Preparation and characterization of intelligent and active bi-layer film based on carrageenan/pectin for monitoring freshness of salmon.

This work aimed to develop and characterize a novel bi-layer film (BIF) for monitoring the freshness of salmon. The indicator layer consists of carrageenan (Car), pectin (PEC) and purple sweet potato anthocyanin (PSPA), and the antibacterial layer consists of Car and magnolol (Mag). The results showed that the Car/Mag2 had the optimal water resistance: the static water contact angle of 80.36 ±â€¯0.92°, moisture content of 31.38 ±â€¯0.86 %, swelling degree of 92.96 ±â€¯0.46 %, and water solubility of 40.08 ±â€¯1.17 %, and showed excellent antibacterial properties against E. coli and S. aureus with antibacterial rate of 86.13 % ±â€¯0.10 % and 97.53 % ±â€¯0.02 %, respectively. Then BIFs with different PSPA concentration were tested. The morphology, mechanical and water vapor properties (WVP) of the BIFs were studied, and its application in salmon preservation was evaluated. The mechanical properties and WVP test results showed that the BIF0.2 had the optimal Tensile strength (TS) and WVP values. The BIFs showed distinguishable color changes between the pH ranges of 3-10. The shelf life of salmon packaged by BIF0.2 was prolonged by 2 days. Moreover, the BIF0.2 was able to effectively monitor salmon freshness. In conclusion, the BIF has great potential for monitoring salmon meat freshness.

Recent advancements in chitosan-based intelligent food freshness indicators: Categorization, advantages, and applications.

With an increasing emphasis on food safety and public health, there is an ongoing effort to develop reliable, non-invasive methods to assess the freshness of diverse food products. Chitosan-based food freshness indicators, leveraging properties such as biocompatibility, biodegradability, non-toxicity, and high stability, offer an innovative approach for real-time monitoring of food quality during storage and transportation. This review introduces intelligent food freshness indicators, specifically those utilizing pH-sensitive dyes like anthocyanins, curcumin, alizarin, shikonin, and betacyanin. It highlights the benefits of chitosan-based intelligent food freshness indicators, emphasizing improvements in barrier and mechanical properties, antibacterial activity, and composite film solubility. The application of these indicators in the food industry is then explored, alongside a concise overview of chitosan's limitations. The paper concludes by discussing the challenges and potential areas for future research in the development of intelligent food freshness indicators using chitosan. Thus, chitosan-based smart food preservation indicators represent an innovative approach to providing real-time data for monitoring food quality, offering valuable insights to both customers and retailers, and playing a pivotal role in advancing the food industry.

Deep Learning Used with a Colorimetric Sensor Array to Detect Indole for Nondestructive Monitoring of Shrimp Freshness.

Intelligent colorimetric freshness indicator is a low-cost way to intuitively monitor the freshness of fresh food. A colorimetric strip sensor array was prepared by p-dimethylaminocinnamaldehyde (PDL)-doped poly(vinyl alcohol) (PVA) and chitosan (Chit) for the quantitative analysis of indole, which is an indicator of shrimp freshness. As a result of indole simulation, the array strip turned from faint yellow to pink or mulberry color with the increasing indole concentration, like a progress bar. The indicator film exhibited excellent permeability, mechanical and thermal stability, and color responsiveness to indole, which was attributed to the interactions between PDL and Chit/PVA. Furthermore, the colorimetric strip sensor array provided a good relationship between the indole concentration and the color intensity within a range of 50-350 ppb. The pathogens and spoilage bacteria of shrimp possessed the ability to produce indole, which caused the color changes of the strip sensor array. In the shrimp freshness monitoring experiment, the color-changing progress of the strip sensor array was in agreement with the simulation and could distinguish the shrimp freshness levels. The image classification system based on deep learning were developed, the accuracies of four DCNN algorithms are above 90%, with VGG16 achieving the highest accuracy at 97.89%. Consequently, a "progress bar" strip sensor array has the potential to realize nondestructive, more precise, and commercially available food freshness monitoring using simple visual inspection and intelligent equipment identification.

Development of xanthan gum intelligent oil-in-water ink and its application in pork freshness preservation.

