Search Results (118)

Canned fish products enable long-term preservation of fish, a vital source of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). Despite research on lipid composition, gaps remain in understanding the bidirectional fatty acid (FA) exchange between fish muscle and coating media during processing and storage. After a systematic literature search across five databases (PubMed, Scopus, Web of Science, Wiley Online Library, Cochrane Library), 20 studies were included examining FA profiles across fish species, filling media (vegetable oils, brine, tomato sauce), and storage durations (up to 5 years). Five studies showed that n-3 FAs migrate from fish to the filling medium, enhancing its nutritional value, while fish muscle absorbs FAs from the oil, increasingly resembling the filling medium. The use of n-6 FA-rich oils (sunflower, soybean) lowered the n-3/n-6 ratio in flesh. Conversely, aqueous media (brine) and tomato sauce maintained better ratios. EPA and DHA content generally decreased due to canning and storage, with retention varying by fish species, filling medium, and sterilization method. This review underscores significant FA exchange between fish and filling media, confirming bidirectional lipid interchange during processing. To optimize health benefits, aqueous packing media are recommended to preserve lipid profiles or to consume the covering oil to recover nutrients. Further research is needed on other factors altering FA content in canned fish such as environmental and geographical variables (including catching season), pre-canning preparation and sterilization steps (such as freezing, steaming, and frying), sterilization conditions (time, temperature, F₀ value) and lipid oxidation induced by thermal processing. Full article

Rapeseed press cake (RPC), the protein-rich residue from edible oil production, is currently underutilized and is primarily used as animal feed. This study aimed to investigate pressurized liquid extraction (PLE) for protein recovery from RPC using response surface methodology (RSM) with precipitation yield (PY) as the response variable. Following alkaline extraction, proteins were precipitated at their isoelectric point, and solid residues were freeze-dried to obtain protein powders. Conventional extraction (CE) under magnetic stirring at room temperature was used as a reference. The results demonstrated that increasing pH from 8 to 11 significantly enhanced protein extraction efficiency for both methods. PLE exhibited superior performance, achieving higher PY compared to CE while drastically reducing extraction time from 120 min (CE) to 6 min (PLE). Optimal conditions were identified at a solid-to-liquid ratio of 0.10 g/mL, 150 °C, and 6 min, yielding a PY of 14.9%, protein recovery in extract (PR_E) of 43.8%, and protein recovery in precipitated mass (PR_P) of 20.0%, with a protein content (PC_P) of 647.2 mg albumin eq./g. RSM analysis identified extraction temperature as the most critical parameter for PLE, highlighting its dominant role in mass transfer. Finally, amino acid (AA) analysis revealed that protein powders were rich in essential AAs, with glutamic and aspartic acids being the most abundant. Additionally, PLE-derived protein powders exhibited enhanced solubility. This study confirms PLE as a highly promising and time-efficient technique for protein recovery from RPC, supporting the potential of sustainable valorization of agro-industrial by-products and promoting a circular economy model within the food industry. Full article

Morinda royoc L. is a shrub with high pharmacological value due to its antimicrobial and metabolic bioactivity. However, its low germination limits its commercial use and conservation. This study characterized the morphophysiology of its diaspores, evaluated pre-germination treatments (mechanical endocarp removal and GA₃ soaking at 600 and 1200 ppm), and determined its storage behavior. The analyses showed a highly significant effect where manual endocarp removal overcame mechanical restriction and eliminated dormancy, achieving 100% germination within two weeks, whereas the control reached only 51.5% after 23 weeks. Treatments with GA₃ did not improve germination compared to the control, and an inhibitory effect was observed at 1200 ppm, although 100% of the non-germinated embryos remained viable. The seeds maintain viability above 90% after three months of storage at 25, 5, and −20 °C. Our findings suggest that M. royoc seeds exhibit orthodox behavior (tolerant to desiccation and freezing) and non-deep physiological dormancy. This study documents, for the first time, a method for the complete release of dormancy in the species, which is essential for successful propagation. Our results provide a rapid and economical germination method that facilitates mass propagation, establishing the technical basis for the transition toward commercial-scale cultivation and ensuring the effective conservation of M. royoc germplasm. Full article

