Herbal-based products are becoming a widespread production trend among manufacturers for the domestic and international markets. As the production increases to meet the market demand, it is very crucial for the manufacturer to ensure that their products have met specific criteria and fulfil the intended quality determined by the quality controller. One famous herbal-based product is herbal tea. This paper investigates bio-inspired flavour assessments in a data fusion framework involving an e-nose and e-tongue. The objectives are to attain good classification of different types and brands of herbal tea, classification of different flavour masking effects and finally classification of different concentrations of herbal tea. Two data fusion levels were employed in this research, low level data fusion and intermediate level data fusion. Four classification approaches; LDA, SVM, KNN and PNN were examined in search of the best classifier to achieve the research objectives. In order to evaluate the classifiers' performance, an error estimator based on k-fold cross validation and leave-one-out were applied. Classification based on GC-MS TIC data was also included as a comparison to the classification performance using fusion approaches. Generally, KNN outperformed the other classification techniques for the three flavour assessments in the low level data fusion and intermediate level data fusion. However, the classification results based on GC-MS TIC data are varied.
A study was conducted to detect and quantify lard stearin (LS) content in canola oil (CaO) using differential scanning calorimetry (DSC). Authentic samples of CaO were obtained from a reliable supplier and the adulterant LS were obtained through a fractional crystallization procedure as reported previously. Pure CaO samples spiked with LS in levels ranging from 5 to 15% (w/w) were analyzed using DSC to obtain their cooling and heating profiles. The results showed that samples contaminated with LS at 5% (w/w) level can be detected using characteristic contaminant peaks appearing in the higher temperature regions (0 to 70°C) of the cooling and heating curves. Pearson correlation analysis of LS content against individual DSC parameters of the adulterant peak namely peak temperature, peak area, peak onset temperature indicated that there were strong correlations between these with the LS content of the CaO admixtures. When these three parameters were engaged as variables in the execution of the stepwise regression procedure, predictive models for determination of LS content in CaO were obtained. The predictive models obtained with single DSC parameter had relatively lower coefficient of determination (R(2) value) and higher standard error than the models obtained using two DSC parameters in combination. This study concluded that the predictive models obtained with peak area and peak onset temperature of the adulteration peak would be more accurate for prediction of LS content in CaO based on the highest coefficient of determination (R(2) value) and smallest standard error.
A simple visual ethanol biosensor based on alcohol oxidase (AOX) immobilised onto polyaniline (PANI) film for halal verification of fermented beverage samples is described. This biosensor responds to ethanol via a colour change from green to blue, due to the enzymatic reaction of ethanol that produces acetaldehyde and hydrogen peroxide, when the latter oxidizes the PANI film. The procedure to obtain this biosensor consists of the immobilization of AOX onto PANI film by adsorption. For the immobilisation, an AOX solution is deposited on the PANI film and left at room temperature until dried (30 min). The biosensor was constructed as a dip stick for visual and simple use. The colour changes of the films have been scanned and analysed using image analysis software (i.e., ImageJ) to study the characteristics of the biosensor's response toward ethanol. The biosensor has a linear response in an ethanol concentration range of 0.01%-0.8%, with a correlation coefficient (r) of 0.996. The limit detection of the biosensor was 0.001%, with reproducibility (RSD) of 1.6% and a life time up to seven weeks when stored at 4 °C. The biosensor provides accurate results for ethanol determination in fermented drinks and was in good agreement with the standard method (gas chromatography) results. Thus, the biosensor could be used as a simple visual method for ethanol determination in fermented beverage samples that can be useful for Muslim community for halal verification.