To optimize the stability of oil-based inks and ensure their wide application in freshness indication, new natural indicator inks were prepared using a stable oil-in-water structure. This study selected natural Lycium ruthenicum anthocyanin as the dye and glucose as the pigment carrier. Soybean oil was introduced as a linker and xanthan gum as a thickener, and an oil-in-water ink with the function of freshness indication was successfully developed. In ensuring the safety of ink labels for use on food packaging, particular attention is paid to the origin and properties of the materials used. All ingredients are of food-grade or bio-friendly provenance, thereby ensuring the safety of the product when in direct contact with food. We measured the viscosity, particle size and fineness of the ink for micro characterization and evaluated its macro printing performance by its printing effect on A4 paper. According to the experimental results, when the water-oil ratio of the ink is 10:5, the average particle size of the emulsion system is 822.83 nm, and the fineness reaches 5 µm. These values are relatively low, which indicates that the stability of the ink system is high at this time, and the ink shows excellent rheological and printing characteristics. With this water-to-oil ratio, the ink can show the best results when printed on A4 paper, clearly displaying image details. In addition, in fresh pork applications, inks with a 10: 5 water-to-oil ratio provide an accurate and highly sensitive indication of the freshness of pork. When the freshness of the pork changes, the ink color responds promptly. This high sensitivity makes the ink ideal for use as a food freshness indication tool, providing consumers with an intuitive and reliable reference for pork freshness. As a further innovation, combining this ink-printed label with a WeChat app not only allows consumers to know the freshness of the food in real-time but also tracks the supply chain information of the food, providing a more comprehensive application prospect for freshness-indicating products.

Development of a Sensitive Colorimetric Indicator for Detecting Beef Spoilage in Smart Packaging.

This study aimed to fabricate and characterize a novel colorimetric indicator designed to detect ammonia (NH3) and monitor meat freshness. The sensing platform was constructed using electrospun nanofibers made from polylactic acid (PLA), which were then impregnated with anthocyanins as a natural pH-sensitive dye, extracted from red cabbage. This research involved investigating the relationship between the various concentrations of anthocyanins and the colorimetric platform's efficiency when exposed to ammonia vapor. Scanning electron microscope (SEM) results were used to examine the morphology and structure of the nanofiber mats before and after the dip-coating process. The study also delved into the selectivity of the indicator when exposed to various volatile organic compounds (VOCs) and their stability under extreme humidity levels. Furthermore, the platform's sensitivity was evaluated as it encountered ammonia (NH3) in concentrations ranging from 1 to 100 ppm, with varying dye concentrations. The developed indicator demonstrated an exceptional detection limit of 1 ppm of MH3 within just 30 min, making it highly sensitive to subtle changes in gas concentration. The indicator proved effective in assessing meat freshness by detecting spoilage levels in beef over time. It reliably identified spoilage after 10 h and 7 days, corresponding to bacterial growth thresholds (107 CFU/mL), both at room temperature and in refrigerated environments, respectively. With its simple visual detection mechanism, the platform offered a straightforward and user-friendly solution for consumers and industry professionals alike to monitor packaged beef freshness, enhancing food safety and quality assurance.

Dual-Mode Ce-MOF Nanozymes for Rapid and Selective Detection of Hydrogen Sulfide in Aquatic Products.

Increasing concern over the safety of consumable products, particularly aquatic products, due to freshness issues, has become a pressing issue. Therefore, ensuring the quality and safety of aquatic products is paramount. To address this, a dual-mode colorimetric-fluorescence sensor utilizing Ce-MOF as a mimic peroxidase to detect H2S was developed. Ce-MOF was prepared by a conventional solvothermal synthesis method. Ce-MOF catalyzed the oxidation of 3,3',5,5'-tetramethylbenzidine (TMB) by hydrogen peroxide (H2O2) to produce blue oxidized TMB (oxTMB). When dissolved, hydrogen sulfide (H2S) was present in the solution, and it inhibited the catalytic effect of Ce-MOF and caused the color of the solution to fade from blue to colorless. This change provided an intuitive indication for the detection of H2S. Through steady-state dynamic analysis, the working mechanism of this sensor was elucidated. The sensor exhibited pronounced color changes from blue to colorless, accompanied by a shift in fluorescence from none to light blue. Additionally, UV-vis absorption demonstrated a linear correlation with the H2S concentration, ranging from 200 to 2300 µM, with high sensitivity (limit of detection, LOD = 0.262 µM). Fluorescence intensity also showed a linear correlation, ranging from 16 to 320 µM, with high selectivity and sensitivity (LOD = 0.156 µM). These results underscore the sensor's effectiveness in detecting H2S. Furthermore, the sensor enhanced the accuracy of H2S detection and fulfilled the requirements for assessing food freshness and safety.

