Stunning pigs with inert gases has reportedly led to differences in pork quality. In the present study, we investigated how different inert gas atmospheres affect the muscle transcriptome, aiming to identify changes in RNA expression that could explain potential differences in meat quality. Therefore, total RNA was extracted from 120 slaughter-weight pigs, which were stunned pairwise either with argon, a nitrogen-argon mixture, or carbon dioxide. To control for potential slaughter day effects, 2 CO2 control groups were included, resulting in 30 animals per stunning condition. Muscle samples from the M. longissimus thoracis et lumborum were collected at 45 min and 36 h postmortem, respectively. For each stunning method and time point, sequencing was performed on 3 pooled samples, each comprising 10 animals. The comparison of the muscle transcriptomes revealed 112 genes to be differentially expressed (absolute Log2Foldchange >1, P adjusted < 0.01) between 45 min and 36 h postmortem across all gas comparisons. Gene set enrichment analysis revealed them to be involved in pathways like the cytoskeleton in muscle cells and protein digestion and absorption. Out of these genes, 24, including Protein phosphatase-1 regulatory subunit 3A (PPP1R3A), were found to be upregulated at 36 h postmortem. When comparing the effects of the different stunning gases, a total of 26 genes were differentially expressed (absolute Log2Foldchange >1, P adjusted < 0.05), although this was not significant in all comparisons. This study is the first to characterize the transcriptome of the M. longissimus thoracis et lumborum in pigs depending on the gas used for stunning. Moreover, we were able to identify distinct transcriptomic profiles at different postmortem time points, providing new insights into the transcriptomic changes occurring in porcine muscle tissue after slaughter.

Traditional meat processing technologies and methods are predominantly manual and labor intensive, but they can be significantly optimized through automation and advanced data-driven systems. Artificial intelligence (AI) and internet of things technologies enable noninvasive, automated, and real-time solutions that enhance efficiency, safety, and consistency, while also reducing labor demands. These capabilities mark an inflection point in meat processing, with AI-driven solutions spanning every stage of the livestock sector, from meat production to quality assessment and market analysis. This review comprehensively explores existing research throughout the meat processing cycle, with a specific focus on data- driven AI applications that perform classification, regression, and image analysis tasks. The analysis emphasizes the types of data collected, the preprocessing strategies employed, and the AI models adopted. It also identifies key challenges, emerging trends, and potential pathways for future development, specifically highlighting opportunities to improve efficiency, safety, and sustainability. The insights presented herein offer valuable guidance for researchers and industry professionals seeking to advance meat processing technologies through AI-driven innovation.

The objective of this study was to develop predictive models for consumer purchase likelihood of beef strip loin steaks based on measures of redness and discoloration. In Phase 1, consumers evaluated steaks from various days of display (0–14 d), ranging in discoloration from 0% to 100%. Phase 2 required consumers to evaluate samples all from the same day of display. For each phase, steaks were removed from their mother-bag and placed in coffin-style retail cases under fluorescent lights for the pre-planned display period (0–14 d) to create differences in both steak color and discoloration. Consumers (N = 200 Phase 1; 176 Phase 2) evaluated steaks for overall appearance liking and whether or not they would purchase the product (yes/no) if it was fully-priced or if it was sold at a discount. Objective color measures (L*, a*, b*, chroma, hue angle), overall redness, % discoloration, and fat color were assessed by a trained descriptive sensory panel. Results showed that all of the objective measures evaluated were predictors (P < 0.05) of consumer purchase intent. The logistic predictive models accounted for 30–78% of the variation in consumer purchase intent for steaks sold at full price and 27–72% of the variation when sold at a discount for Phase 1. In Phase 2, the models were less predictive, with models only accounting for 11–57% of the variation in consumer purchase likelihood in both full-priced and discounted samples. a* and discoloration scores were some of the best predictors evaluated. These models indicated that even minimal discoloration has a large impact on consumer purchasing likelihood, with steaks having 12 to 22% discoloration only having a 50% chance of being purchased at full price. These results underscore the importance of ensuring steaks sold at retail are maintained at a bright-red color and free of discoloration.