A total of 126 food samples, categorised into three groups (seafood and seafood products, meat and meat products, as well as milk and dairy products) from Malaysia were analysed for polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs). The concentration of PCDD/Fs that ranged from 0.16 to 0.25 pg WHO05-TEQ g(-1) fw was found in these samples. According to the food consumption data from the Global Environment Monitoring System (GEMS) of the World Health Organization (WHO), the dietary exposures to PCDD/F from seafood and seafood products, meat and meat products, as well as milk and dairy products for the general population in Malaysia were 0.064, 0.183 and 0.736 pg WHO05-TEQ kg(-1) bw day(-1), respectively. However, the exposure was higher in seafood and seafood products (0.415 pg WHO05-TEQ kg(-1) bw day(-1)) and meat and meat products (0.317 pg WHO05-TEQ kg(-1) bw day(-1)) when the data were estimated using the Malaysian food consumption statistics. The lower exposure was observed in dairy products with an estimation of 0.365 pg WHO05-TEQ kg(-1) bw day(-1). Overall, these dietary exposure estimates were much lower than the tolerable daily intake (TDI) as recommended by WHO. Thus, it is suggested that the dietary exposure to PCDD/F does not represent a risk for human health in Malaysia.
A study was conducted to differentiate lard, chicken fat, beef fat and mutton fat using Gas Chromatography Mass Spectrometry (GC-MS) and Elemental Analyzer-Isotope Ratio Mass Spectrometry (EA-IRMS). The comparison of overall fatty acid data showed that lard and chicken fat share common characteristics by having palmitic, oleic and linoleic acid as major fatty acids while beef and mutton fats shared common characteristics by possessing palmitic, stearic and oleic acid as major fatty acids. The direct comparisons among the fatty acid data, therefore, may not be suitable for discrimination of different animal fats. When the fatty acid distributional data was subjected to Principle Component Analysis (PCA), it was demonstrated that stearic, oleic and linoleic acids as the most discriminating parameters in the clustering of animal fats into four subclasses. The bulk carbon analysis of animal fats using EA-IRMS showed that determination of the carbon isotope ratios (δ¹³C) would be a good indicator for discriminating lard, chicken fat, beef fat and mutton fat. This would lead to a faster and more efficient method to ascertain the source of origin of fats used in food products.
In dairy product sector, butter is one of the potential sources of fat soluble vitamins, namely vitamin A, D, E, K; consequently, butter is taken into account as high valuable price from other dairy products. This fact has attracted unscrupulous market players to blind butter with other animal fats to gain economic profit. Animal fats like mutton fat (MF) are potential to be mixed with butter due to the similarity in terms of fatty acid composition. This study focused on the application of FTIR-ATR spectroscopy in conjunction with chemometrics for classification and quantification of MF as adulterant in butter. The FTIR spectral region of 3910-710 cm⁻¹ was used for classification between butter and butter blended with MF at various concentrations with the aid of discriminant analysis (DA). DA is able to classify butter and adulterated butter without any mistakenly grouped. For quantitative analysis, partial least square (PLS) regression was used to develop a calibration model at the frequency regions of 3910-710 cm⁻¹. The equation obtained for the relationship between actual value of MF and FTIR predicted values of MF in PLS calibration model was y = 0.998x + 1.033, with the values of coefficient of determination (R²) and root mean square error of calibration are 0.998 and 0.046% (v/v), respectively. The PLS calibration model was subsequently used for the prediction of independent samples containing butter in the binary mixtures with MF. Using 9 principal components, root mean square error of prediction (RMSEP) is 1.68% (v/v). The results showed that FTIR spectroscopy can be used for the classification and quantification of MF in butter formulation for verification purposes.
This paper reports the design of an electronic nose (E-nose) prototype for reliable measurement and correct classification of beverages. The prototype was developed and fabricated in the laboratory using commercially available metal oxide gas sensors and a temperature sensor. The repeatability, reproducibility and discriminative ability of the developed E-nose prototype were tested on odors emanating from different beverages such as blackcurrant juice, mango juice and orange juice, respectively. Repeated measurements of three beverages showed very high correlation (r > 0.97) between the same beverages to verify the repeatability. The prototype also produced highly correlated patterns (r > 0.97) in the measurement of beverages using different sensor batches to verify its reproducibility. The E-nose prototype also possessed good discriminative ability whereby it was able to produce different patterns for different beverages, different milk heat treatments (ultra high temperature, pasteurization) and fresh and spoiled milks. The discriminative ability of the E-nose was evaluated using Principal Component Analysis and a Multi Layer Perception Neural Network, with both methods showing good classification results.