The Impact of the Fresh Pork Display Lamps on the Sensory Response of Consumers to Fresh Pork.

Two studies were conducted to investigate the impact of fresh pork display lamps on consumers' sensory responses to pork products. In the first experiment, 63 participants were asked to evaluate pork products with different degrees of freshness under four fresh pork display lights and two LED lights. In the other experiment, 30 subjects participated in the Farnsworth-Munsell 100 hue test under the same lamps, with the aim of showing whether the fresh pork display lamps impaired color discrimination. The results showed that the light source had a significant effect on the color appearance evaluation of the pork products. The ratings for perceived freshness under the fresh pork display lamps were significantly higher than those of ordinary LED lamps, while the color discrimination performance of the subjects under those lamps was poor. It was demonstrated that improper component proportions of the light spectrum could influence observers' assessment of meat quality and weaken the observers' ability to distinguish the freshness level. Through this study, the authors would like to remind lamp users and manufacturers to not only be concerned about the improvement in the color appearance of pork, but also the need for consumers to be aware of the authentic freshness of the pork products.

Comprehensive Evaluation of the Physicochemical Attributes, Antioxidant Capacity, and pH-Responsive Behavior of Starch Films Enhanced by Laver Incorporation.

Herein, a new starch film incorporating laver was developed to address issues related to inadequate water resistance and suboptimal preservation quality in food packaging. The integration of laver into starch film formulations offers a compelling avenue for creating biodegradable, active, and smart food packaging. Scanning electron microscope (SEM) analysis revealed that the starch film with a laver concentration of 70% exhibited a uniformly flat microstructure, as expected. Fourier-transform infrared spectroscopy (FTIR) confirmed the presence of intermolecular interactions and hydrogen bonding between the starch and laver. Viscoelastic tests demonstrated the superior film-forming performance of the starch/laver composite films. Moreover, it was found that the most favorable concentration of incorporated laver was 10%. Specifically, the S7-3 film emerged as a promising candidate for food packaging applications, boasting the highest contact angle (CA) value of 114.98 ± 1.28°, the lowest water solubility (WS) value of 15.38%, and a reduced water vapor transmission rate (WVTR) value of 2.52 g/m2 × h. Additionally, the S3-7 film displayed an extraordinary tensile strength of 32.47 MPa, an elongation at break of 19.04%, and a Young's modulus of 606.83 MPa. Furthermore, the starch/laver composite films exhibited outstanding UV-blocking capabilities, exceptional pH-responsive behavior, and significant antioxidant activity, underscoring their potential for packaging applications with laver integration.

Enhancement of Colorimetric pH-Sensitive Film Incorporating Amomum tsao-ko Essential Oil as Antibacterial for Mantis Shrimp Spoilage Tracking and Fresh-Keeping.

Anthocyanin-based smart packaging has been widely used for food freshness monitoring, but it cannot meet the requirements of smart films with antibacterial properties. This study aimed to enhance the antibacterial properties of intelligent films by incorporating Amomum tsao-ko essential oil (AEO) for mantis shrimp spoilage tracking and keeping the product fresh. A smart film was designed by introducing AEO and purple potato anthocyanin (PPA) to a polyvinyl alcohol/cellulose nanocrystal (PVA/CNC) polymer matrix. Our findings revealed that APP and AEO imparted the smart film with a favorable oxygen barrier, UV protection, mechanical properties, and antioxidant and pH/NH3-sensitive functions. Interestingly, the PVA/CNC-AEO-PPA film achieved 45.41% and 48.25% bactericidal efficacy against S. putrefaciens and V. parahaemolyticus, respectively. Furthermore, a visual observation confirmed that the target film (PVA/CNC-AEO-PPA) changed color significantly during mantis shrimp spoilage: rose red-light red-pink-light gray-dark gray. Meanwhile, the PVA/CNC-AEO-PPA film retarded the quality deterioration of the mantis shrimp effectively. The PVA/CNC-AEO-PPA film shows great application potential in mantis shrimp preservation and freshness monitoring; it is expected to become a rapid sensor for detecting seafood quality non-destructively and a multifunctional film for better preservation of product quality.

Colorimetric indicator films based on carboxymethyl cellulose and anthocyanins as a visual indicator for shrimp freshness tracking.

