M. T. Kidd’s research while affiliated with University of Arkansas at Fayetteville and other places

Object tracking, especially animal tracking, is one of the key topics that attract a lot of attention due to its benefits of animal behavior understanding and monitoring. Recent state-of-the-art tracking methods are founded on deep learning architectures for object detection, appearance feature extraction and track association. Despite the good tracking performance, these methods are trained and evaluated on common objects such as human and cars. To perform on the animal, there is a need to create large datasets of different types in multiple conditions. The dataset construction comprises of data collection and data annotation. In this work, we put more focus on the latter task. Particularly, we renovate the well-known tool, LabelMe, so as to assist common user with or without in-depth knowledge about computer science to annotate the data with less effort. The new tool named as TrackMe inherits the simplicity, high compatibility with varied systems, minimal hardware requirement and convenient feature utilization from the predecessor. TrackMe is an upgraded version with essential features for multiple object tracking annotation.

Chick sexing, the process of determining the gender of day-old chicks, is a critical task in the poultry industry due to the distinct roles that each gender plays in production. While effective traditional methods achieve high accuracy, color, and wing feather sexing is exclusive to specific breeds, and vent sexing is invasive and requires trained experts. To address these challenges, we propose a novel approach inspired by facial gender classification techniques in humans: facial chick sexing. This new method does not require expert knowledge and aims to reduce training time while enhancing animal welfare by minimizing chick manipulation. We develop a comprehensive system for training and inference that includes data collection, facial and keypoint detection, facial alignment, and classification. We evaluate our model on two sets of images: Cropped Full Face and Cropped Middle Face, both of which maintain essential facial features of the chick for further analysis. Our experiment demonstrates the promising viability, with a final accuracy of 81.89%, of this approach for future practices in chick sexing by making them more universally applicable.

Bacterial chondronecrosis with osteomyelitis (BCO) lameness is a major welfare issue for broiler production worldwide affecting approximately 1.5% of broilers over 42 days old. Excessive body weight gain causes mechanical stress on long bones, leading to micro-fractures. This condition induces a bacterial infection of fractures, resulting in bone necrosis and eventual BCO lameness. Increasing gut integrity and supporting Calcium metabolism contribute to the optimal bone structure and subsequently reduce BCO lameness. Probiotics thus provide an excellent strategy for alleviating BCO due to the improvement of intestinal integrity and barrier function. Accordingly, the present study investigated the lameness reduction through the feed supplementation of a selected probiotic. Broiler chickens were assigned to three treatments, including a control litter group (FL), a PoultryStar®Bro probiotic fed group (BRO), and a control wire-flooring group (CW) designed to induce BCO lameness. The probiotic significantly decreased lameness by 46% compared to the control group (p < 0.05). The most predominant bacteria identified from the BCO lesions were Staphylococcus cohnii and Staphylococcus lentus. Moreover, significant increments of tight junction gene expression in jejunum and ileum, plus numerical improvements of body weight gain (BW; +360 g) and feed conversion ratio (FCR; −12 pts) were observed in BRO-supplemented birds.

Two experiments were conducted to assess the efficiency of including the novel protein ingredient Arthrospira platensis or improved soybean meal in a broiler diet. The first experiment aimed to determine the feeding value of soybean meal produced from varieties of soybeans bred for increased amino acid content (SBAA) and improved oil content (SBO) compared to a conventional soybean variety in an ANOVA design fed to Cobb 500 female broilers for 28–42 d. The SBAA and SBO soybeans contained overall higher amino acid content and lower oligosaccharide content compared to the conventional soybean variety in addition to improved oil quality. The second experiment assessed the novel protein ingredient microalgae, Arthrospira platensis (algae), and was conducted to evaluate algae and corn distillers’ grain (DDGS) inclusion on broiler performance for a 28–42 d feeding period in Cobb CF05 male broilers with a 2 × 2 factorial treatment array. Prior to the experimental period, all birds were reared on common feeds. In Experiment 1, birds were fed a diet containing 20% inclusion of an experimental soybean source in the form of full-fat soybean meal. In Experiment 2, the four dietary treatments consisted of diets containing algae at inclusion levels of either 0 or 2% and DDGS at inclusion levels of 0% and 8%. Diets were fed to 288 female broilers (Experiment 1) and 384 male broilers (Experiment 2), placed in eight replicate pens of twelve birds, and live performance was assessed from d 28 to 42. At d 42, six birds from each pen were randomly selected and processed for evaluation of carcass traits and incidence of woody breast. For Experiment 1, all performance data were analyzed using a one-way ANOVA using JMP Pro 16 software with diet as the fixed effect and block as a random effect. Statistical significance was considered at p ≤ 0.05. No significant responses were observed for any recorded measurement for live performance, carcass traits, or woody breast. All data in Experiment 2 were analyzed as a full factorial with a mixed model using JMP software with algae, DDGS, and algae × DDGS as fixed effects and block as a random effect. The F-protected Fisher’s LSD test was used to separate means when p ≤ 0.05. No significant responses were observed for the algae, DDGS, and algae × DDGS influences on BWG, FI, and FCR or processing characteristics; the ingredient source did not affect bird performance. Experimental soybean lines developed at the University of Arkansas were able to be incorporated into broiler diets without decreasing performance. Algae has the potential to be a protein-contributing ingredient for broilers.

