Total Viable Count Prediction Research Articles

Hyperspectral imaging combined with fluorescence for the prediction of microbial growth in chicken breasts under different packaging conditions

The crucial feature of chicken is real-time microbial growth detection along the supply chain. Hyperspectral imaging (HSI) is an online and nondestructive technique that is well suited to microorganism analysis applications. In this study, visible and near-infrared (Vis/NIR, 400–1000 nm) HSI were utilized to establish prediction models of the total viable count (TVC) of chicken breasts stored in vacuum, pallet, and combined packaging using partial least squares regression (PLSR) and support vector machine (SVM) after spectral pretreatments and optimal wavelength selection. The SVM algorithm was used to develop the best TVC prediction models for storage, with the following outcome: vacuum (R2p = 0.91, RMSEp = 0.47, RPD = 2.44), pallet (R2p = 0.90, RMSEp = 0.54, RPD = 2.32), and combined (R2p = 0.89, RMSEp = 0.58, RPD = 2.21) packs. Successively, HSI combined with fluorescence (F–HSI) was used to predict the TVC and coliforms of chicken breasts in combined packaging. Compared to HSI, F–HSI achieved the best-predicted results for TVC (R2p = 0.93, RMSEp = 0.44, RPD = 2.74) and coliforms (R2p = 0.92, RMSEp = 0.46, RPD = 2.57). These results indicated that fluorescence, as a complementary technique for HSI, will be widely used in the chicken supply chain to predict microbial growth.

Determination of the total viable count of Chinese meat dishes by near‐infrared spectroscopy: A predictive model

The total viable count (TVC) is an important index that reflects the freshness, contamination, and spoilage of dishes. Presently, near-infrared spectroscopy (NIR) is widely used for the assessment of TVC in raw meat, but there are few reports on the application of NIR spectroscopy to evaluate the TVC of Chinese meat dishes. Also, in the existing research, the modeling samples are industrial products, which makes it lack extensive representation, resulting in insufficient generalization ability of the model. Therefore, we prepared a relatively complex modeling sample to improve the application ability of the model. In this research, braised beef was investigated as an example to establish a predictive model of TVC in complex food matrices using NIR spectroscopy. For this purpose, five groups of 472 braised beef samples with different seasonings were used to collect NIR spectra (900–1,700 nm). For five groups, the periods of storage were tested using spectroscopy every day, after which the TVC was measured directly. Furthermore, we established a PLS prediction model. Later, the model established by using the best pre-processed spectral data for band selection, and the model established by band selection followed by variable processing methods were compared. The model using SPA band selection method followed by median center processing exhibited a good prediction ability, with the coefficient of correlation (R) of .8507, root mean squared error of prediction (RMSEP) of .2707 lg CFU/g, ratio of prediction to deviation (RPD) of 2.00, and ratio error range (RER) of 9.46. These results prove that the developed model can be used in the prediction of TVC in Chinese braised beef. Novelty impact statement Took Chinese braised beef as an example, this research has used NIR spectroscopy to predict the TVC in meat dishes that contained a range of seasonings and were stored for various periods. The model using the SPA band selection method followed by median center pre-processing exhibited a good prediction ability. The developed model can be used in the prediction of TVC in Chinese braised beef and to make a preliminary judgment on the edibility of Chinese dishes.

Visible and NIR hyperspectral imaging and chemometrics for prediction of microbial quality of beef Longissimus dorsi muscle under simulated normal and abuse storage conditions

There is a need to develop a rapid technique to provide real time information on the microbial load of meat along the supply chain. Hyperspectral imaging (HSI) is a rapid, non-destructive technique well suited to food analysis applications. In this study, HSI in both the visible and near infrared spectral ranges, and chemometrics were studied for prediction of the bacterial growth on beef Longissimus dorsi muscle (LD) under simulated normal (4 °C) and abuse (10 °C) storage conditions. Total viable count (TVC) prediction models were developed using partial least squares regression (PLS-R), spectral pre-treatments, band selection and data fusion methods. The best TVC prediction models developed for storage at 4 (RMSEp 0.58 log CFU/g, RPDp 4.13, R2p 0.96), 10 °C (RMSEp 0.97 log CFU/g, RPDp 3.28, R2p 0.94) or at either 4 or 10 °C (RMSEp 0.89 log CFU/g, RPDp 2.27, R2p 0.86) were developed using high-level data fusion of both spectral regions. The use of appropriate spectral pre-treatments and band selection methods was key for robust model development. This study demonstrated the potential of HSI and chemometrics for real time monitoring to predict microbial growth on LD along the meat supply chain.