This study aimed to evaluate the response efficiency of colorimetric indicator films based on carboxymethyl cellulose (CMC) incorporated with different anthocyanins [Karanda alone (CMC/AK), butterfly pea alone (CMC/AB), and a mixture of anthocyanins from Karanda and butterfly pea (CMC/AK75/AB25)] for tracking shrimp freshness during storage at different temperatures and times (4 °C for 8 days and 25 °C for 30 h). The mathematical models were also applied to predict their freshness and shelf life. The CMC/AK75/AB25 indicator film was the most sensitive and clearly changed color, which could be distinguished by the naked eye. Color changes indicated the shrimp deterioration processes: dark purple (fresh), purplish gray or gray (semi-fresh), and olive green or brown (spoilage). During shrimp storage at temperatures of 4 and 25 °C, the pH reached 7.52 and 8.14, TVB-N 35.98 and 72.72 mg/100 g, and TVC 5.75 and 7.88 log CFU/g, respectively, indicating shrimp had completely deteriorated. Furthermore, there was a positive correlation between the ΔE value of the indicator film and both TVB-N and TVC. These findings suggest that the CMC/AK75/AB25 indicator film could serve as a real-time visual indicator for tracking shrimp freshness and could enhance the guarantee of shrimp safety.

Ammonia/pH super-sensitive colorimetric labels based on gellan gum, sodium carboxymethyl cellulose, and dyes for monitoring freshness of lamb meat.

This study aimed to develop an ammonia and pH super-sensitive label by incorporating methyl red and bromothymol blue (MR-BTB, MB) into gellan gum/sodium carboxymethyl cellulose (GG/CMC-Na, GC). Furthermore, E-nose as an auxiliary tool combined with the labels to monitor meat freshness. Results showed that MB had more color change than pure MR or BTB, and the detection limit of ammonia about the MR-BTB (1:2) group was only 2.82 ppm. The addition of MB significantly increased tensile strength, moisture content, and water solubility, but decreased elongation at break and transmittance of the GC label (p < 0.05). The result of FTIR and SEM indicated the formation of hydrogen bonds and well compatibility between MB and GC. Furthermore, the color of the GC-10.0MB label was constantly obviously changing during meat storage, indicating that the GC-10.0MB label had great potential for monitoring the freshness of the lamb meat. A high correlation was found between ΔE of GC-10.0MB label and TVB-N (R2 = 0.9092) and pH (R2 = 0.9114) of meat. Interestingly, the high correlation between ΔE of GC-10.0 MB label and the response value of S2 (R2 = 0.7531), S6 (R2 = 0.9921), and S7 sensor (R2 = 0.8325) of E-nose was also found.

A smartphone-adaptable fluorescent probe for visual monitoring of fish freshness and its application in fluorescent dyes.

Due to increasing food safety issues, developing intelligent, on-site, and visual methods for detecting fish freshness has attracted significant attention. Here, we have prepared a benzo[h]chromene derivative BCN that can visually detect 12 biogenic amines (BAs) with high sensitivity. The mechanism for recognizing cadaverine (Cad) is that the probe reacts with Cad to produce a Schiff base derivative, which alters the charge distribution within the molecule, resulting in significant colorimetric and fluorescence changes. The sensing label BCN/FPS was prepared by loading the probe BCN on filter paper, and a visual detection platform was constructed by combining it with a smartphone. By monitoring the correspondence between label color and TVB-N content, a working curve of (R + B)/(R + B + G) with TVB-N content was obtained, enabling visual evaluation of salmon freshness using only a mobile phone. In addition, based on the good solubility and processability of BCN, its application in fluorescent dyes including impregnating dyes, printing inks, coatings, and flexible films has been explored, which opens up new directions for the application of BCN. Therefore, BCN has the potential for real-time monitoring of meat freshness and preparation of fluorescent materials.

Active and smart biomass film with curcumin Pickering emulsion stabilized by chitosan-adsorbed laurate esterified starch for meat freshness monitoring.

The multifunctional active smart biomass film was prepared by incorporating chitosan-adsorbed laurate esterified starch curcumin Pickering emulsion into the starch film matrix, with nano-cellulose serving as reinforcing agents. The mechanical and functional properties of the film were studied, and the film was used to monitor the freshness of pork. The results demonstrated a relatively uniform distribution of curcumin and Pickering emulsion droplets within the film matrix. Furthermore, the thermal stability was minimally impacted by the introduction of curcumin Pickering emulsion, while the tensile strength and tensile strain of the film were increased, and both its hydrophobicity and antioxidant properties were improved. The free radical scavenging rate reached 56.01 %, with sustained high antioxidant capacity even after 8 days. Additionally, the presence of curcumin provided the film with pH indicating ability and delayed pork spoilage. Therefore, this work provides an attractive strategy for constructing green, active, and smart biomass packaging films for meat packaging applications.

