Hasil untuk "Animal biochemistry"

A. P. Megha, Surej Joseph Bunglavan, K. Ally et al.

The present study evaluated the effect of supplementation of emulsifier and probiotic supplement with sodium butyrate in milk replacer on haircoat condition and blood biochemical profile in pre-ruminant crossbred calves. Twelve healthy calves, seven days old, were randomly assigned to two treatment groups (n 6) for a 90-day feeding trial following step-down feeding regimen. The T1 group received milk replacer containing lysolecithin at 4gcalfday and Lactobacillus rhamnosus GG (1010 CFU) at 1gcalf day, while the T2 group was provided the same diet with sodium butyrate at 2 gcalf day. Calf starter and green fodder were offered ad libitum to all animals. Alopecia was assessed weekly using a 03 point scoring scale and blood biochemical indices including haemoglobin, serum glucose, total protein, albumin, globulin, A:G ratio, blood urea nitrogen, calcium, and phosphorus were measured at the end of the trial. Results showed that alopecia was evident in both groups between the 2nd and 6th weeks, but scores were significantly (P

Xinyu Shi, Li-Yi Wei, Nanxuan Zhao et al.

We introduce the concept of notational animating, an interaction paradigm for animation authoring where users sketch high-level notations over static drawings to indicate intended motions, which are then interpreted by automatic methods (e.g., GenAI models) to generate animation keyframes. Sketched notations have long served as cognitive instruments for animators, capturing forces, poses, dynamics, paths, and other animation features. However, such notations are often context-dependent, non-categorical, ambiguous, and composable based on our analysis of real-world animator-produced sketches. To facilitate interpretation, we first formalize these notations into a structured animation representation (i.e., source, path, and target). We then built an animation authoring system that translates high-level notations into the formalized intended animation, provides dynamic UI widgets for fine-grained parameter control, and establishes a closed feedback loop to resolve ambiguity. Finally, through a preliminary study with animators, we assess the usability of notational animating, reflect its affordance, and identify its contexts of use.

Alejandro Belanche, André Bannink, Jan Dijkstra et al.

ABSTRACT: This publication aims to provide guidelines of the knowledge required and the potential research to be conducted in order to understand the mode of action of antimethanogenic feed additives (AMFA). In the first part of the paper, we classify AMFA into 4 categories according to their mode of action: (1) lowering dihydrogen (H2) production; (2) inhibiting methanogens; (3) promoting alternative H2-incorporating pathways; and (4) oxidizing methane (CH4). The second part of the paper presents questions that guide the research to identify the mode of action of an AMFA on the rumen CH4 production from 5 different perspectives: (1) microbiology; (2) cell and molecular biochemistry; (3) microbial ecology; (4) animal metabolism; and (5) cross-cutting aspects. Recommendations are provided to address various research questions within each perspective, along with examples of how aspects of the mode of action of AMFA have been elucidated before. In summary, this paper offers timely and comprehensive guidelines to better understand and reveal the mode of action of current and emerging AMFA.

Monika Toneva, Nikola Kostadinov, Zhani Yanev et al.

