Periparturient cows adapt physiologically to high nutrient demands; however, a poor transition results in economic losses due to decreased production and increased disease incidence. This study investigated the effects of a control diet (CN) and a negative dietary cation–anion difference (DCAD) diet (ND) during close-up on the postpartum performance of low (LBCS)- and high (HBCS)-body-condition-score (BCS: 1-5) cows. Forty Holstein cows were enrolled at −21 d relative to calving into one of the four treatments (n=10/group): LBCS-CN, LBCS-ND, HBCS-CN, and HBCS-ND. The LBCS and HBCS cows had ≤ 3.00 and ≥ 3.25 BCSs, whereas the DCADs in the CN and ND groups were +100 and −100 mEq/kg of dry matter, respectively. Chlorides and sulfates of Mg and Ca were used, where the DCAD=[(Na+K)-(Cl+S)]. This study was approved by the ethical committee for animal welfare at the University of Veterinary and Animal Sciences, Lahore, Pakistan. Repeated measures analysis was conducted using the GLIMMIX procedure of SAS. Milk production, milk composition, and calf birth weight were not different between LBCS and HBCS groups (p>0.05). Prepartum, the ND versus the CN produced numerically a 3.20 kg/d higher amount of milk when fed to HBCS cows, but this increase was only 0.90 kg/d when fed to LBCS cows (p=0.28). However, prepartum DCAD level had no interaction with the BCS group of the cows for any of the observed parameters (p>0.05). The serum concentrations of β-hydroxybutyrate (p=0.03) and free fatty acids (FFA;p=0.01) were also increased in the HBCS cows versus those in the LBCS cows over the 9 wk of lactation. Prepartum DCAD level had no effect on pre- and postpartum BCS and daily rumination time (p>0.05). The ND decreased postpartum concentrations of serum β-hydroxybutyrate (p=0.01) and FFA (p=0.04) compared with those under the CN. In conclusion, a negative DCAD during the close-up period is equally beneficial in low- and high-BCS cows in terms of decreased β-hydroxybutyrate and FFA during the postpartum period.

Modern agriculture is increasingly challenged by the need for precision, sustainability, and reduced human intervention. This work presents an autonomous robotic solution for phenotyping row crops, combining embedded sensors, onboard vision, and fuzzy logic. A differential-drive mobile robot was developed and tested in a 3D-modeled agricultural field, structured in crop rows. It is equipped with infrared sensors for detecting plant obstacles and a front-facing camera for capturing images for phenotypic analysis. To ensure smooth navigation while preserving the integrity of the crops, fuzzy logic is employed to manage sensor uncertainty and dynamically adapt the robot's movements. This approach enables effective autonomous exploration while avoiding damaging contact with plants. The results highlight the relevance of combining sensors and artificial intelligence in the context of smart agriculture, particularly for non-destructive tasks such as automated phenotyping. This work contributes to the advancement of agricultural robotics by demonstrating the potential of intelligent embedded systems in realistic simulated environments for future field applications.

This study evaluated the effects of two lactic acid bacteria (LAB) strains on Zi geese. Ninety 28-day-old healthy Zi geese were randomly divided into three groups (n=30/group), with three replicates (pens) per group and 10 geese (5 males, 5 females) per pen. The control group (CON) received a basal diet. Experimental groups received the basal diet supplemented with 10⁹ CFU/kg Lactobacillus plantarum (LAC) or 10⁹ CFU/kg Pediococcus acidilactici B2 (PED). The trial included a 7-day pre-test and 28-day test period. The pen served as the experimental unit. Data were analysed by one-way ANOVA using SPSS; significant differences (P<0.05) between treatment means were determined by Duncan's test. Compared to CON, the PED group significantly increased average daily gain (ADG) (P<0.05), while the LAC group showed a non-significant increase (P>0.05). Both LAC and PED significantly reduced the feed-to-gain ratio (F/G) (P<0.05 and P<0.01, respectively). Serum urea content was significantly decreased in both LAC (P<0.05) and PED (P<0.01). Serum alkaline phosphatase activity was significantly decreased in both groups (P<0.05). Fecal lactic acid bacteria counts (MRS agar) significantly increased in both LAC (P<0.05) and PED (P<0.01). Fecal Escherichia coli (MacConkey agar) and Salmonella (SS agar) counts significantly decreased in both groups (P<0.05 and P<0.01, respectively). The PED group generally exhibited more pronounced effects than the LAC group. Dietary supplementation with L. plantarum and P. acidilactici B2, particularly the latter, improved growth performance, modulated serum biochemistry, and enhanced the fecal microflora profile in Zi geese.

