scholarly journals OptiFlex: Multi-Frame Animal Pose Estimation Combining Deep Learning With Optical Flow

2021 ◽  
Vol 15 ◽  
Author(s):  
XiaoLe Liu ◽  
Si-yang Yu ◽  
Nico A. Flierman ◽  
Sebastián Loyola ◽  
Maarten Kamermans ◽  
...  
Keyword(s):  
Deep Learning ◽  
Pose Estimation ◽  
Body Shape ◽  
Animal Species ◽  
Fruit Fly ◽  
Four Dimensions ◽  
Animal Body ◽  
Video Frames ◽  
Wide Range ◽  
Evaluation Metric

Animal pose estimation tools based on deep learning have greatly improved animal behaviour quantification. These tools perform pose estimation on individual video frames, but do not account for variability of animal body shape in their prediction and evaluation. Here, we introduce a novel multi-frame animal pose estimation framework, referred to as OptiFlex. This framework integrates a flexible base model (i.e., FlexibleBaseline), which accounts for variability in animal body shape, with an OpticalFlow model that incorporates temporal context from nearby video frames. Pose estimation can be optimised using multi-view information to leverage all four dimensions (3D space and time). We evaluate FlexibleBaseline using datasets of four different lab animal species (mouse, fruit fly, zebrafish, and monkey) and introduce an intuitive evaluation metric—adjusted percentage of correct key points (aPCK). Our analyses show that OptiFlex provides prediction accuracy that outperforms current deep learning based tools, highlighting its potential for studying a wide range of behaviours across different animal species.

scholarly journals OptiFlex: video-based animal pose estimation using deep learning enhanced by optical flow

2020 ◽  
Author(s):  
XiaoLe Liu ◽  
Si-yang Yu ◽  
Nico Flierman ◽  
Sebastian Loyola ◽  
Maarten Kamermans ◽  
...  
Keyword(s):  
Deep Learning ◽  
Optical Flow ◽  
Pose Estimation ◽  
Body Shape ◽  
Fruit Fly ◽  
Four Dimensions ◽  
Animal Body ◽  
Video Frames ◽  
Wide Range ◽  
Evaluation Metric

AbstractDeep learning based animal pose estimation tools have greatly improved animal behaviour quantification. However, those tools all make predictions on individual video frames and do not account for variability of animal body shape in their model designs. Here, we introduce the first video-based animal pose estimation architecture, referred to as OptiFlex, which integrates a flexible base model to account for variability in animal body shape with an optical flow model to incorporate temporal context from nearby video frames. This approach can be combined with multi-view information, generating prediction enhancement using all four dimensions (3D space and time). To evaluate OptiFlex, we adopted datasets of four different lab animal species (mouse, fruit fly, zebrafish, and monkey) and proposed a more intuitive evaluation metric - percentage of correct key points (aPCK). Our evaluations show that OptiFlex provides the best prediction accuracy amongst current deep learning based tools, and that it can be readily applied to analyse a wide range of behaviours.

scholarly journals Antimicrobial Usage Surveillance Through Sales at Veterinary Drug Shops Intended for Livestock in Vietnam

2021 ◽  
Vol 5 ◽  
Author(s):  
Le Thi Thu Ha ◽  
Chalalai Rueanghiran ◽  
Nguyen Thi Huong Giang ◽  
Doan Phuong Thuy ◽  
Doan Hoang Phu ◽  
...  
Keyword(s):  
Body Weight ◽  
Data Collection ◽  
Animal Species ◽  
Veterinary Drug ◽  
Surveillance Systems ◽  
Middle Income ◽  
Animal Body ◽  
Country Level ◽  
Drug Shops ◽  
Wide Range