<b>Background/Objectives</b>: Acetaminophen (paracetamol or APAP) overdose is a major cause of acute liver injury mediated by oxidative stress, inflammation, and hepatocellular necrosis. The present study investigates the in vivo hepatoprotective potential of morin (M), lignin nanoparticles (LN), and morin-encapsulated lignin nanoparticles (LMN) against APAP-induced hepatotoxicity in mice. The specific goal was to determine whether LMN could strengthen hepatic antioxidant and anti-inflammatory defenses prior to toxic insult, which aligns with a prophylactic model rather than a post-injury clinical rescue approach. This study was guided by the primary hypothesis that LMN pretreatment would markedly reduce APAP-induced hepatic injury. <b>Methods</b>: Experimental groups included control, APAP, M, LN, LMN, M+APAP, LN+APAP, and LMN+APAP treatments. Serum hepatic biomarkers, oxidative stress parameters, and inflammatory cytokines were analyzed to assess protective responses. <b>Results</b>: APAP exposure markedly elevated aspartate aminotransferase (AST) and alkaline phosphatase (ALP) levels, indicating severe hepatic dysfunction, accompanied by increased lipid peroxidation and pro-inflammatory cytokine production. LMN+APAP treatment significantly restored hepatic enzyme levels to approximately normal values and suppressed malondialdehyde (MDA) formation, while enhancing superoxide dismutase (SOD), catalase (CAT), and glutathione peroxidase (GPx) activities. LMN also downregulated interleukin 6 (IL-6), tumor necrosis factor α (TNF-α), and interleukin 1β (IL-1β), while upregulating interleukin 10 (IL-10), suggesting effective attenuation of inflammatory signaling. Correlation analyses demonstrated positive interactions between MDA, cytokines, and hepatic enzymes, whereas antioxidant enzyme levels were inversely correlated with liver injury markers. Histopathological analysis revealed that treatment with LMN enhanced hepatoprotection, demonstrating predominantly mild, reversible lesions and suggesting a synergistic antioxidant and immunomodulatory effect. <b>Conclusions</b>: It could be concluded that LMN provided superior hepatoprotection compared to M or LN. These findings establish LMN as a promising bio-based nanotherapeutic agent for mitigating drug-induced hepatotoxicity through coordinated antioxidant and anti-inflammatory mechanisms.

Nesma Maher, Gehad Ezzat, Samah Darwish et al.

Seafood is a known reservoir for Vibrio species, some of which carry virulence factors and can be acquired via handling and cross-contamination in the retail markets, which may pose public health risks. This study investigated the prevalence, virulence-associated genes, and spoilage effects of Vibrio spp. isolated from 400 samples from eight seafood species collected in Egypt. Microbiological and molecular identification were conducted using conventional methods and confirmatory PCR assays targeting species-specific and virulence-associated genes. Vibrio spp. were detected in 51.75% of the samples, from which 66.6% were identified as virulent species: V. parahaemolyticus (33.3%), V. cholerae (22.2%), and V. vulnificus (11.1%). Gene screening revealed the presence of toxR and vvh in all V. parahaemolyticus and V. vulnificus isolates, while ctx and hylA were found in 60% and 10% of V. cholerae isolates, respectively. Quality assessments of contaminated seafood showed elevated levels of total volatile base nitrogen (TVB-N) and moderate changes in pH, indicating spoilage. Thiobarbituric acid reactive substances (TBA-RS) values remained within acceptable limits, suggesting limited lipid oxidation. These findings bring to light the public health and quality concerns associated with Vibrio-contaminated seafood and emphasize the need for routine monitoring and comprehensive risk assessments.

Jingyang Ke, Feiyang Wu, Jiyi Wang et al.

Traditional approaches to studying decision-making in neuroscience focus on simplified behavioral tasks where animals perform repetitive, stereotyped actions to receive explicit rewards. While informative, these methods constrain our understanding of decision-making to short timescale behaviors driven by explicit goals. In natural environments, animals exhibit more complex, long-term behaviors driven by intrinsic motivations that are often unobservable. Recent works in time-varying inverse reinforcement learning (IRL) aim to capture shifting motivations in long-term, freely moving behaviors. However, a crucial challenge remains: animals make decisions based on their history, not just their current state. To address this, we introduce SWIRL (SWitching IRL), a novel framework that extends traditional IRL by incorporating time-varying, history-dependent reward functions. SWIRL models long behavioral sequences as transitions between short-term decision-making processes, each governed by a unique reward function. SWIRL incorporates biologically plausible history dependency to capture how past decisions and environmental contexts shape behavior, offering a more accurate description of animal decision-making. We apply SWIRL to simulated and real-world animal behavior datasets and show that it outperforms models lacking history dependency, both quantitatively and qualitatively. This work presents the first IRL model to incorporate history-dependent policies and rewards to advance our understanding of complex, naturalistic decision-making in animals.