Early detection of health issues in sows is critical to reducing mortality, controlling disease spread, and improving herd welfare. Traditional monitoring methods are labor-intensive, often ineffective in early-stage detection, and can stress animals during handling. This study introduces a non-invasive ear-cuff heart rate monitor designed for daily use during farmers’ routine checks, enabling real-time, stress-free monitoring.

A prototype device was developed featuring a waterproof compact housing with an adjustable hypoallergenic strap positioned at the sow’s ear base. A photoplethysmography (PPG) sensor detects blood volume changes, and a NodeMCU microcontroller processes the data to trigger RGB LED indicators—blue for low (<70 bpm), green for normal (70–120 bpm), and red for high (>120 bpm). Intended for daily use during farmers’ routine checks, the system allows real-time monitoring without restraining the animal, reducing stress and enabling timely interventions. A purposive survey of 30 pig farmers from Cebu City and Province was conducted using structured questionnaires covering demographics, farm practices, monitoring challenges, and technology adoption attitudes. The findings show 76.67% consider traditional methods ineffective, 46.67% cite inadequate technology, 93.33% express strong interest in the device, and 90% are willing to join pilot testing. The proposed ear-cuff heart rate monitor provides a potential solution that is cost-effective, user-friendly, and non-invasive for integrating heart rate monitoring into the farmer's daily routines, supporting earlier interventions and improved swine health outcomes.

Crop water requirement, the water a crop needs for optimal growth and yield throughout its growing cycle, including transpiration, evaporation, and other losses, must be accurately determined for irrigation scheduling, water resources management, and environmental analysis. Traditionally, this is performed using methods that depend on detailed climate data. However, in many areas, this data may not be available, and the process can take a lot of time. In such cases, using models to predict crop water needs is a good alternative. Machine Learning (ML), a kind of Artificial Intelligence (AI), offers tools that can learn from existing data and make future predictions. This study aimed to predict the water requirement of maize crop of the Samastipur district of Bihar, India, using ML models like Random Forest (RF), Multivariate Adaptive Regression Splines (MARSs), and Support Vector Machine (SVM). It used 20 years (2001–2020) of daily weather data, including maximum and minimum temperature, humidity, wind speed, and solar radiation. The water requirement was first calculated using the FAO-56 Penman–Monteith method combined with crop coefficients. The Gamma test helped choose the best input variables. The data was split into 80% for training the models and 20% for testing. To evaluate the models, this study used three performance measures: the Coefficient of Determination (R²), Root Mean Square Error (RMSE), and Nash-Sutcliffe Efficiency (NSE). Results showed that choosing the right model reduces errors and improves prediction accuracy. Among the models tested, Random Forest performed the best in both training and testing, followed by MARS and then SVM. These results highlight how effective ML models can be for accurately predicting crop water needs.

Improving the economic efficiency of sheep farming depends significantly on increasing animal productivity. In meat sheep breeds, one key productivity indicator is slaughter yield, a trait influenced by numerous biological parameters and regulated by multiple genes involved in diverse molecular pathways. Genetic variation in these pathways can be utilized through molecular marker-assisted selection, especially when markers show significant associations within specific populations.