In this study, a rapid, specific and sensitive multi-residue method based on acetonitrile extraction followed by dispersive solid-phase extraction (d-SPE) clean-up was implemented and validated for multi-class pesticide residues determination in palm oil for the first time. Liquid-liquid extraction followed by low-temperature precipitation procedure was evaluated in order to study the freezing-out clean-up efficiency to obtain high recovery yield and low co-extract fat residue in the final extract. For clean-up step, d-SPE was carried out using a combination of anhydrous magnesium sulphate (MgSO(4)), primary secondary amine, octadecyl (C(18)) and graphitized carbon black. Recovery study was performed at two concentration levels (10 and 100 ng g(-1)), yielding recovery rates between 74.52% and 97.1% with relative standard deviation values below 10% (n = 6) except diuron. Detection and quantification limits were lower than 5 and 9 ng g(-1), respectively. In addition, soft matrix effects (≤±20%) were observed for most of the studied pesticides except malathion that indicated medium (20-50%) matrix effects. The proposed method was successfully applied to the analysis of suspected palm oil samples.
Yellow alkaline noodles (YAN) prepared by partial substitution of wheat flour with soy protein isolate and treated with microbial transglutaminase (MTG) and ribose were investigated during cooking. Cooking caused an increase in lightness but a decrease in redness and yellowness, pH, tensile strength and elasticity values of noodles. The extents of these changes were influenced by formulation and cross-linking treatments. The pH and lightness for YAN-ribose were lowest but the yellowness and redness were the highest whilst the tensile strength and elasticity values remained moderate. For YAN-MTG, the color and pH values were moderate, but tensile strength and elasticity values were the highest. YAN prepared with both cross-linking agents had physical values between YAN-ribose and YAN-MTG. Although certain sensory parameters showed differences in score, the overall acceptability of all 10-min-cooked YAN was similar. It is possible to employ cross-linking agents to improve physical properties of cooked YAN.
This study was conducted to evaluate the total carotene content (TCC) and beta carotene (BC) in the selected underutilized tropical fruits. TCC of underutilized fruits estimated by spectrophotometric method was in the range of 1.4-19.8 mg/100 g edible portion. The TCC of these fruits decreased in the order: Jentik-jentik > Durian Nyekak 2 > Durian Nyekak 1 > Cerapu 2 > Cerapu 1 > Tampoi Kuning > Bacang 1 > Kuini > Jambu Mawar > Bacang 2 > Durian Daun > Bacang 3 > Tampoi Putih > Jambu Susu. BC contents estimated by HPLC method were highest in Jentik-jentik, followed by Cerapu 2, Durian Nyekak 2, Tampoi Kuning, Durian Nyekak 1, and Cerapu 1, which had a range of 68-92% of BC in TCC. These underutilized fruits have an acceptable amount of carotenoids that are potential antioxidant fruits.