There is a pressing need to establish surveillance systems for antimicrobial use (AMU) intended for animal production particularly in many low- and middle-income countries. This is an extremely challenging task, notably due to the wide range of animal species, production types and antimicrobials available in the market. In Vietnam, farmers commonly buy antimicrobials from veterinary drug shops. Therefore, veterinary drug shops are a potential target for data collection on AMU. We collected antimicrobial sales data at veterinary drug shops and estimated the amount of AMU in different animal species by antimicrobial active ingredient (AAI) class using different measurement metrics. We compiled information on all antimicrobials licensed in Vietnam and used this information to develop a mobile application to capture sales of antimicrobials intended for use in poultry, pig, and ruminant. We provided tablets with this application to 60 veterinary drug shops in two provinces of the country (Bac Giang in the north, Dong Thap in the south; three districts and 30 shops per province) for data collection over 3 weeks. Total sales of antimicrobials were extrapolated to 1 year, and these amounts were related to three different denominator estimates in each province including standing animal body weight, animal biomass, and Population Correction Unit (PCU). A total of 3,960 transactions [2,577 (median 75.5 per shop) in Bac Giang; 1,383 (median 28.5 per shop) in Dong Thap] of 831 different antimicrobial-containing products were recorded in the 3-week period. Sales of 57 AAIs belonging to 17 classes were recorded. In the three Bac Giang districts, we estimated that 242.0 kg of AAI were hypothetically sold over 1 year. Of those, 202.2 kg (83.6%) were intended for poultry, 19.8 kg (8.1%) for pigs, and 20.0 kg (8.3%) for ruminants. In Dong Thap, an estimated 48.4 kg of antimicrobials were sold, including 28.9 kg (59.7%) for poultry, 16.0 kg (33.1%) for pigs, and 3.5 kg (7.2%) for ruminants. After standardized by different animal population denominators, AMU in Bac Giang amounted to 1129.2 mg/kg standing animal body weight, 480.2 mg/kg biomass, and 636.1 mg/kg PCU. In Dong Thap, AMU figures were 1211.0 mg/kg standing animal body weight, 595.8 mg/kg biomass and 818.5 mg/kg PCU. We discuss the observed differences between species, location and metrics, as well as the potential advantages and limitations (including potential sources of bias) of this methodology and its applicability at country level. Retail level data collection can effectively be integrated into AMU surveillance systems that help identify priority AMU management areas (species, regions, and antimicrobial classes), establish national benchmarks and reduction targets.

Registration of animal rabies in Krasnodar region

2020 ◽  
pp. 3-4
Author(s):  
Oleg Yu. Chernykh ◽  
◽  
Vadim A. Bobrov ◽  
Sergey N. Zabashta ◽  
Roman A. Krivonos ◽  
...  
Keyword(s):  
Urban Form ◽  
Animal Species ◽  
Farm Animals ◽  
Natural Focus ◽  
Dead End ◽  
Wide Range ◽  
Epizootic Process ◽  
Krasnodar Region ◽  
Spread Of Infection ◽  
Reporting Data

Rabies remains a constant threat to humanity in many parts of the world. At the same time, scientifically grounded antiepizootic measures should be based on the peculiarities of the regional epizootology of this zooanthroponosis. The authors studied the epizootological and statistical reporting data of the Kropotkin Regional Veterinary Laboratory, presented an analysis of the registration of rabies in animals in Krasnodar region. From the obtained data, it should be noted that despite the wide range of animals involved in the epizootic process of rabies infection in Krasnodar region, dogs, cats and foxes play a major role in the reservation and spread of infection, which account for 78.6. Of the total number of registered cases, 15.5% falls on foxes, that indicates the natural focus of the disease, along with the manifestation of the disease in an urban form. At the same time, stray and neglected dogs and cats, which occupy a significant place among the total number of sick animals, are also sources and spread of the infection. Thus farm animals (8.3% of the total number of infected animals) are a biological dead end for the infection. Isolated cases of the disease were noted in muskrat, donkey, raccoon, raccoon dog, marten, ferret and jackal. The authors also established the specific morbidity of various animal species with rabies infection, that is an important aspect in the development and implementation of antiepizootic measures complex