The objective of this study was to evaluate the effect of supplementing α-tocopheryl-acetate (ATA) and γ-tocopherol (GT) vitamin E isoforms with corn oil (CO) and tallow (TW) on carcass characteristics and meat quality characteristics of pigs grown to heavy weights (150 kg). Individually fed pigs (n = 72; 36 barrows, 36 gilts) were randomly assigned to 12 dietary treatments in a 2 × 6 factorial arrangement. Fat treatments were 5% TW and 5% CO. The vitamin E treatments included 4 levels of ATA (11, 40, 100, and 200 ppm) and 2 levels of mixed tocopherols (primarily GT; 40 and 100 ppm). Pigs were humanely slaughtered at approximately 150 kg. Carcass characteristics, pH, primal cuts, and meat quality attributes were evaluated. Slaughter weight (P = .04) increased with increasing dietary ATA. Dressing percentage was lower (P = .04) for pigs fed CO. Increasing dietary ATA had a quadratic effect on 45-min pH (P = .02) and 24-h pH (P = .02). Fresh bellies from pigs fed fat sources with higher saturated fat acids displayed a greater belly depth (P = .04), a larger belly angle (P < .01), and a greater lateral (P < .05) and a lower vertical (P < .05) belly flex. In general, loins from pigs fed 40 ppm GT supplementation exhibited greater L* (P < .05) and lower a* (P < .05) values during shelf life than their counterparts fed 40 ppm ATA. In contrast, feeding GT at 100 ppm resulted in loins with lower (P < .05) L* and greater (P < .05) a* values than 100 ppm ATA. Lipid oxidation in loins from pigs fed CO diets increased at a greater (P < .05) rate than in loins from pigs fed TW diets. Lipid stability in loin improved (P < .01) when dietary ATA increased over 40 ppm. The results indicated that TW can be used as a dietary strategy for firmer and desirable pork bellies from heavy-weight pigs. Additionally, inclusion of GT at 100 ppm or more could be exploited as a preharvest strategy to improve color of loins from heavy-weight pigs during retail display.

This study aimed to test the hypothesis that pork loin pH at 1-day postmortem influences fresh pork quality and degradation of desmin. Fresh pork loins from a commercial facility were evaluated based on 1-day postmortem pH value and ultimately classified as high/normal (5.90–6.46, x̄ = 6.09; n = 25) and low/normal (5.59–5.71, x̄ = 5.67; n = 25). Instrumental color and drip loss were determined on 1-day postmortem chops (2.54 cm). The remaining chops were individually vacuum packaged, aged 1, 7, or 14 d postmortem at 4°C, and then frozen until further analyses. Intramuscular lipid and moisture content was determined through proximate analysis. Instrumental star probe tenderness was evaluated on cooked chops (68°C) from each aging time. Trained sensory panel evaluation and cooking loss were determined on cooked, 14-d postmortem chops. Western blot analysis examined the presence of desmin degradation products within low ionic strength extracts of longissimus thoracis. Based on the pH categories, star probe tenderness, drip loss, cook loss, palatability traits, meat color measurements, and the presence of desmin degradation products were significantly different, whereas intramuscular lipid content was not different. High/normal pH chops had lower L values at 1-d (P < 0.01) and at 14-d postmortem (P < 0.01). Drip loss was greater (P < 0.01) in 1-d low/normal pH chops, as was cooking loss at each aging time (P < 0.01). The trained sensory panel rated high/normal pH loins as juicier (P < 0.01), more tender (P < 0.01), and less chewy (P < 0.05). High/normal pH chops had lower star probe values compared to low/normal pH chops at 1 (P < 0.05), 7 (P < 0.01), and 14 (P < 0.01) d postmortem. High/normal pH chops exhibited a greater proportion of samples containing desmin degradation products within the low ionic strength extract at each aging time (P < 0.01). Variation in 1-d postmortem pH contributes to differences in measurable quality characteristics of fresh pork loin and impacts the presence of desmin degradation products in low ionic strength extracts.