Carrageenan-based functional films hybridized with carbon dots and anthocyanins from rose petals for smart food packaging applications.

Multifunctional smart biopolymeric films were fabricated using rose petal anthocyanin (RPA) and carrageenan (CAR) doped with rose petal-derived carbon dots (RP-CDs). Response surface-optimized RPA showed the highest total anthocyanins and radical scavenging ability. Produced RP-CD exhibited UV absorption and high fluorescence with antibacterial/antioxidant abilities. Enrichment with 2 % RP-CD and 5 % RPA in the CAR matrix results in improved physicochemical, i.e., water contact angle, water vapor permeability, and UV-blocking properties of the fabricated material. Results showed that nanocomposite films scavenged radicals better than the neat CAR films. Zeta potential, FTIR, SEM, and XPS suggested improved compatibility/stability and enhanced elemental configuration of RP-CDs/RPA additives in the CAR polymer matrix. Perishable food packaging (minced pork and shrimp) demonstrated that nanocomposite films work efficiently and non-destructively and are promising tools for monitoring real-time freshness through interpretable visual changes from red to yellow. The CAR/RP-CDs/RPA-based nanocomposite indicator films are expected to be applied as various smart packaging materials. These films possess the ability to promptly detect changes in quality, preserve the quality, and prolong the shelf life of packaged foods.

Potato oxidized hydroxypropyl starch/pectin-based indicator film with Clitoria ternatea anthocyanin and silver nanoparticles for monitoring chilled beef freshness.

Potato oxidized hydroxypropyl starch (POHS)/pectin (P) functional and smart beef freshness indicator films were prepared using butterfly pea (Clitoria ternatea) anthocyanin (BA) and silver nanoparticles (AgNPs). BA exhibited significant pH-responsive color changes. BA and AgNPs were evenly distributed within a polymer matrix to create a compatible film with POHS/P. The films containing BA and AgNPs had good UV resistance and maintained strong mechanical strength, barrier properties, and color stability. The color of the indicator film changed from purple to green when exposed to ammonia, with the 1 % POHS/P/BA/AgNPs film showing the most sensitive response. The films also demonstrated strong antibacterial and antioxidant properties. The freshness of beef was monitored using 1 % POHS/P/BA/AgNPs films and was identified as sub-fresh and spoiled on days 4 and 7, respectively. The relationship between the color change of the indicator label and the freshness of chilled beef was established: purple for fresh meat, blue for less fresh meat, and green for spoiled meat. Thus, the new POHS/P/BA/AgNPs film can serve as a smart packaging material to indicate food freshness and extend shelf life. These results suggest that POHS/P/BA/AgNPs films have significant potential as an active and smart food packaging material.

Effect of metal cation crosslinking on the mechanical properties and shrimp freshness monitoring sensitivity of pectin/carboxymethyl cellulose sodium/anthocyanin intelligent films.

Although many preparation methods have been reported so far, it is still a great challenge for intelligent packaging films with both excellent mechanical properties and very high sensitivity. Herein, we report a facile method to prepare performance-enhanced pectin (PC)/carboxymethyl cellulose sodium (CMC)/anthocyanins (ACNs)/metal ion films by crosslinking with metal ions (Zn2+, Mg2+ and Ca2+). Cross-linking reaction between PC/CMC and metal ions significantly improved water resistance and mechanical properties of composite films (P < 0.05). Even at high relative humidity (RH = 84 %), cross-linking of Ca2+, Mg2+, and Zn2+ significantly increased the tensile index of the films by 1.37, 1.41, and 1.52 times (P < 0.05), respectively. Moreover, the complexation of metal ions/polysaccharides with ACNs reduced the decomposition rate of ACNs, improved the storage stability and antioxidant capacity of ACNs, and also increased the sensitivity of the colorimetric response of the indicator films in monitoring shrimp freshness. Thus, with this high sensitivity, the Red, Green and Blue (RGB) values of the films can be determined using a mobile phone application to monitor shrimp safety in real time. These results suggest that ACNs-metal cation-polysaccharide composite films have great potential for smart packaging applications.

Designing cellulose nanofibrils/carbon dots intelligent label with colorimetric and fluorescent dual responsiveness for real-time monitoring of food freshness.