The effects of thermal and non-thermal pretreatments combined with different drying methods on the drying kinetics, physicochemical properties, and bioactive compounds of the Chilean wild mushroom Morchella conica were investigated. Fresh samples were subjected to hot-air drying (HAD, 60 °C), freeze-drying (FD), and a hybrid process (FD–HAD), applied directly or after pretreatments including thermal pre-drying (55 and 75 °C), ultrasound (US, 10 and 20 min), and high hydrostatic pressure (HHP, 600 MPa). Drying curves were successfully fitted using the Weibull model (R² > 0.987), showing that HAD combined with thermal and ultrasound pretreatments increased drying rates, while FD–HAD reduced total drying time. Freeze-drying better preserved color (ΔE < 2) and minimized shrinkage (<8%), whereas HAD produced darker samples and greater structural deformation. Water activity decreased below 0.30 in most treatments, ensuring microbiological stability. Thermal pretreatments enhanced total phenolic content, while FD preserved antioxidant capacity. Principal component analysis explained 62.2% of the total variance, revealing distinct quality profiles among drying methods. Overall, FD and hybrid FD–HAD combined with moderate pretreatments showed the best balance between drying efficiency and quality preservation, while HHP improved antioxidant properties under specific conditions. These findings highlight the potential of integrating innovative pretreatments with drying technologies to optimize processing of Morchella conica. Full article

Cross-topic stance detection degrades when encoders entangle stance signals with topic-specific vocabulary, causing representations that fail to transfer to unseen targets. Existing methods commit to either topic-invariant or topic-aware representations and apply the same strategy uniformly to every input, sacrificing complementary information. We propose ReDyGait, a three-stage framework that disentangles these two types of signals through dedicated contrastive pre-training and recombines them adaptively at inference time. Stage 1 trains a topic-invariant encoder with supervised contrastive loss over cross-topic positives. Stage 2 trains a topic-contextual encoder with bidirectional pair contrastive loss over within-topic positives; both stages employ topic-aware hard negative mining to prevent shortcut learning. Stage 3 freezes the two contrastive encoders and learns a gating network that produces per-instance weights over invariant, contextual, and base-encoder pathways. On VAST, ReDyGait achieves a macro-averaged F1 of 0.782 in the zero-shot setting and 0.752 in the few-shot setting, improving over the strongest baseline by 1.1 points in both; on SEM16t6 in a leave-one-target-out setup, ReDyGait reaches an average F1 of 0.612. Analysis of the learned gate weights shows that the model shifts toward the invariant pathway for unfamiliar topics and toward the contextual pathway when topic-specific patterns are available, confirming that the disentanglement operates as intended. Full article

Drying significantly influences the quality of micellar casein (MC) powder. This study investigated the effects of three pre-freezing temperatures (−20 °C, −80 °C, and liquid nitrogen) prior to freeze drying on the structure and rennet-induced gelation properties of MC powder. The results showed that as the pretreatment temperature decreased, the degree of disruption to the secondary and tertiary protein structures was reduced, while the particle size gradually increased. In terms of rennet-induced gel properties, the untreated raw MC consistently outperformed all MC powder samples. Among the MC powders, the sample pre-frozen at −80 °C and then freeze-dried (FD-80) exhibited the highest gel strength and a relatively shorter rennet coagulation time. The observed microstructures of the rennet-induced gel were consistent with the rheological results, showing that samples with smaller particle sizes formed more regular and compact gel networks. In conclusion, the MC powder prepared via pre-freezing at −80 °C and then freeze-drying better preserved protein structure and demonstrated superior rennet-induced gel properties, which were closely related to particle size. This study provides theoretical insights for the application of MC powder in products such as cheese, processed cheese, and protein-fortified foods. Full article