The European beef carcass grading scheme prioritizes meat yield over consumer eating experience, in contrast to the Meat Standards Australia (MSA) grading scheme, which is more focused on eating quality. The Italian–French beef production system, mostly characterized by young bulls and heifers imported from France and then fattened within Italian specialized fattening units, has been underexplored using the MSA grading system. This study examines the impact of animal sex on performance, MSA traits, and predicted MSA quality scores (MQ4 and MSA index) using carcasses from an Italian commercial abattoir. It also assesses how animal performance and MSA traits influence predicted beef eating quality. A Random Forest classifier demonstrated high performance with an accuracy of .98, a specificity of .99, and a sensitivity of .97 after 10-fold cross-validations, confirming key traits such as hot carcass weight, European carcass grading scheme fatness score, and general muscular characteristics as being significantly different due to sex. Entire males had greater muscle development, whereas females had notably higher MSA marbling scores, MQ4 scores, and MSA index (P < .05). Furthermore, categorizing the MSA index into 2 classes and applying a similar Random Forest classifier approach revealed that MSA marbling was the primary factor influencing variability of the MSA index. These findings suggest that beef cuts from females may better meet consumer expectations when using the MSA grading scheme in Europe, making them a suitable category for premium beef branding. Grading carcasses from entire male animals for sensory quality is also important, as it demonstrates their potential to produce good-quality beef with a higher lean meat yield; however, their sensory quality tends to be lower than that of heifers and this is in part reflected by their lower marbling score resulting from reduced intramuscular fat deposition.

Krzywicki’s equations to determine myoglobin forms on meat surfaces allow quantification without creating myoglobin form standards. However, Krzywicki’s equations can result in unrealistic estimations of the percentage numbers of the myoglobin forms. In this study, we revisited Krzywicki’s individual equations of [Oxy]%, [Deoxy]%, and [Met]% based on tissue optics and reexamined the practice of deriving [Oxy]% indirectly using [Deoxy]%, and [Met]%. In addition, the use of baseline correction at 730 nm in myoglobin quantification equations was also evaluated. The modified Krzywicki’s equations were tested on 44 loins (longissimus lumborum, n = 22; psoas major, n = 14; semitendinosus muscles, n = 8). Testing on colorimeter measurements indicated that Krzywicki’s equation for directly estimating [Oxy]% could be sensitive to the baseline reference projected for 730 nm. A new algorithm converting the spectral reflectance to spectral absorption outperformed that which converts the spectral reflectance to reflex attenuance, in implementing Krzywicki’s equations. The modified Krzywicki’s equations performed better, with or without including 730 nm, in specifically resolving a more realistic increase of the [Met]% during the retail display. For example, the new algorithm applied to psoas major muscles revealed a pronounced rapid increase of the [Met]% as the chroma reduced to ∼30, compared to a steady increase of the [Met]% in longissimus lumborum muscles at similar numbers of chroma. The results suggest that modified Krzywicki’s equations provide a more realistic estimation of the percentage numbers of myoglobin forms on steak surfaces than the original equations.