Thi Thu Thuy Nguyen, Duc Thanh Nguyen

Smart data selection is becoming increasingly important in data-driven machine learning. Active learning offers a promising solution by allowing machine learning models to be effectively trained with optimal data including the most informative samples from large datasets. Wildlife data captured by camera traps are excessive in volume, requiring tremendous effort in data labelling and animal detection models training. Therefore, applying active learning to optimise the amount of labelled data would be a great aid in enabling automated wildlife monitoring and conservation. However, existing active learning techniques require that a machine learning model (i.e., an object detector) be fully accessible, limiting the applicability of the techniques. In this paper, we propose a model-agnostic active learning approach for detection of animals captured by camera traps. Our approach integrates uncertainty and diversity quantities of samples at both the object-based and image-based levels into the active learning sample selection process. We validate our approach in a benchmark animal dataset. Experimental results demonstrate that, using only 30% of the training data selected by our approach, a state-of-the-art animal detector can achieve a performance of equal or greater than that with the use of the complete training dataset.

Magdalena Materniak-Kornas, Katarzyna Piórkowska, Katarzyna Ropka-Molik et al.

Small ruminant lentiviruses (SRLVs) infect sheep, causing a multiorganic disease called maedi-visna or ovine progressive pneumonia, which significantly affects the production and welfare of sheep, generating serious economic losses. Although not all infected animals develop fully symptomatic disease, they constantly spread the virus in the flock. Since the infection is incurable and no vaccine is available, another approach is necessary to control SRLV infections. In recent years, an alternative for culling infected animals has become the approach based on identifying genetic markers for selecting SRLV-resistant individuals. Recent reports revealed several candidates, including gene encoding transmembrane protein 154 (TMEM154). Several single nucleotide polymorphisms (SNPs) are found within this gene in sheep of different breeds, but only some can be considered as resistant markers. This study aimed to investigate the presence of single polymorphic sites in TMEM154 gene in sheep of selected Polish flocks and assess their association with the infection and proviral load in the context of susceptibility to SRLV infection. In total 107 sheep, representing three breeds, were screened for their SRLV infection status by serological and PCR testing. All these animals were also genotyped to characterize the presence of SNPs in TMEM154 gene and estimate their potential of being the SRLV-resistance marker. The frequency of identified alleles differed among breeds. Moreover, the positive association between TMEM154 genotype and SRLV status was found for E35K polymorphism and two polymorphic sites in 5′UTR in one of analyzed breed. However, when the relationship between SNPs and SRLV proviral load was analyzed, five had a strong association, considering the whole population of tested sheep. Presented data, for the first time, identified the presence of SNPs in TMEM154 gene in sheep housed in Polish flocks and suggested that selecting SRLV-resistant animals based on this analysis might be possible, but further validation in a larger group of sheep is required.

Fabrizia Valleriani, Andrea Polci, Federica Iapaolo et al.

West Nile virus (WNV) (Flaviviridae, Flavivirus) infection is a mosquito-borne zoonosis able of causing disease and death in humans and animals. Over the past decade, WNV infections have been a significant public health concern in Europe, and Italy has been among the most affected countries since 2008. The 2022 vector season has been characterized by an intense and early circulation of WNV. This report describes cases of co-circulation of WNV L1 and of WNV L2 occurring at the end of the 2022 vector season in Sicily and Tuscany, regions where no strains had ever been sequenced. The phylogenetic analysis of the detected strains confirmed the peculiar WNV scenario that has characterized the Italian West Nile disease (WND) epidemic since its appearance. The circulation observed in Tuscany was in fact a consequence of the spread of endemic strains to new areas while the Sicilian episodes were linked to new introductions of WNV L1 and L2 strains likely from other European countries.