Red rice is a fermented product of Monascus spp. It is widely consumed by Malaysian Chinese who believe in its pharmacological properties. The traditional method of red rice preparation disregards safety regulation and renders red rice susceptible to fungal infestation and mycotoxin contamination. A preliminary study was undertaken aiming to determine the occurrence of mycotoxigenic fungi and mycotoxins contamination on red rice at consumer level in Selangor, Malaysia. Fifty red rice samples were obtained and subjected to fungal isolation, enumeration, and identification. Citrinin, aflatoxin, and ochratoxin-A were quantitated by ELISA based on the presence of predominant causal fungi. Fungal loads of 1.4 × 10(4) to 2.1 × 10(6) CFU/g exceeded Malaysian limits. Monascus spp. as starter fungi were present in 50 samples (100%), followed by Penicillium chrysogenum (62%), Aspergillus niger (54%), and Aspergillus flavus (44%). Citrinin was present in 100% samples (0.23-20.65 mg/kg), aflatoxin in 92% samples (0.61-77.33 μg/kg) and Ochratoxin-A in 100% samples (0.23-2.48 μg/kg); 100% citrinin and 76.09% aflatoxin exceeded Malaysian limits. The presence of mycotoxigenic fungi served as an indicator of mycotoxins contamination and might imply improper production, handling, transportation, and storage of red rice. Further confirmatory analysis (e.g., HPLC) is required to verify the mycotoxins level in red rice samples and to validate the safety status of red rice.
Sensorial analysis of pineapple breads (conventionally baked, Cpb; fully baked frozen, Fpb and partially baked, Ppb) showed no significant differences in terms of aroma and taste. On the contrary, the scores for the overall quality between the partially baked and conventionally baked breads showed significant (p < 0.05) differences. At the same time, headspace analysis using a solid-phase microextraction (SPME) method identified 59 volatile compounds. The results of the aroma extracts dilution analysis (AEDA) revealed 19 most odour-active compounds with FD factors in the range of 32-128 as the key odourants of the pineapple breads. Further analysis of the similarities and differences between the pineapple breads in terms of the key odourants were carried out by the application of PLS-DA and PLS-regression coefficients. Results showed that Ppb exhibited strong positive correlations with most of the volatile- and non-volatile compounds, while the Cpb showed significant positive correlations with hexanal and 4-hydroxy-2,5-dimethyl-3(2H)-furanone, and the Fpb had strong positive correlations with lactic acid, benzoic acid, benzaldehyde and ethyl propanoate.
A reversed-phased HPLC method that allows the separation and simultaneous determination of the preservatives benzoic (BA) and sorbic acids (SA), methyl- (MP) and propylparabens (PP) is described. The separations were effected by using an initial mobile phase of methanol-acetate buffer (pH 4.4) (35:65) to elute BA, SA and MP and changing the mobile phase composition to methanol-acetate buffer (pH 4.4) (50:50) thereafter. The detector wavelength was set at 254 nm. Under these conditions, separation of the four components was achieved in less than 23 min. Analytical characteristics of the separation such as limit of detection, limit of quantification, linear range and reproducibility were evaluated. The developed method was applied to the determination of 67 foodstuffs (mainly imported), comprising soft drinks, jams, sauces, canned fruits/vegetables, dried vegetables/fruits and others. The range of preservatives found were from not detected (nd)--1260, nd--1390, nd--44.8 and nd--221 mg kg(-1) for BA, SA, MP and PP, respectively.
An experimental nanosilver-coated low-density polyethylene (LDPE) food packaging was incubated with food simulants using a conventional oven and tested for migration according to European Commission Regulation No. 10/2011. The commercial LDPE films were coated using a layer-by-layer (LbL) technique and three levels of silver (Ag) precursor concentration (0.5%, 2% and 5% silver nitrate (AgNO3), respectively) were used to attach antimicrobial Ag. The experimental migration study conditions (time, temperature and food simulant) under conventional oven heating (10 days at 60°C, 2 h at 70°C, 2 h at 60°C or 10 days at 70°C) were chosen to simulate the worst-case storage period of over 6 months. In addition, migration was quantified under microwave heating. The total Ag migrant levels in the food simulants were quantified by inductively coupled plasma-atomic emission spectroscopy (ICP-AES). Mean migration levels obtained by ICP-AES for oven heating were in the range 0.01-1.75 mg l(-1). Migration observed for microwave heating was found to be significantly higher when compared with oven heating for similar temperatures (100°C) and identical exposure times (2 min). In each of the packaging materials and food simulants tested, the presence of nanoparticles (NPs) was confirmed by scanning electron microscopy (SEM). On inspection of the migration observed under conventional oven heating, an important finding was the significant reduction in migration resulting from the increased Ag precursor concentration used to attach Ag on the LDPE LbL-coated films. This observation merits further investigation into the LbL coating process used, as it suggests potential for process modifications to reduce migration. In turn, any reduction in NP migration below regulatory limits could greatly support the antimicrobial silver nanoparticle (AgNP)-LDPE LbL-coated films being used as a food packaging material.