This study aimed to identify potential molecular markers associated with slaughter yield in intensively fattened lambs of sheep breeds raised in Latvia. A total of 160 lambs (59.5% Latvian Dark-Head breed) from the most common local breeds were included in a controlled fattening program. Slaughter yield (%) was measured (44.45±2.52%) at age 149.65±14.26 days, and genotyping of 57 SNPs across eight candidate genes was conducted. Association and regression analyses were performed to assess the relationship between genotypes and slaughter yield.

Among the analyzed SNPs, 23 (40.35%) showed statistically significant associations with slaughter yield in the Latvian sheep population. The strongest associations were found for two SNPs in the UPC2 gene (rs412180048 A>G and rs405808821 C>T) and two in the MTOR gene (rs419418343 C>T and rs160776285 T>C). In both UPC2 SNPs, the highest slaughter yields were observed in lambs homozygous for the common allele, although this genotype was rare in the Latvian population. In the case of MTOR, the highest yield was found in both homozygous common and heterozygous genotypes, which were present in only 25% of the sampled animals.

These results suggest that SNPs in UPC2 and MTOR genes have strong potential as molecular markers for improving slaughter yield through marker-assisted selection in Latvian sheep breeding programs.

Acknowledgements: This research was supported by the Latvian Council of Science (LZP-2021/1-0489) and the University of Latvia Postdoctoral Project No. 1.1.1.9/LZP/1/24/027 “Genetic markers as a basis for the excellence of carcass quality traits in the Latvian dark-headed sheep breed”.

Infectious diseases are a major problem for all livestock farming, including swine. They are responsible for numerous economic losses and, in the case of zoonotic diseases, also for public health problems. In this study, the seroprevalence of 10 common swine pathogens was evaluated, evaluating any risk factors related to increased exposure risks. A total of 420 blood samples were collected in the Campania region, southern Italy, and tested with specific commercial ELISAs. The highest seroprevalences were found for hepatitis E virus (HEV, 41.4%), porcine reproductive and respiratory virus (PRRSV, 16.7%), and porcine epidemic diarrhea virus (PEDV, 14.8%). Lower prevalences were found for transmissible gastroenteritis (TGEV, 5.5%), porcine respiratory coronavirus (PRCV, 0.9%), Mycobacterium avium paratubercolosis (MAP, 3.5%), bovine viral diarrhea virus (BVDV, 3%), Schmallenberg virus (SBV, 5.3%), and Coxiella burnetii (4.1%). No animal had antibodies against Brucella suis. Furthermore, statistical tests correlated sex, age, and type of farm with higher exposure to HEV, SBV, PEDV, and TGEV. Therefore, in the study area, exposure to pathogens causing damage to the swine industry economy, such as PRRSV and PEDV, as well as to those with zoonotic potential, such as Coxiella and HEV, was frequent. The results of this study underline the importance of continuous surveillance in swine farming in order to understand the main circulating pathogens, risk factors, and measures to be taken.