Aml Ibrahim, Ola Hegab, Neveen Soliman

Milk rice pudding (MRP) is a commercial and popular dairy dessert, but owing to its characteristics and valuable ingredients, it may be contaminated by many pathogenic and spoilage microorganisms. So, this study aimed to improve the quality and safety of MRP by using cinnamon and clove essential oils. Concerning the evaluation of the minimum inhibitory concentration (MIC) for both oils with 0.2, 0.5, and 1% concentrations, B. cereus and B. subtilis were sensitive (+) to cinnamon and clove 0.5%, with inhibition zones of 13.3 and 14 mm for cinnamon and 11.3 and 12 mm for clove EO, respectively. While both bacteria were very sensitive (++) to cinnamon 1% (18.8 and 19.5 mm) and clove 1% (17.3 and 18.7 mm), respectively. Therefore, MRP was prepared by adding cinnamon and clove EOs at 0.6%. Treatments containing EOs showed a significant reduction of tested microorganisms compared to controls. B. cereus wasn’t detected in clove and cinnamon EO treatments at day 21 of the storage period, while B. subtilis vanished on day 14 for the cinnamon treatment and on day 21 for clove MRP. Moreover, the results revealed the enhancement of sensory characteristics of MRP supplemented with EOs without any significant alteration in their pH values. This study recommends the addition of cinnamon and clove EOs (0.6%) to MRP, as it isn’t only an excellent substitution of chemical preservatives with powerful antibacterial efficiency but also improves the overall acceptance of the product.

Mengrui Li, Stephanie Vanegas, Mia R. Gonzalgo et al.

We report a subcutaneous mastocytoma in a mid-aged Italian greyhound dog with a small 41 bp genomic deletion of the c-kit gene leading to skipping of the authentic 3′-splice junction of intron 10. The shift to an alternative splice junction in exon 11 leads to a mis-spliced in-frame mRNA transcript with a 27 bp deletion of exon 11 coding for 9 amino acids in the juxtamembrane negative regulatory domain of c-kit tyrosine kinase. In the tumor, c-kit was activated as revealed by more pronounced c-kit-regulated signaling by the PI3K/Akt and G-coupled receptor pathways. The same 9 amino acids deletion was reported in several human gastrointestinal stromal tumors (GIST) pointing to a remarkable similarity of c-kit activation by small deletions and aberrant splicing in humans and dogs, independent of exact sequence context, tumor type and location. Interestingly, the alternative splice junction in exon 11 has been conserved in Primates but less in other Orders with increased body temperature such as ruminants. We hypothesize that elevated body temperature has acted as evolutionary pressure to eliminate the alternative splice site at this hotspot. At a molecular level, hyperthermia may increase the frequency of small deletions in the c-kit gene by facilitating base slipping or hindering repair. An RT-qPCR assay was developed to detect c-kit alternative splicing in tumors and cell lines exposed to hyperthermia. The molecular mechanisms of tumorigenesis are discussed.

Marian Eskander, Salma El Samannoudy, Aya Hendawy et al.

