (RBF) provided the most consistent performance (Accuracy = 97.2%), with RF showing marginal superiority in selective performance metrics. Feature importance and effect-size analyses revealed the matrix-specific nature of the discriminative color parameters. The tool reliably detected Rhodamine-b (Rh-b) at concentrations as low as 25 ppm in five out of six matrices. Independent method comparison with UV–visible spectroscopy confirmed superior performance of the developed tool at lower concentrations.

Graphical Abstract

查看原文本刊更多论文

CIE-L*a*b*结合AI/ML快速检测食品基质中罗丹明-b

Rh-b是一种被印度食品安全与标准局（FSSAI）禁止的合成染料，它继续被非法用于加工即食食品中，以增加视觉吸引力。传统的检测方法，如高效液相色谱法和紫外可见光谱法，需要昂贵的仪器和复杂的样品处理。这项概念验证研究提出了一种非侵入性AI/ML工具，用于快速定性检测加工食品中的Rh-b，而无需从食品中分离染料。六种结构和成分不同的食物基质（辣椒粉、辣椒酱、棉花糖、硬糖、软糖和戈壁满洲）在已知的浓度范围内加入了Rh-b；使用颜色阅读器记录比色数据（CIE-L*a*b*），并通过方差分析（带效应量）和主成分分析（PCA）进行统计验证。四种机器学习算法(随机森林（RF），支持向量机（SVM-（RBF)）， k近邻（KNN）和逻辑回归）被用于模型开发的基准。RF和SVM （RBF）提供了最一致的性能（准确率= 97.2%），RF在选择性性能指标上显示出边际优势。特征重要性和效应大小分析揭示了区分颜色参数的矩阵特异性。该工具在六种基质中的五种中可靠地检测到浓度低至25 ppm的罗丹明-b （Rh-b）。独立方法与紫外-可见光谱的比较证实了该工具在较低浓度下的优越性能。图形抽象

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

Food Analytical Methods 农林科学-食品科技

CiteScore

6.00

自引率

3.40%

发文量

244

审稿时长

3.1 months

期刊介绍： Food Analytical Methods publishes original articles, review articles, and notes on novel and/or state-of-the-art analytical methods or issues to be solved, as well as significant improvements or interesting applications to existing methods. These include analytical technology and methodology for food microbial contaminants, food chemistry and toxicology, food quality, food authenticity and food traceability. The journal covers fundamental and specific aspects of the development, optimization, and practical implementation in routine laboratories, and validation of food analytical methods for the monitoring of food safety and quality.