Raman spectroscopy based analysis of milk using random forest classification

Amjad, A.; Ullah, Rahat; Khan, Saranjam; Bilal, Muhammad; Khan, Asifullah

doi:10.1016/j.vibspec.2018.09.003

Cited by 64 publications

(36 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The laser wavelength is important, especially in food matrices; as was stated in the study of Amjad et al, owing to the high fluorescence background, some extra data processing had to be performed in order to eliminate the fluorescence signal. 14 In the present study, interference of fluorescence was limited, and this can improve the classification power. Also, contrary to the methods based on laserinduced breakdown spectroscopy, the proposed method has prominent features, such as no need for any sample preparation steps.…”

Section: Discrimination According To Heat Treatment By Using Pls-damentioning

confidence: 75%

“…26 Also, there are some variations among the spectra of milk samples in the 343-386 cm −1 , 913-921, and 1470-1520 cm −1 regions. The band in the 343-386 cm −1 region represents the stretching and bending vibrations of the C O C, 25 the 913-921 cm −1 region represents the stretching vibration of the ν(C C), 14 and the 1470-1521 cm −1 region represents the methylene scissoring deformation ⊐(CH 2 )sc. 8,26 Furthermore, it was found that the spectra had minimal variations above a waveumber of 1600 cm −1 .…”

Section: Discussionmentioning

confidence: 99%

“…In previous studies, discrimination was performed for only raw or pasteurized milks. 9,30 Also, different from the study of Amjad et al, in which the discrimination of milk species according to their species as cow, goat, buffalo, or human by using Raman spectroscopy, 14 the present study incorporates ewe milk instead of buffalo and human milk. In addition, the laser wavelength of the Raman spectroscopy used in the present study was 785 nm, whereas it was 532 nm in that study.…”

Section: Discrimination According To Heat Treatment By Using Pls-damentioning

confidence: 92%

“…In that study, origin of cow, goat, buffalo, and human milk was determined through random forest classification. 14 In the present study, Raman spectroscopy coupled with partial least squares discriminant analysis (PLS-DA) was used. Although a variety of chemometric models have been successfully used for the discrimination of samples or determination of adulterations, including soft independent model of class analogy (SIMCA), k-nearest neighbor (k-NN), and PLS-DA, as far as we know, this is the first use of PLS-DA together with Raman spectroscopy for the discrimination of milk species.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Discrimination of milk species using Raman spectroscopy coupled with partial least squares discriminant analysis in raw and pasteurized milk

et al. 2020

View full text Add to dashboard Cite

BACKGROUND Heat treatment is the most common practice for the microbiological safety of milk; hence, determination of the heat treatment of milk is essential. Also, mislabeling or adulteration of expensive milk samples, like ewe or goat milk, with cow's milk is a growing problem in the dairy market. Thus, the determination of the authenticity of milk samples has crucial importance for both producers and consumers. The aim of this study was to discriminate milk samples using Raman spectroscopy with partial least squares discriminant analysis (PLS‐DA), first with regard to whether the milk was heat‐treated or not, and second with regard to species (cow, goat, ewe, mixture (adulterated)) in both raw and pasteurized milk. RESULTS First, discrimination of milk samples as raw or pasteurized was achieved using PLS‐DA. Both in calibration and prediction models, high sensitivity and specificity values were obtained for raw and pasteurized milk samples. Second, the proposed method also discriminated milk samples according to their species (cow, goat, ewe, and mixture) for both raw and pasteurized milk. In both calibration and prediction models, the sensitivity and specificity values were above 0.857 and 0.897 respectively. Also, the accuracy values were above 0.915. The results obtained denote satisfactory accurate classification of the samples. CONCLUSION The results suggest that Raman spectroscopy coupled with PLS‐DA can be successfully used to discriminate milk samples according to heat treatment (raw/pasteurized) and their species within 20 s per sample. It was seen that Raman spectra provide valuable information to be used especially for discrimination of milk samples according to their origin. © 2020 Society of Chemical Industry

show abstract

Section: Discrimination According To Heat Treatment By Using Pls-damentioning

confidence: 75%

Section: Discussionmentioning

confidence: 99%

Section: Discrimination According To Heat Treatment By Using Pls-damentioning

confidence: 92%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Discrimination of milk species using Raman spectroscopy coupled with partial least squares discriminant analysis in raw and pasteurized milk

et al. 2020

View full text Add to dashboard Cite

show abstract

“…The analysis of principal components (PCA) was used as a reduction technology in combination with RF to emphasize changes in Raman spectra that can differentiate milk samples from different species. The proposed method shows a sufficient opportunity to distinguish samples of cow milk from different species due to an average accuracy of about 94%, a specificity of about 97%, and a sensitivity of about 93% [94].…”