Heat stress is Heat stress is a  life-threatening condition with a detrimental impact on the physiological functions of both humans and animals. The present study aimed to investigate the effect of date palm fruit on physiological functions and bone density and to document the protective effect of date palm in mitigating the negative impact of heat stress in rats. Thirty-two mature male Sprague Dawley rats weighing 170–200 g were randomly divided into four groups (8 rats/group): group 1 (control) was provided with standard diet pellets; group 2 (heat stress) rats received a standard diet and were exposed to artificial heat stress (43°C for 60 min/day); group 3 (date palm fruit) rats were given date palm fruit at a dose of 1 g/kg body weight; and group 4 (date palm fruit + heat stress) rats received date palm fruit with the same dose and were exposed to the same protocol of heat stress. Diet protocols started from the beginning of the experiment and continued till the end of the study at 2 months. Heat stress was induced (groups 2 and 4) daily during the second month of the experiment. Hematobiochemical and oxidative stress parameters, histopathological examination of the liver, kidney, and adrenal gland, and quantitative evaluation of radiographic bone density were evaluated at the end of the study. Results demonstrated heat stress resulted in significantly increased leucocyte count, decreased RBCs, platelet count, and Hb concentration (P< 0.01). Significant increases in ALT, AST, ALP, urea and creatinine levels (P < 0.01) with concurrent histopathological changes in the liver and kidney were also recorded in heat-stressed rats. Oxidative stress biomarkers, glucocorticoids, were increased with heat stress (P< 0.01). Serum calcium level and radiographic bone density were significantly decreased in the heat stress group. Rats supplemented with date palm and subjected to heat stress exhibited an insignificant change in physiologic parameters (P> 0.05) compared to control rats. In conclusion, date palm fruit demonstrated a protective effect against the deleterious changes of heat stress in rats by maintaining physiological parameters and improving bone turnover. Dietary supplementation with date palm prior to exposure to heat stress is safe and effective in protecting against the life-threatening adverse effects of heat stress.

Jenna Kline, Austin O'Quinn, Tanya Berger-Wolf et al.

Platforms that run artificial intelligence (AI) pipelines on edge computing resources are transforming the fields of animal ecology and biodiversity, enabling novel wildlife studies in animals' natural habitats. With emerging remote sensing hardware, e.g., camera traps and drones, and sophisticated AI models in situ, edge computing will be more significant in future AI-driven animal ecology (ADAE) studies. However, the study's objectives, the species of interest, its behaviors, range, habitat, and camera placement affect the demand for edge resources at runtime. If edge resources are under-provisioned, studies can miss opportunities to adapt the settings of camera traps and drones to improve the quality and relevance of captured data. This paper presents salient features of ADAE studies that can be used to model latency, throughput objectives, and provision edge resources. Drawing from studies that span over fifty animal species, four geographic locations, and multiple remote sensing methods, we characterized common patterns in ADAE studies, revealing increasingly complex workflows involving various computer vision tasks with strict service level objectives (SLO). ADAE workflow demands will soon exceed individual edge devices' compute and memory resources, requiring multiple networked edge devices to meet performance demands. We developed a framework to scale traces from prior studies and replay them offline on representative edge platforms, allowing us to capture throughput and latency data across edge configurations. We used the data to calibrate queuing and machine learning models that predict performance on unseen edge configurations, achieving errors as low as 19%.

Mubin Ul Haque, Joel Janek Dabrowski, Rebecca M. Rogers et al.

Detecting flying animals (e.g., birds, bats, and insects) using weather radar helps gain insights into animal movement and migration patterns, aids in management efforts (such as biosecurity) and enhances our understanding of the ecosystem.The conventional approach to detecting animals in weather radar involves thresholding: defining and applying thresholds for the radar variables, based on expert opinion. More recently, Deep Learning approaches have been shown to provide improved performance in detection. However, obtaining sufficient labelled weather radar data for flying animals to build learning-based models is time-consuming and labor-intensive. To address the challenge of data labelling, we propose a self-supervised learning method for detecting animal movement. In our proposed method, we pre-train our model on a large dataset with noisy labels produced by a threshold approach. The key advantage is that the pre-trained dataset size is limited only by the number of radar images available. We then fine-tune the model on a small human-labelled dataset. Our experiments on Australian weather radar data for waterbird segmentation show that the proposed method outperforms the current state-of-the art approach by 43.53% in the dice co-efficient statistic.

Yingxue Yu, Vidit Vidit, Andrey Davydov et al.

Animal Re-ID is crucial for wildlife conservation, yet it faces unique challenges compared to person Re-ID. First, the scarcity and lack of diversity in datasets lead to background-biased models. Second, animal Re-ID depends on subtle, species-specific cues, further complicated by variations in pose, background, and lighting. This study addresses background biases by proposing a method to systematically remove backgrounds in both training and evaluation phases. And unlike prior works that depend on pose annotations, our approach utilizes an unsupervised technique for feature alignment across body parts and pose variations, enhancing practicality. Our method achieves superior results on three key animal Re-ID datasets: ATRW, YakReID-103, and ELPephants.