Food forgery has posed considerable risk to public health, religious rituals, personal budget and wildlife. Pig, dog, cat, rat and monkey meat are restricted in most religions, but their sporadic adulteration are rampant. Market controllers need a low-cost but reliable technique to track and trace suspected species in the food chain. Considering the need, here we documented a lab-on-a-chip-based multiplex polymerase chain reaction (PCR) assay for the authentication of five non-halal meat species in foods. Using species-specific primers, 172, 163, 141, 129 and 108-bp sites of mitochondrial ND5, ATPase 6 and cytochrome b genes were amplified to detect cat, dog, pig, monkey and rat species under complex matrices. Species-specificity was authenticated against 20 different species with the potential to be used in food. The targets were stable under extreme sterilisation (121°C at 45 psi for 2.5 h) which severely degrades DNA. The assay was optimised under the backgrounds of various commercial meat products and validated for the analysis of meatballs, burgers and frankfurters, which are popular fast food items across the globe. The assay was tested to detect 0.1% suspected meats under commercial backgrounds of marketed foods. Instead of simplex PCR which detects only one species at a time, such a multiplex platform can reduce cost by at least fivefolds by detecting five different species in a single assay platform.
A sensitive and selective gas chromatography with mass spectrometry method was developed for the simultaneous determination of three organophosphorus pesticides, namely, chlorpyrifos, malathion, and diazinon in three different food commodities (milk, apples, and drinking water) employing solid-phase extraction for sample pretreatment. Pesticide extraction from different sample matrices was carried out on Chromabond C18 cartridges using 3.0 mL of methanol and 3.0 mL of a mixture of dichloromethane/acetonitrile (1:1 v/v) as the eluting solvent. Analysis was carried out by gas chromatography coupled with mass spectrometry using selected-ion monitoring mode. Good linear relationships were obtained in the range of 0.1-50 μg/L for chlorpyrifos, and 0.05-50 μg/L for both malathion and diazinon pesticides. Good repeatability and recoveries were obtained in the range of 78.54-86.73% for three pesticides under the optimized experimental conditions. The limit of detection ranged from 0.02 to 0.03 μg/L, and the limit of quantification ranged from 0.05 to 0.1 μg/L for all three pesticides. Finally, the developed method was successfully applied for the determination of three targeted pesticides in milk, apples, and drinking water samples each in triplicate. No pesticide was found in apple and milk samples, but chlorpyrifos was found in one drinking water sample below the quantification level.
The authentication of food products from the presence of non-allowed components for certain religion like lard is very important. In this study, we used proton Nuclear Magnetic Resonance ((1)H-NMR) spectroscopy for the analysis of butter adulterated with lard by simultaneously quantification of all proton bearing compounds, and consequently all relevant sample classes. Since the spectra obtained were too complex to be analyzed visually by the naked eyes, the classification of spectra was carried out.The multivariate calibration of partial least square (PLS) regression was used for modelling the relationship between actual value of lard and predicted value. The model yielded a highest regression coefficient (R(2)) of 0.998 and the lowest root mean square error calibration (RMSEC) of 0.0091% and root mean square error prediction (RMSEP) of 0.0090, respectively. Cross validation testing evaluates the predictive power of the model. PLS model was shown as good models as the intercept of R(2)Y and Q(2)Y were 0.0853 and -0.309, respectively.