LW-NIR hyperspectral imaging for rapid prediction of TVC in chicken flesh

Total viable count (TVC) is often used as an important indicator for chicken freshness evaluation. In this study, 112 fresh chicken flesh samples were acquired after slaughtered and their hyperspectral images were collected in the LW-NIR (900-1700 nm) range. The full LW-NIR spectra (486 wavebands) within the images were extracted and applied to related to reference TVC values measured in different storage period, using partial least squares regression (PLSR) algorithm, resulting in high correlation coefficients (R) and low root mean square errors (RMSE), for either raw spectra or pretreatment spectra. By using regression coefficients (RC) method, 20, 18, 17 and 20 optimal wavebands were respectively selected from raw spectra, baseline correction (BC) spectra, Savitzky-Golay convolution smoothing (SGCS) spectra and standard normal variate (SNV) spectra and applied for the optimization of original full waveband PLSR model. By comparison, RC-PLSR model based on the SGCS spectra showed a better performance in TVC prediction with RC of 0.98 and RMSEC of 0.35 log10 CFU/g in calibration set, and RP of 0.98 and RMSEP of 0.44 log10 CFU/g in prediction set. At last, by transferring the best RC-PLSR model, the dynamic TVC change during the storage was visualized by color maps to indicate the TVC spoilage degree. The overall study revealed that LW-NIR hyperspectral imaging combined with PLSR could be used to predict the freshness of chicken flesh. Keywords: hyperspectral imaging, chicken, TVC, partial least square regression (PLSR) DOI: 10.25165/j.ijabe.20191203.4444 Citation: H Wang, H J He, H J Ma, F S Chen, Z L Kang, M M Zhu, et al. LW-NIR hyperspectral imaging for rapid prediction of TVC in chicken flesh. Int J Agric & Biol Eng, 2019; 12(3): 180–186.

Using deep learning and hyperspectral imaging to predict total viable count (TVC) in peeled Pacific white shrimp

In this study, deep learning method coupled with near-infrared (NIR) hyperspectral imaging (HSI) technique was used for nondestructively determining total viable count (TVC) of peeled Pacific white shrimp. Firstly, stacked auto-encoders (SAE) was conducted as a big data analytical method to extract 20 deep hyperspectral features from NIR hyperspectral image (900–1700 nm) of peeled shrimp stored at 4 °C, and the extracted features were used to predict TVC by fully-connected neural network (FNN). The SAE–FNN method obtained high prediction accuracy for determining TVC, with R P 2 = 0.927. Additionally, TVC spatial distribution of peeled shrimp during storage could be visualized via applying the established SAE–FNN model. The results demonstrate that SAE–FNN combined with HSI technique has a potential for non-destructive prediction of TVC in peeled shrimp, which supply a novel method for the hygienic quality and safety inspections of shrimp product.

LW-NIR hyperspectral imaging for rapid prediction of TVC in chicken flesh

Total viable count (TVC) is often used as an important indicator for chicken freshness evaluation. In this study, 112 fresh chicken flesh samples were acquired after slaughtered and their hyperspectral images were collected in the LW-NIR (900-1700 nm) range. The full LW-NIR spectra (486 wavebands) within the images were extracted and applied to related to reference TVC values measured in different storage period, using partial least squares regression (PLSR) algorithm, resulting in high correlation coefficients (R) and low root mean square errors (RMSE), for either raw spectra or pretreatment spectra. By using regression coefficients (RC) method, 20, 18, 17 and 20 optimal wavebands were respectively selected from raw spectra, baseline correction (BC) spectra, Savitzky-Golay convolution smoothing (SGCS) spectra and standard normal variate (SNV) spectra and applied for the optimization of original full waveband PLSR model. By comparison, RC-PLSR model based on the SGCS spectra showed a better performance in TVC prediction with RC of 0.98 and RMSEC of 0.35 log10 CFU/g in calibration set, and RP of 0.98 and RMSEP of 0.44 log10 CFU/g in prediction set. At last, by transferring the best RC-PLSR model, the dynamic TVC change during the storage was visualized by color maps to indicate the TVC spoilage degree. The overall study revealed that LW-NIR hyperspectral imaging combined with PLSR could be used to predict the freshness of chicken flesh. Keywords: hyperspectral imaging, chicken, TVC, partial least square regression (PLSR) DOI: 10.25165/j.ijabe.20191203.4444 Citation: H Wang, H J He, H J Ma, F S Chen, Z L Kang, M M Zhu, et al. LW-NIR hyperspectral imaging for rapid prediction of TVC in chicken flesh. Int J Agric & Biol Eng, 2019; 12(3): 180–186.