Qianyi Deng, Oishi Deb, Amir Patel et al.

Animal pose estimation (APE) aims to locate the animal body parts using a diverse array of sensor and modality inputs (e.g. RGB cameras, LiDAR, infrared, IMU, acoustic and language cues), which is crucial for research across neuroscience, biomechanics, and veterinary medicine. By evaluating 176 papers since 2011, APE methods are categorised by their input sensor and modality types, output forms, learning paradigms, experimental setup, and application domains, presenting detailed analyses of current trends, challenges, and future directions in single- and multi-modality APE systems. The analysis also highlights the transition between human and animal pose estimation, and how innovations in APE can reciprocally enrich human pose estimation and the broader machine learning paradigm. Additionally, 2D and 3D APE datasets and evaluation metrics based on different sensors and modalities are provided. A regularly updated project page is provided here: https://github.com/ChennyDeng/MM-APE.

R. Ramanathan, S. Suman, C. Faustman

Appearance is an important sensory property that significantly influences consumers' perceptions of fresh meat quality. Failure to meet consumer expectations can lead to rejection of meat products, concomitant loss in value, and potential production of organic waste. Immediately after animal harvest, skeletal muscle metabolism changes from aerobic to anaerobic. However, anoxic postmortem muscle is biochemically active, and biomolecular interaction between myoglobin, mitochondria, metabolites, and lipid oxidation determines meat color. This review examines how metabolites and mitochondrial activity can influence myoglobin oxygenation and metmyoglobin reducing activity. Further, the review highlights recent research that has examined myoglobin redox dynamics, sarcoplasmic metabolite changes, and/or postmortem biochemistry.

Philip A. Sapp, Jeremy R. Townsend, Trevor O. Kirby et al.

More than 57% of US adults take dietary supplements, with the most common being daily multivitamins. Daily multivitamins are typically formulated in a pill or tablet form; however, new options, including powder form that can be mixed with water, are being utilized to increase bioavailability. While limited data are available, the theory is that multivitamin tablets must be adequately dissolved before entering the small intestine for assimilation, while powders come pre-dissolved before consumption, which theoretically ensures enhanced bioavailability. Our aim was to investigate the bioaccessibility and bioavailability of minerals (magnesium (Mg), zinc (Zn), calcium (Ca), and potassium (K)) using a novel foundational nutrition supplement called AG1 compared to a multivitamin tablet. AG1 contains vitamins and minerals comparable to multivitamin tablets but also includes prebiotics, probiotics, and phytonutrients. We employed the adapted Simulator of Human Intestinal Microbial Ecosystem (SHIME^®) model to assess the bioaccessibility and bioavailability of this study’s products using a simulated stomach and small intestine physiological environment equipped with a dialysis membrane (methylcellulose) to emulate absorption. Luminal contents were collected at the end of the stomach, duodenum, and 1-, 2-, and 3 h after small intestine absorption simulation (dialysis) to assess bioaccessibility. The dialysis solution was measured at 1-, 2-, and 3 h to assess bioavailability. A significantly higher (<i>p</i> < 0.05) percentage of the total amount of all minerals given at the baseline was present at the end of the stomach and duodenum portion for the powder form vs. the tablet. There was a significantly higher % maximal concentration (C_max), bioavailability, and bioaccessibility for Mg, Ca, and Zn for AG1 vs. the tablet. These preclinical data demonstrate that a greater proportion of minerals in AG1 enter the small intestine, have a higher C_max, and several are more bioaccessible and bioavailable than a tablet multivitamin in vitro.

Harry F. Mbacham, Diange M Mosume, Tobias O. Apinjoh et al.