Section: Raman Spectroscopy-based Analysismentioning

confidence: 99%

Reconnoitering Milk Constituents of Different Species, Probing and Soliciting Factors to Its Soundness

Aqib¹,

Ijaz²,

Anjum³

et al. 2019

Milk Production, Processing and Marketing

View full text Add to dashboard Cite

Milk composition and production varies from species to species, reflecting its diversified benefits on health. Lipids from caprine and ovine milk are anti-obesity and anti-atherogenic while prebiotic in the case of caprine. Higher contents of selenium from caprine and iron from camel milk play a role in immune system and oxygen transport system, respectively, whereas enriched vitamins like riboflavin, folic acid, B6, vitamin A of bovine, and foliate of cattle are effective in the synthesis of hemoglobin, and high niacin content of caprine is anti-cancerous. Camel milk is found to have characteristics of anti-carcinogenic, antidiabetic, and autoimmune therapeutic. Various processing techniques like pasteurization, skim milk powder processing, and ultra-high temperature processing are necessary for safe provision of milk to meet consumers' demand. Change in flavor, loss of micronutrients, biofilm production, and spore-forming bacteria are prominent challenges during processing. Antimicrobial resistance and disease conditions are exaggerating factors of milk deterioration with respect to quality and quantity. Preclinical trials like somatic cell count, California mastitis test, proteomic analysis, Raman spectroscopy-based analysis, and X-ray fluorescence analysis are helpful in avoiding the spread of disease and controlling of economic losses. This chapter focuses differential functions of bioactive of milk, issues arising during processing techniques, and preclinical studies of milk for safer production and consumption of milk.

show abstract

An Alloy Platform of Dual‐Fingerprints for High‐Performance Stroke Screening

Shu

Zhang

et al. 2022

Adv Funct Materials

View full text Add to dashboard Cite

A multi‐modal serum profiling platform holds promise for precision diagnosis of diseases. Still, advanced tools are in demand to deliver the multi‐modal serum profiling. Herein, a bimodal spectrometric protocol is designed for stoke serum profiling using an alloy platform, by integrating label‐free surface‐enhanced Raman spectroscopy (SERS) and laser desorption/ionization mass spectrometry (LDI‐MS). The PdAu@Au concave cube with a wide localized surface plasmonic resonance (LSPR) range simultaneously enhances the signals from SERS and LDI‐MS, enabling high‐throughput co‐detection of vibrational and metabolic fingerprints of 0.1 µL serum in 2 min with simple pretreatment. Further, a dual‐fingerprints screening model of stroke is constructed, by adaptive machine learning with a programming nonlinear fitting model. The area under the curve are 0.949 (0.917–0.977, 95% confidence interval (CI)) and 0.911 (0.812–0.984, 95% CI), in the discovery and validation cohorts, respectively. Finally, five metabolites are identified that correlated to SERS signals and mapped the relevant pathways. This study features high performance in terms of throughput, speed, sample volume, and accuracy, providing new insight into the construction of multiplexed characterization platforms for precision diagnostic.

show abstract

Raman spectroscopy based analysis of milk using random forest classification

Cited by 64 publications

References 19 publications

Discrimination of milk species using Raman spectroscopy coupled with partial least squares discriminant analysis in raw and pasteurized milk

Discrimination of milk species using Raman spectroscopy coupled with partial least squares discriminant analysis in raw and pasteurized milk

Reconnoitering Milk Constituents of Different Species, Probing and Soliciting Factors to Its Soundness

An Alloy Platform of Dual‐Fingerprints for High‐Performance Stroke Screening

Contact Info

Product

Resources

About