Wider availability but lack of legal market trades has given feline meat a high potential for use as an adulterant in common meat and meat products. However, mixing of feline meat or its derivatives in food is a sensitive issue, since it is a taboo in most countries and prohibited in certain religions such as Islam and Judaism. Cat meat also has potential for contamination with of severe acute respiratory syndrome, anthrax and hepatitis, and its consumption might lead to an allergic reaction. We developed a very short-amplicon-length (69 bp) PCR assay, authenticated the amplified PCR products by AluI-restriction digestion followed by its separation and detection on a lab-on-a-chip-based automated electrophoretic system, and proved its superiority over the existing long-amplicon-based assays. Although it has been assumed that longer DNA targets are susceptible to breakdown under compromised states, scientific evidence for this hypothesis has been rarely documented. Strong evidence showed that shorter targets are more stable than the longer ones. We confirmed feline-specificity by cross-challenging the primers against 10 different species of terrestrial, aquatic and plant origins in the presence of a 141-bp site of an 18S rRNA gene as a universal eukaryotic control. RFLP analysis separated 43- and 26-bp fragments of AluI-digest in both the gel-image and electropherograms, confirming the original products. The tested detection limit was 0.01% (w/w) feline meat in binary and ternary admixed as well as meatball matrices. Shorter target, better stability and higher sensitivity mean such an assay would be valid for feline identification even in degraded specimens.
Being the third-largest primate population has not made macaque (Macaca fascicularis sp.) monkeys less exposed to threats and dangers. Despite wildlife protection, they have been widely hunted and consumed in several countries because of their purported nutritional values. In addition to trading as pure bush meats in several places, monkey meat has been sold in meatball and soup products in Indonesia. Thus the possibility of macaque meat trafficking under the label of common meats is quite high. This paper reports the development of a polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) assay with the shortest amplicon length for the confirmed detection of monkey meat under compromised states which are known to degrade DNA. We amplified a 120-bp region of d-loop gene using a pair of macaque-specific primers and confirmed their specificity for the target species through cross-challenging against 17 different species using a 141-bp site of an 18 S rRNA gene as an endogenous control for eukaryotes. This eliminated the possibilities of any false-negative detection with complex matrices or degraded specimens. The detection limit was 0.00001 ng DNA in a pure state and 0.1% of meat in mixed matrices and commercial meatball products. RFLP analysis further authenticated the originality of the PCR product and distinctive restriction patterns were found upon AluI and CViKI-1 digestion. A micro-fluidic lab-on-a-chip automated electrophoretic system separated the fragments with high resolution. The assay was validated for screening commercial meatball products with sufficient internal control.
An amperometric enzyme-electrode was introduced where glucose oxidase (GOD) was immobilized on chitosan membrane via crosslinking, and then fastened on a platinum working electrode. The immobilized enzyme showed relatively high retention activity. The activity of the immobilized enzyme was influenced by its loading, being suppressed when more than 0.6 mg enzyme was used in the immobilization. The biosensor showing the highest response to glucose utilized 0.21 ml/cm2 thick chitosan membrane. The optimum experimental conditions for the biosensors in analysing glucose dissolved in 0.1 M phosphate buffer (pH 6.0) were found to be 35°C and 0.6 V applied potential. The introduced biosensor reached a steady-state current at 60 s. The apparent Michaelis-Menten constant ([Formula: see text]) of the biosensor was 14.2350 mM, and its detection limit was 0.05 mM at s/n > 3, determined experimentally. The RSD of repeatability and reproducibility of the biosensor were 2.30% and 3.70%, respectively. The biosensor was showed good stability; it retained ~36% of initial activity after two months of investigation. The performance of the biosensors was evaluated by determining the glucose content in fruit homogenates. Their accuracy was compared to that of a commercial glucose assay kit. There was no significance different between two methods, indicating the introduced biosensor is reliable.