Abstract Background Plasmodium falciparum resistance to intermittent preventive treatment with sulfadoxine-pyrimethamine (IPTp-SP) continues to spread throughout sub-Saharan Africa. This study assessed the occurrence of microscopic and sub-microscopic P. falciparum parasitaemia, dihydropteroate synthase mutations associated with resistance to SP and maternal anaemia in the Mount Cameroon area. Methods Consenting pregnant women living in semi-rural and semi-urban/urbanized settings were enrolled in this cross-sectional study. Socio-demographic, antenatal and clinical data were documented. Microscopic and sub-microscopic parasitaemia were diagnosed using peripheral blood microscopy and nested polymerase chain reaction (PCR) respectively. The dhps mutations were genotyped by restriction fragment length polymorphism analysis. The presence of A437G, K540E, and A581G was considered a marker for high-level resistance. Haemoglobin levels and anaemia status were determined. Results Among the women, the prevalence of microscopic and sub-microscopic P. falciparum infection were 7.7% (67/874) and 18.6% (93/500) respectively. Predictors of microscopic infection were younger age (< 21 years) (AOR = 2.89; 95% CI 1.29–6.46) and semi-rural settings (AOR = 2.27; 95% CI 1.31–3.96). Determinants of sub-microscopic infection were the rainy season (AOR, 3.01; 95% CI 1.77–5.13), primigravidity (AOR = 0.45; 95% CI 0.21–0.94) and regular ITN usage (AOR = 0.49; 95% CI 0.27–0.90). Of the145 P. falciparum isolates genotyped, 66.9% (97) carried mutations associated with resistance to SP; 33.8% (49), 0%, 52.4% (76) and 19.3% (28) for A437G, K540E, A581G and A437G + A581G respectively. The A581G mutation was associated with ≥ 3 SP doses evident only among sub-microscopic parasitaemia (P = 0.027) and multigravidae (P = 0.009). Women with microscopic infection were more likely from semi-rural settings (AOR = 7.09; 95% CI 2.59–19.42), to report history of fever (AOR = 2.6; 95% CI 1.07–6.31), to harbour parasites with double resistant mutations (AOR = 6.65; 95% CI 1.85–23.96) and were less likely to have received 2 SP doses (AOR = 0.29; 95% CI 1.07–6.31). Microscopic infection decreased Hb levels more than sub-microscopic infection. Conclusion The occurrence of sub-microscopic P. falciparum parasites resistant to SP and intense malaria transmission poses persistent risk of malaria infection during pregnancy in the area. ITN usage and monitoring spread of resistance are critical.

Michael Perez, Corey Toler-Franklin

Classifying the behavior of humans or animals from videos is important in biomedical fields for understanding brain function and response to stimuli. Action recognition, classifying activities performed by one or more subjects in a trimmed video, forms the basis of many of these techniques. Deep learning models for human action recognition have progressed significantly over the last decade. Recently, there is an increased interest in research that incorporates deep learning-based action recognition for animal behavior classification. However, human action recognition methods are more developed. This survey presents an overview of human action recognition and pose estimation methods that are based on convolutional neural network (CNN) architectures and have been adapted for animal behavior classification in neuroscience. Pose estimation, estimating joint positions from an image frame, is included because it is often applied before classifying animal behavior. First, we provide foundational information on algorithms that learn spatiotemporal features through 2D, two-stream, and 3D CNNs. We explore motivating factors that determine optimizers, loss functions and training procedures, and compare their performance on benchmark datasets. Next, we review animal behavior frameworks that use or build upon these methods, organized by the level of supervision they require. Our discussion is uniquely focused on the technical evolution of the underlying CNN models and their architectural adaptations (which we illustrate), rather than their usability in a neuroscience lab. We conclude by discussing open research problems, and possible research directions. Our survey is designed to be a resource for researchers developing fully unsupervised animal behavior classification systems of which there are only a few examples in the literature.