This study utilized 1,2,4-triaminobenzene dihydrochloride and NaOH as precursors to prepare the pH optical sensor based on carbon dots (CDs). By incorporating CDs into pineapple peel cellulose nanofibrils (CNF) matrix, an intelligent label (CNF/CDs label) with colorimetric and fluorescent dual responsiveness was created for real-time monitoring of food freshness. The CNF/CDs labels exhibited remarkable sensitivity and recognizability towards pH changes from 1 to 12. They also demonstrated excellent reversibility during acid-alkali cycling. Moreover, these labels exhibited exceptional responsiveness to the alkaline and acidic gas environments formed by ammonium hydroxide and acetic acid solutions, respectively. These responses were visually distinguishable through visible color changes and ultraviolet (UV) fluorescence alterations. Encouragingly, the developed labels were successfully applied to monitor the freshness of prawns and fruits, enabling timely assessment of food freshness levels. The dual-mode response of color and fluorescence provided double assurance for the accuracy of the results.

Novel intelligent packaging films based on starch/PVA with Cu-ICA nanocrystal as functional compatibilizer for monitoring food freshness.

Considering public health and environmental safety, the development of reliable and efficient monitoring methods is essential to ensure food quality and safety. Herein, a new Cu-based metal organic framework (Cu-ICA) nanocrystal with ammonia-sensitive performance was built up and then introduced as a functional compatibilizer of starch/polyvinyl alcohol (STA/PVA) blend to develop high-performance intelligent packaging films for food freshness monitoring. The introduction of Cu-ICA upgraded the compatibility, mechanical strength (42.9 MPa), UV-protection (with UV transmittance of only 2.8 %), and moisture/oxygen barrier performances of STA/PVA film. Furthermore, the developed STA/PVA/Cu-ICA films presented long-term colour stability, outstanding antibacterial efficacy (over 99.5 %) toward both Escherichia coli and Staphylococcus aureus bacteria, as well as remarkable ammonia-sensitive discoloration capability. The STA/PVA/Cu-ICA films possessed visually identifiable colour change during the monitoring of shrimp spoilage. These findings indicate that the developed STA/PVA/Cu-ICA film possesses tremendous potential as an intelligent active packaging material.

High-performance carboxymethyl starch/PVA based intelligent packaging films engineered with Cu-Trp nanocrystal as functional compatibilizer.

A spindle-like Cu-based framework (Cu-Trp, Trp = L-Tryptophan) nanocrystal with ammonia-responsiveness was fabricated via simple aqueous solution approach, and it was subsequently explored as a functional compatibilizer of carboxymethyl starch/polyvinyl alcohol (CMS/PVA) blend toward constructing high-performance intelligent packaging films. The results showed that incorporation of Cu-Trp nanocrystal into CMS/PVA blend resulted in significant promotions regarding to the compatibility, mechanical strength (42.92 MPa), UV-blocking (with UV transmittance of only 2.4%), and water vapor barrier effectiveness of the blend film. Besides, the constructed CMS/PVA/Cu-Trp nanocomposite film exhibited superb long-term color stability, favorable antibacterial capacity (over 98.0%) toward both E. coli and S. aureus bacteria, as well as color change ability under ammonia environment. Importantly, the application trial confirmed that the CMS/PVA/Cu-Trp nanocomposite film is capable of visually monitoring shrimp spoilage during storage. These results implied that the CMS/PVA/Cu-Trp nanocomposite film holds tremendous potential as an intelligent active packaging material.

Deep learning models with optimized fluorescence spectroscopy to advance freshness of rainbow trout predicting under nonisothermal storage conditions.

This study established long short-term memory (LSTM), convolution neural network long short-term memory (CNN_LSTM), and radial basis function neural network (RBFNN) based on optimized excitation-emission matrix (EEM) from fish eye fluid to predict freshness changes of rainbow trout under nonisothermal storage conditions. The method of residual analysis, core consistency diagnostics, and split-half analysis of parallel factor analysis was used to optimize EEM data, and two characteristic components were extracted. LSTM, CNN_LSTM, and RBFNN models based on characteristic components of EEM used to predict the freshness indices. The results demonstrated the relative errors of RBFNN models with an R2 above 0.96 and relative errors less than 10% for K-value, total viable counts, and volatile base nitrogen, which were better than those of LSTM and CNN_LSTM models. This study presents a novel approach for predicting the freshness of rainbow trout under nonisothermal storage conditions.