Estimating the post-mortem interval (PMI) using microRNAs (miRNAs) in vitreous humor (VH) is a promising technique in forensic pathology. However, the reliability of quantitative Real-Time PCR (qPCR) data in this matrix is currently constrained by a critical methodological challenge: the lack of a rigorously validated endogenous reference gene (normalizer) capable of correcting for non-biological variations without being influenced by decomposition. This study aimed to identify a robust reference gene for VH analysis by performing a comparative validation of two candidates proposed in the literature: miR-222-3p and miR-96-5p. VH samples were collected from 47 forensic autopsy cases with estimated PMIs ranging from 3 to 24 h. The validation process assessed three key parameters: amplification detectability, expression stability (Coefficient of Variation, CV), and statistical independence from both the PMI and the pre-analytical freezing interval using regression models. MiR-222-3p was rejected as a normalizer due to poor detectability, failing to reach the detection threshold (Cq < 35) in 61.7% of cases (29/47). Conversely, hsa-miR-96-5p was validated as a stable reference gene. It demonstrated high detectability and expression stability (CV = 9.07%) among valid samples. Crucially, linear regression analysis showed no significant correlation between hsa-miR-96-5p levels and either the PMI (p = 0.69) and the pre-freezing time (p = 0.70). This study demonstrates that miR-222-3p is unsuitable for forensic casework in VH due to instability. We identified and validated hsa-miR-96-5p as a robust endogenous reference gene. Its adoption is recommended to standardize future molecular thanatochronology studies and improve the accuracy of PMI estimation models. Full article

Network slicing is a cornerstone of 5G/6G vertical services, yet practical deployments require mobile network operators (MNOs) to adjust slice service level agreement (SLA) weights based on quality of experience (QoE), causing rapid non-stationary objective changes that can destabilize deep reinforcement learning (DRL) slicing policies and necessitate retraining. This paper proposes Continual Mixture of Experts (CoMEx) for fast policy adaptation. CoMEx pre-trains and freezes multiple expert policies under diverse SLA preferences, explicitly appends the SLA weight vector to observations, and trains a DRL-based gating network to fuse expert actions at the step level for fast adaptation to unseen SLA configurations. To broaden coverage without degrading existing experts, CoMEx further incorporates a masked expert expansion mechanism that incrementally adds new experts and fine-tunes the gate. Step-level DRL gating demonstrates superior generalization in RAN slicing, attaining a mean score of 78.95 under unseen SLA weights—surpassing episode-level and supervised gating by 2.40% and 27.67%, respectively. Moreover, CoMEx’s extensibility is highlighted by a 7.08% performance boost (reaching 84.54) upon the addition of a fourth expert. Such results confirm the framework’s capacity for timely and robust policy adaptation in non-stationary SLA environments. Full article

Metabolic profiling enables comprehensive characterisation of the small molecules that are part of the biochemical composition of biological fluids. The most widely profiled biofluids include serum and plasma. Additionally synovial fluid provides a direct reflection of the metabolomic environment of joints and holds promise for biomarker discovery in arthropathies. However, the reproducibility of metabolomics data is highly sensitive to pre-analytical variation, and at the present time, standardised protocols for synovial fluid remain underdeveloped. This review aims to identify and evaluate the existing literature on effects of biofluid pre-analytical handling treatments on metabolic profiles. This review was conducted and reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) guidelines. A search was carried out to identify studies employing LC-MS, GC-MS, and NMR spectroscopy for the investigation of factors including sample collection variables, pre-centrifugation conditions, centrifugation parameters, post-centrifugations conditions, sample storage conditions, and freeze/thaw cycling. Best practice recommendations emerging from this review include the use of additive free serum and heparin plasma tubes, the centrifugation of samples within two hours of collection, immediate storage of samples at −80 °C, and avoidance of repeated freeze/thaw cycling. However, while pre-analytical influences have been extensively characterised for plasma and serum, evidence for synovial fluid remains limited. Overall, the findings highlight the existing recommendations for plasma and serum and demonstrate the need for standardised pre-analytical protocols and validation of quality control markers to advance synovial fluid metabolomics. Full article