The sweet potato (Ipomoea batatas (L.)) is a herbaceous, perennial plant belonging to the Convolvulaceae family, native to Central America. It is a cosmopolitan plant of worldwide importance. The evaluation of Distinctness, Uniformity, and Stability is an important step for the protection of cultivars, aiming to assess whether the selected genotypes meet technical requirements. These must be clearly distinguishable, with relevant homogeneous characteristics in the population, including standard features and uniformity, to maintain their traits. The objective was to carry out Distinctness, Uniformity, and Stability assessments in sweet potato genotypes for the purposes of registration and protection of cultivars. The trial was set up in an area belonging to UNESP, on the Jaboticabal campus, from September 2022 to December 2023, in a randomized block design with three repetitions. Five experimental genotypes and five sweet potato cultivars were evaluated: Cerat25-01, Cerat35-11, Cerat51-30, Cerat60-22, Cerat60-25, Amélia, Brasilândia Roxa, Beauregard, Gaita, Princesa, and Rubissol. The useful area of the plot corresponded to 40 plants, with 30 central plants being evaluated, and harvesting was performed 90 days after planting. The variables assessed included root shape, flesh color, number of lobes in the leaf, and anthocyanin pigmentation in the internode. Data analysis was conducted using the Genes Software through mean tests. In both the genotypes and the cultivars, there was no significant difference in the means, indicating standard behavior of the genotypes and cultivars. The general shape characteristic of the root was such that 99% of the genotypes showed a table market shape and 1% showed an industrial market shape. Of these, 54% had orange flesh and 46% had white and yellow flesh. Among the evaluated genotypes, 73% had between five and seven lobes and 27% had no lobes. Regarding anthocyanin pigmentation in the internode, 18% of the genotypes had a medium-to-strong presence of pigmentation and 82% had no pigmentation. The results obtained highlight the differences between genotypes and cultivars, meeting the requirements of the Ministry of Agriculture, Livestock and Supply (MAPA) for intellectual protection.

Introduction:

The integration of artificial intelligence (AI) and digital technologies into the agricultural sector represents a potentially transformative development for increasing productivity, optimizing resource use, and promoting sustainable practices. However, the diffusion of such innovations is far from uniform: significant disparities are emerging between small and large-scale farms, as well as between developed and marginal rural areas. This scenario raises serious concerns regarding existing structural inequalities and the risk of digital exclusion for a substantial portion of the farming population, particularly in disadvantaged contexts.

Methods:

This study presents a systematic review of recent empirical and conceptual literature (2019–2025), focusing on the adoption of smart farming technologies in sub-Saharan Africa. It analyzes eight key studies from Scopus-indexed sources, assessing infrastructure availability, economic barriers, and farmers' digital literacy through qualitative and quantitative approaches.

Results:

The findings reveal a substantial digital divide. Smallholder farmers report limited access to ICTs due to high costs, poor connectivity, and lack of technical support. In South Africa, 78.8% of surveyed farmers perceived digital tools as too expensive and 81% lacked the required skills (Bontsa et al. 2024). These barriers are compounded by perceptions of limited reliability and low trust in digital systems.

Conclusions:

While AI-driven farming promises increased efficiency, its benefits remain unequally distributed. Without targeted interventions, such as infrastructure development, affordable technology, and context-specific training, digital agriculture risks reinforcing existing inequalities. To ensure an inclusive agricultural transformation, policies must prioritize digital equity, particularly for marginalized rural producers.

No prior studies have compared the in situ degradation of fresh and ensiled Guinea grass cultivars grown in the microclimate of Visayas State University, Baybay City, Leyte, the Philippines. This study evaluated the in situ degradation characteristics of two Guinea grass cultivars—Local Guinea grass (LG) and Mombasa grass (MG)—harvested at 30 and 45 days. Six test diets were prepared: fresh LG (FLG), fresh MG (FMG), and ensiled LG and MG harvested at 30 and 45 days (LG30, LG45, MG30, and MG45, respectively). Three rumen-fistulated Brahman heifers (180 ± 10 kg bodyweight) were dewormed, pre-conditioned, and incubated with the test diets using the sequential addition method at 0, 24, 48, and 72 hours. At 0 and 24 hours, all treatments showed a comparable degradation of DM and ADF (p>0.05). The NDF degradation of all treatments was comparable only at 0 hours. At 24 hours, MG30 showed significantly greater NDF degradation compared to most other treatments (p<0.01). At 48 and 72 hours, MG30 showed a significantly greater degradation of DM, NDF, and ADF compared to most other treatments (p<0.01). MG30 showed a significantly greater effective degradability of DM, NDF, and ADF compared to most other treatments (p<0.01). These findings indicate that MG30 has superior in situ degradability characteristics, making it a promising forage option for ruminant diets.