The study aimed to perform biomapping and enviromapping of microbial indicators and pathogens to track microbial changes on carcasses and various surfaces throughout the facility, from harvest to fabrication. Additionally, the study sought to evaluate the effectiveness of interventions and food safety programs in 3 distinct areas of the facility. Samples were collected on 2 separate days to track the fabrication of carcasses from cattle harvested the previous day. On the first day, samples were taken from the lairage area and the harvest floor. On the second day, samples were collected from the fabrication floor. The surfaces tested included walls, equipment, handrails, drains, conveyor belts, cutting boards, and carcasses. MicroTally® cloths (FREMONTA), EZ Reach™ sponges (World Bioproducts), and boot covers (VWR® Basic Protection Anti-Skid Shoe cover, Avantor) were used to collect samples depending on the surface. Most samples, except those from the lairage area, were enumerated using the TEMPO® system (bioMérieux). All samples were tested for pathogens using the BAX® System (Hygiena). The microbial indicators for this study were aerobic counts, Enterobacteriaceae, and Escherichia coli Biotype I, and for pathogens were Escherichia coli O157:H7 (Shiga toxin-producing Escherichia coli [STEC] O157), non-O157 STEC (O26, O45, O103, O111, O121, and O145), and Salmonella spp. As anticipated, statistical reductions among different areas occurred for all 3 locations (P<0.05). The results showed a trend of decreasing microbial loads, starting with high levels of indicators and gradually reducing in the later stages of the process. Salmonella and STEC were detected in the lairage and harvest areas, with some also found on surfaces in the fabrication area before the final interventions were applied. Biomapping and enviromapping are valuable tools for assessing the efficacy of interventions and the effectiveness of food safety programs, enabling manufacturers to make data-driven decisions.

Beef color and tenderness are 2 of the most important attributes contributing to the perception of meat quality and palatability. Studies in search of biomarkers explaining mechanisms regulating meat color stability and tenderness have identified biology in common with both traits. Therefore, the objective of this study was to evaluate the contribution of metabolic factors to animal variation in beef longissimus lumborum tenderness and color stability. Beef carcasses (n = 96) were selected based on tenderness and color stability predictions at grading. Longissimus lumborum muscles were obtained from both sides of each carcass and aged until 12 or 26 d postmortem. Overall tenderness ratings, slice shear force, desmin degradation, and sarcomere length were used to characterize tenderness. Instrumental color values measured during simulated retail display, oxygen consumption, and nitric oxide metmyoglobin-reducing ability were used to characterize color stability. Metabolic traits including pH, glycolytic intermediates, proportion of Type I fibers, sarcoplasmic and myofibrillar carbonyls, and myoglobin concentration were determined on each muscle. Relationships of metabolic traits to tenderness and color stability were assessed with correlation analysis, analysis of variance across clusters derived from tenderness and color stability data, and principal component analysis. Increased nitric oxide metmyoglobin-reducing ability was associated with decreased (P < 0.05) tenderness ratings and desmin degradation and longer sarcomere lengths. Oxygen consumption was associated with decreased tenderness ratings and increased slice shear force (P < 0.05). Both tenderness and lean color stability were negatively related to increased oxidative metabolism. These results indicated that relationships exist between tenderness and lean color stability. Moreover, animal variation in both tenderness and color stability is influenced by muscle metabolism.

Although various studies have evaluated individual factors that influence meat discoloration, limited research has assessed the effects of multiple ground beef cold chain parameters on color. This study evaluated the combined effects of storage temperature, storage time, display temperature, and display time on ground beef color and economic losses due to poor cold chain management. Fresh ground beef chubs (81% lean) were randomly assigned to combinations of 3 storage temperatures (0, 4.5, and 8.9°C), 4 storage times (0, 4, 8, and 12 d), and 3 display temperatures (0, 4.5, and 8.9°C). Visual color was measured every 24 h and a* was measured every 6 h during the 48 h display period. Storage at 0°C minimized discoloration during display compared with storage at 4.5 and 8.9°C. Longer storage times at 0°C did not increase discoloration, whereas prolonged storage at 4.5 and 8.9°C decreased color stability during display. Economic analysis suggested that expected sales loss due to discoloration is positively correlated with storage days, storage temperature, and display temperature. Ground beef stored and displayed at 0°C is shown to minimize expected sales loss with an average loss of $0.29/kg or 2.8% of average retail value. Our results suggest that the total losses to U.S. retailers from cold chain mismanagement are expected to range from $630 million to $1.33 billion when compared with storage and display at 0°C. Simultaneous control of all cold chain parameters is essential for maximizing ground beef color and shelf life