Background: Following gonadotropin-releasing hormone (GnRH) agonist trigger and “freeze all” in order to prevent ovarian hyper-stimulation syndrome (OHSS), patients are usually anxious to continue immediately with a frozen embryo transfer (FET). Currently, the preferred FET protocol in based on natural or induced ovulation. Objectives: Do ovarian stimulation and GnRH-a, used to trigger final oocyte maturation, affect the reproductive axis in the next natural cycle? Design: An observational case series of 100 subsequent in vitro fertilization (IVF) patients to whom GnRH-a (Triptorelin 0.2 mg) was given for final oocyte maturation in the context of ovarian hyper-stimulation syndrome prevention, followed by embryos “freeze all”. Methods: In the next natural cycle, patients were followed to detect a dominant follicle (≥17 mm), at which time ovulation was triggered with human chorionic gonadotropin (hCG, 250 µg), and FET was scheduled according to embryo’s age on freezing day. Results: Whereas natural ovulation according to pre-IVF treatment was predicted to be on cycle day 14, the actual hCG-scheduled ovulation in our patients was on day 21. In eight patients, follicular activity was not detected after 15–28 days; therefore, the natural cycle frozen embryo transfer approach was abandoned. Conclusions: Ovarian stimulation and GnRH-a used to trigger final oocyte maturation in IVF patients inhibits the reproductive axis for days. Therefore, natural ovulation in the subsequent cycle may be deferred for about one week relative to the patient’s pre-IVF menstrual cycle pattern. This may help schedule clinic visits to optimize monitoring efficiency. Full article

Controlling pre-analytical conditions for medical biology tests, particularly during transport, is crucial for complying with the ISO 15189 standard and ensuring high-quality medical services. The use of drones, also known as unmanned aerial vehicles, to transport clinical samples is growing in scale, but requires prior validation to verify that there is no negative impact on the test results provided to doctors. This study aimed to establish a secure, high-quality solution for transporting biological samples by drone in a coastal region of France. The 80 km routes passed over several densely populated urban areas, with take-off and landing points within hospital grounds. The analytical and clinical impact of this mode of transport was compared according to two protocols: an interventional clinical trial on 30 volunteers compared to the reference transport by car, and an observational study on samples from 126 hospitalized patients compared to no transport. The system enabled samples to be transported without damage by maintaining freezing, refrigerated, and room temperatures throughout the flight, without any significant gain in travel time. Analytical variations were observed for sodium, folate, GGT, and platelet levels, with no clinical impact on the interpretation of the results. There is a risk of time-dependent alterations of blood glucose measurements in heparin tubes, which can be corrected by using fluoride tubes. This demonstrated the feasibility and security of transporting biological samples over long distances in line with the ISO 15189 standard. Controlling transport times remains crucial to assessing the quality of analyses. It is imperative to devise contingency plans for backup solutions to ensure the continuity of transportation in the event of inclement weather. Full article

Household food waste is a complex issue shaped by socioeconomic conditions, household size, time and resource constraints, and routine food management behaviors. Understanding the practices, attitudes, barriers, and motivators that influence food waste is crucial for designing effective and sustainable interventions for low-income households experiencing high rates of food insecurity. Guided by community input, a food waste reduction education program was developed and piloted in seven California counties. In total, 50 adults were enrolled; 40 completed pre/post surveys, 17 completed food waste audits, and 14 responded to a four-month follow-up survey. Survey results showed significant increases in key food management behaviors: making and using a shopping list, freezing food, and using leftovers in future meals. The percentage of participants discarding food because of package dates declined from 53% to 30%. All measures of barriers and self-efficacy improved. Food audit results indicated the volume and weight of solid and liquid food waste decreased, although the changes were not statistically significant. At follow-up, all respondents reported checking their refrigerator and cupboards before shopping, making a shopping list, and storing and reheating food safely all or most of the time. Overall, the findings demonstrate that practical, skills-based education can help low-income households reduce food waste. Full article