A Nondestructive Real-Time Detection Method of Total Viable Count in Pork by Hyperspectral Imaging Technique

A nondestructive method was developed for assessing total viable count (TVC) in pork during refrigerated storage by using hyperspectral imaging technique in this study. The hyperspectral images in the visible/near-infrared (VIS/NIR) region of 400–1100 nm were acquired for fifty pork samples, and their VIS/NIR diffuse reflectance spectra were extracted from the images. The reference values of TVC in pork samples were determined by classical microbiological plating method. Both partial least square regression (PLSR) model and support vector machine regression model (SVR) of TVC were built for comparative analysis to achieve better results. Different transformation methods and filtering methods were applied to improve the models. The results show that both the optimized PLSR model and SVR model can predict the TVC very well, while the SVR model based on second derivation was better, which achieved with RP (correlation coefficient of prediction set) = 0.94 and SEP (standard error of prediction set) = 0.4570 log CFU/g in the prediction set. An image processing algorithm was then developed to transfer the prediction model to every pixel of the image of the entire sample; the visualizing map of TVC would be displayed in real-time during the detection process due to the simplicity of the model. The results demonstrated that hyperspectral imaging is a potential reliable approach for non-destructive and real-time prediction of TVC in pork.

A comparative study for improving prediction of total viable count in beef based on hyperspectral scattering characteristics

The objectives of this study were to investigate the feasibility of hyperspectral scattering imaging to predict the bacterial contamination in meat nondestructively, and propose an optimal approach for detecting low levels of total viable count (TVC) contamination in beef. Fresh beef samples were obtained from a commercial slaughtering plant, and stored at 4°C for 0–12days. The visible/near-infrared (VIS/NIR) hyperspectral images in the backscattering mode were acquired from 3–5 beef samples on each day of the experiment, in parallel with microbiological analysis to enumerate the TVC population. Lorentzian function was used to resolve the light scattering information within the hyperspectral image and consequently Lorentzian parameters, which represented different hyperspectral scattering characteristics were extracted. In this study, not only the individual Lorentzian parameters but also the parameter combinations were used to establish the multivariate statistical models for predicting beef TVC, based on the modeling methods of principal component regression (PCR), partial least squares regression (PLSR), and back propagation neural network (BPNN), respectively. The models established using individual Lorentzian parameters did not perform well in predicting low levels of TVC contamination in beef, and the best prediction result could only achieved with the correlation coefficient of prediction set (RP) and root mean squared error of prediction set (RMSEP) of 0.81 and 1.27 log CFU/g, respectively. Based on the parameter combinations, the best modeling results were achieved with RP and RMSEP of 0.86 and 0.93 log CFU/g, 0.87 and 0.79 log CFU/g, 0.90 and 0.88 log CFU/g by PCR, PLSR, and BPNN methods, respectively, which confirmed the superiority of the parameter combination method. The results of this study demonstrated for the first time that hyperspectral scattering imaging combined with Lorentzian function and the proposed parameter combination method could be used to detect low levels of bacterial contamination in beef nondestructively.

A nondestructive method for fish freshness determination with electronic tongue combined with linear and non-linear multivariate algorithms

Electronic tongue coupled with linear and non-linear multivariate algorithms was attempted to address the drawbacks of fish freshness detection. Parabramis pekinensis fish samples stored at 4°C were used. Total volatile basic nitrogen (TVB-N) and total viable count (TVC) of the samples were measured. Fisher liner discriminant analysis (Fisher LDA) and support vector machine (SVM) were applied comparatively to classify the samples stored at different days. The results revealed that SVM model was better than Fisher LDA model with a higher identification rate of 97.22% in the prediction set. Partial least square (PLS) and support vector regression (SVR) were applied comparatively to predict the TVB-N and TVC values. The quantitative models were evaluated by the root mean square error of prediction (RMSEP) and the correlation coefficient in the prediction set (R<sub>pre</sub>). The results revealed that SVR model was superior to PLS model with RMSEP = 5.65 mg/100 g, R<sub>pre</sub> = 0.9491 for TVB-N prediction and RMSEP = 0.73 log CFU/g, R<sub>pre</sub> = 0.904 for TVC prediction. This study demonstrated that the electronic tongue together with SVM and SVR has a great potential for a convenient and nondestructive detection of fish freshness.  