In the food industry, assessing the quality of poultry carcasses during processing is a crucial step. This study proposes an effective approach for automating the assessment of carcass quality without requiring skilled labor or inspector involvement. The proposed system is based on machine learning (ML) and computer vision (CV) techniques, enabling automated defect detection and carcass quality assessment. To this end, an end-to-end framework called CarcassFormer is introduced. It is built upon a Transformer-based architecture designed to effectively extract visual representations while simultaneously detecting, segmenting, and classifying poultry carcass defects. Our proposed framework is capable of analyzing imperfections resulting from production and transport welfare issues, as well as processing plant stunner, scalder, picker, and other equipment malfunctions. To benchmark the framework, a dataset of 7,321 images was initially acquired, which contained both single and multiple carcasses per image. In this study, the performance of the CarcassFormer system is compared with other state-of-the-art (SOTA) approaches for both classification, detection, and segmentation tasks. Through extensive quantitative experiments, our framework consistently outperforms existing methods, demonstrating re- markable improvements across various evaluation metrics such as AP, AP@50, and AP@75. Furthermore, the qualitative results highlight the strengths of CarcassFormer in capturing fine details, including feathers, and accurately localizing and segmenting carcasses with high precision. To facilitate further research and collaboration, the source code and trained models will be made publicly available upon acceptance.

Further reductions of crude protein in broiler diets may cause essential amino acids traditionally considered to be supplied at adequate levels to become limiting. Requirement data for histidine are currently scarce and this amino acid is uniquely able to be stored within the body. Thus, the objectives of this study were to evaluate female Cobb 500 broiler growth performance and carcass characteristic responsiveness to increasing digestible histidine to digestible lysine ratios (dHis:dLys) during a 22- to 42-day feeding period and determine if dHis:dLys in preceding feeds influence the former responses. Starter diets were formulated to contain dHis:dLys of 33% or 38% and grower diets were formulated to contain dHis:dLys of 28%, 33%, 38%, or 43%, which gave rise to a 2 × 4 factorial arrangement of treatments. No interactions (p > 0.05) between the starter or grower diet dHis:dLys occurred. Increasing the dHis:dLys from 22 to 42 days resulted in body weight gain, feed conversion ratio, and total breast meat yield quadratic responses (p ≤ 0.05). Additionally, it appears that dHis:dLys between 33% and 38% fed during the starter phase does not influence broiler responsiveness to different dHis:dLys during the grower phase.

Our prior research demonstrated a 20% to 25% reduction in bacterial chondronecrosis with osteomyelitis (BCO) lameness in broilers with organic Zn, Mn, and Cu (Availa® ZMC) supplementation. Expanding on this, we investigated the optimal timing for Availa® ZMC feeding to mitigate BCO lameness and reduce feed additive costs in the poultry industry. In this study, we compared the application of 0.15% Availa® ZMC for 56 days, the first 28 days, and the last 28 days. The experimental design was a randomized block design involving 1560 one-day-old chicks distributed across two wire-floor pens as BCO source infection and four treatment groups with six replicates. The source of BCO infection exhibited a cumulative lameness incidence of 83%, whereas the negative control group showed a 77% cumulative incidence of lameness (p = 0.125). Administering 0.15% of Availa® ZMC during the initial 28 d resulted in a 41.3% reduction in BCO incidence, significantly different from the supplementation during the last 28 d (p < 0.05). However, this reduction did not differ substantially (p > 0.05) from the 56d application period. Hence, administering 0.15% Availa® ZMC during the first four weeks emerges as the optimal timing protocol, providing a defense against lameness comparable to the continuous supplementation throughout the complete production duration. Implementing this feeding approach reduces the cost of feed additive, promotes the health of skeletal bones, and effectively protects against BCO lameness in broilers, offering a valuable consideration for producers seeking optimal outcomes in the poultry industry.