Opportunities to capture anticipated niche markets for diverse populations continue to rise. Okra (Abelmoschus esculentus L.), considered a high-value crop, is rich in nutritional and medicinal properties; however, fresh okra is highly perishable. This study examined the effects of thermal pre-treatments and drying processes in combination on the nutritional quality of dried okra. The experiment consisted of two thermal treatments (steam-blanched and hot water-blanched, and the control) and three drying treatments (freeze-dried, hot air-dried, and infrared-dried). Okra was grown in black plastic mulch, harvested twice per week, and processed three times throughout the growing season. The study analyzed moisture content, water activity, phytonutrients, ascorbic acid, β-carotene, and antioxidant activities. No significant differences were observed in moisture content and water activity among the treatments. Significant differences were observed among treatments and harvest time for total phenolic and flavonoid contents and antioxidant activity. Notable differences in β-carotene content were observed across all treatments. Based on the findings, the steam-blanched freeze-dried treatment was the most effective preservation technique for maintaining the nutritional and functional quality of dried okra. Hot water-blanching, hot air-drying, and infrared-drying were the least effective for the development of a high-value, nutrient-dense dried okra value-added product. Full article

Respiratory diseases pose a significant threat to intensive pig farming, and cough recognition serves as a key indicator for early intervention. However, its practical application is constrained by the scarcity of labeled samples and the complex acoustic conditions of farm environments. To address these challenges, this study proposes a lightweight pig cough recognition method based on a pre-trained model. By freezing the backbone of a pre-trained audio neural network and fine-tuning only the classifier, our approach achieves effective knowledge transfer and domain adaptation with very limited data. We further enhance the model’s ability to capture temporal–spectral features of coughs through a time–frequency dual-stream module. On a dataset consisting of 107 cough events and 590 environmental noise clips, the proposed method achieved an accuracy of 94.59% and an F1-score of 92.86%, significantly outperforming several traditional machine learning and deep learning baseline models. Ablation studies validated the effectiveness of each component, with the model attaining a mean accuracy of 96.99% in cross-validation and demonstrating good calibration. The results indicate that our framework can achieve high-accuracy and well-generalized pig cough recognition under small-sample conditions. The main contribution of this work lies in proposing a lightweight fine-tuning paradigm for small-sample audio recognition in agricultural settings, offering a reliable technical solution for early warning of respiratory diseases on farms. It also highlights the potential of transfer learning in resource-limited scenarios. Full article

Pre-trained models have played important roles in many tasks, such as domain adaptation and out-of-distribution generalization, by transferring matured knowledge. In this paper, we study Neural Architecture Search (NAS) in the feature space level and observe that low-level features of NAS-based networks (generated networks from a NAS space) become stable in the earlier stage of training. In addition, these low-level features are similar to those from hand-crafted networks such as VGG, ResNet, and DenseNet. This phenomenon is consistent over different search spaces and datasets. Motivated by these observations, we propose a new architectural method for NAS, called Knowledge-Transfer NAS, which utilizes the features from a pre-trained hand-crafted network. Specifically, we replace the first few cells of NAS-based networks with pre-trained manually designed blocks and freeze them, and then only train the remaining cells. We perform extensive experiments using various NAS algorithms and search spaces, and show that Knowledge-Transfer NAS achieves higher/comparable performance while requiring less memory footprint and search time, offering a new perspective on the applicability of pre-trained models for improved NAS algorithms. Full article