A Nondestructive Method for Prediction of Total Viable Count in Pork Meat by Hyperspectral Scattering Imaging

Current methods for detecting the bacterial contamination of meat are time-consuming, labor-intensive, and giving retrospective information; therefore, the objective of this study was to investigate the feasibility of hyperspectral scattering imaging for rapid and nondestructive determination of total viable count (TVC) in pork meat. Fresh pork meat was purchased from a local market and stored at 10 °C for 1–15 days. In total, 59 samples were used in this study, and three to four samples were taken out randomly for hyperspectral scattering imaging and conventional microbiological analysis on each day of the experiment. Both the Lorentzian function and the Gompertz function were exploited to interpret the scattering profiles of pork meat samples, and good fitting results were obtained between 472 and 1,000 nm. Stepwise multiple linear regression (SMLR) method was performed to establish the prediction models, and moving average method with the filter size ranging from 3-point to 15-point was applied to improve the modeling results, respectively. Among the models established, the models developed by the Lorentzian parameter b and the Gompertz parameter β performed best for predicting pork meat TVC, with the correlation coefficient of validation set (Rv) of 0.94 and 0.93, respectively, after 13-point and 11-point moving average. The Lorentzian parameter a and the Gompertz parameters α and δ can also give good prediction results, with Rv of 0.83, 0.88, and 0.82, respectively. The results demonstrated that hyperspectral scattering imaging combined with the Lorentzian function and the Gompertz function can be a powerful tool for evaluating the microbial safety of meat in the future.

Rapid detection of total viable count (TVC) in pork meat by hyperspectral imaging

Total viable count (TVC) of bacteria is one of the most important indexes in evaluation of quality and safety of meat. In this work, the TVC in pork meat was detected by hyperspectral imaging technology. First, the spectra were extracted from 3-D datacube of hyperspectral image and 100 characteristic variables were selected by synergy interval PLS (SI-PLS) algorithm. Meanwhile, principal component analysis (PCA) was implemented on the 3-D datacube to determine 3 characteristic pictures. And, 5 characteristic variables were extracted using texture analysis from each characteristic picture. PCA was implemented on 111 spectra variables, 15 image variables and data fusion (126 variables), and the top principal components (PCs) were extracted for developing the TVC prediction model, respectively. Experimental results show that the model based on data fusion is superior to others, which was achieved with RMSEP=0.243lgCFU/g and Rp2=0.8308 in the prediction set. This work demonstrates that HSI technique, as a nondestructive analytical tool, has the potential in nondestructive detection of TVC in pork meat.

Potential of time series-hyperspectral imaging (TS-HSI) for non-invasive determination of microbial spoilage of salmon flesh

This study investigated the potential of using time series-hyperspectral imaging (TS-HSI) in visible and near infrared region (400–1700nm) for rapid and non-invasive determination of surface total viable count (TVC) of salmon flesh during spoilage process. Hyperspectral cubes were acquired at different spoilage stages for salmon chops and their spectral data were extracted. The reference TVC values of the same samples were measured using standard plate count method and then calibrated with their corresponding spectral data based on two calibration methods of partial least square regression (PLSR) and least-squares support vector machines (LS-SVM), respectively. Competitive adaptive reweighted sampling (CARS) was conducted to identify the most important wavelengths/variables that had the greatest influence on the TVC prediction throughout the whole wavelength range. As a result, eight variables representing the wavelengths of 495nm, 535nm, 550nm, 585nm, 625nm, 660nm, 785nm, and 915nm were selected, which were used to reduce the high dimensionality of the hyperspectral data. On the basis of the selected variables, the models of PLSR and LS-SVM were established and their performances were compared. The CARS-PLSR model established using Spectral Set I (400–1000nm) was considered to be the best for the TVC determination of salmon flesh. The model led to a coefficient of determination (rP2) of 0.985 and residual predictive deviation (RPD) of 5.127. At last, the best model was used to predict the TVC values of each pixel within the ROI of salmon chops for visualizing the TVC distribution of salmon flesh. The research demonstrated that TS-HSI technique has a potential for rapid and non-destructive determination of bacterial spoilage in salmon flesh during the spoilage process.

Potential prediction of the microbial spoilage of beef using spatially resolved hyperspectral scattering profiles

Spoilage in beef is the result of decomposition and the formation of metabolites caused by the growth and enzymatic activity of microorganisms. There is still no technology for the rapid, accurate and non-destructive detection of bacterially spoiled or contaminated beef. In this study, hyperspectral imaging technique was exploited to measure biochemical changes within the fresh beef. Fresh beef rump steaks were purchased from a commercial plant, and left to spoil in refrigerator at 8 °C. Every 12 h, hyperspectral scattering profiles over the spectral region between 400 and 1100 nm were collected directly from the sample surface in reflection pattern in order to develop an optimal model for prediction of the beef spoilage, in parallel the total viable count (TVC) per gram of beef were obtained by classical microbiological plating methods. The spectral scattering profiles at individual wavelengths were fitted accurately by a two-parameter Lorentzian distribution function. TVC prediction models were developed, using multi-linear regression, on relating individual Lorentzian parameters and their combinations at different wavelengths to log 10(TVC) value. The best predictions were obtained with r 2 = 0.95 and SEP = 0.30 for log 10(TVC). The research demonstrated that hyperspectral imaging technique showed potential for real-time and non-destructive detection of bacterial spoilage in beef.