Cultural Accommodation of Internet-Based Interventions for Substance Use and Related Disorders: A Proposed Comprehensive Framework (Preprint)

2019 ◽  
Author(s):  
Keren Gueta ◽  
Yossi Harel-Fisch ◽  
Sophie D. Walsh
Keyword(s):  
Substance Use ◽  
Ecological Validity ◽  
Limited Attention ◽  
Diverse Range ◽  
Four Dimensions ◽  
Service Engagement ◽  
Wide Range ◽  
Cultural Accommodation ◽  
Abuse Treatment ◽  
Comprehensive Framework

BACKGROUND Despite the low utilization rates of substance use and related disorders services, and the ability of internet-based interventions for substance use and related disorders (IBIS) to address challenges related to service engagement, limited attention has been placed on the processes for the accommodation of these interventions to diverse cultural settings. OBJECTIVE The purpose of this study was to develop a conceptual framework for the cultural accommodation of IBIS across populations, settings, and countries. METHODS A pilot study of cultural accommodation of an existing internet intervention for alcohol use (Down Your Drink (DYD)), focus groups and daily online surveys of prospective consumers (N=24) and interviews with experts (N=7) in the substance abuse treatment field were conducted. RESULTS Thematic analysis revealed a wide range of themes identified as needing to be addressed in the process of DYD accommodation. It also emphasized that accommodation needs to incorporate both technical and contents themes, shaped by both the general Israeli cultural as well as by the specific Israeli drinking subculture. A combined mixed emic–etic theoretical approach incorporating the pilot findings together with a scoping literature review was employed to develop a framework for cultural accommodation of IBIS. A comprehensive framework for cultural accommodation of IBIS is introduced consisting of five chronological stages of IBIS accommodation and four dimensions of accommodation. CONCLUSIONS The proposed framework can serve as a guide for the cultural accommodation of existing IBIS across a diverse range of cultural and geographical settings thus augmenting the ecological validity of IBIS and reducing health disparities worldwide.

scholarly journals A Generalization Performance Study Using Deep Learning Networks in Embedded Systems

Sensors ◽  
2021 ◽  
Vol 21 (4) ◽  
pp. 1031
Author(s):  
Joseba Gorospe ◽  
Rubén Mulero ◽  
Olatz Arbelaitz ◽  
Javier Muguerza ◽  
Miguel Ángel Antón
Keyword(s):  
Deep Learning ◽  
Embedded Systems ◽  
Embedded System ◽  
General Purpose ◽  
Learning Networks ◽  
Performance Study ◽  
Learning Techniques ◽  
Wide Range ◽  
Learning Architectures

Deep learning techniques are being increasingly used in the scientific community as a consequence of the high computational capacity of current systems and the increase in the amount of data available as a result of the digitalisation of society in general and the industrial world in particular. In addition, the immersion of the field of edge computing, which focuses on integrating artificial intelligence as close as possible to the client, makes it possible to implement systems that act in real time without the need to transfer all of the data to centralised servers. The combination of these two concepts can lead to systems with the capacity to make correct decisions and act based on them immediately and in situ. Despite this, the low capacity of embedded systems greatly hinders this integration, so the possibility of being able to integrate them into a wide range of micro-controllers can be a great advantage. This paper contributes with the generation of an environment based on Mbed OS and TensorFlow Lite to be embedded in any general purpose embedded system, allowing the introduction of deep learning architectures. The experiments herein prove that the proposed system is competitive if compared to other commercial systems.

scholarly journals Data augmentation for computed tomography angiography via synthetic image generation and neural domain adaptation

2020 ◽  
Vol 6 (1) ◽  
Author(s):  
Malte Seemann ◽  
Lennart Bargsten ◽  
Alexander Schlaefer
Keyword(s):  
Computed Tomography ◽  
Neural Networks ◽  
Deep Learning ◽  
Medical Imaging ◽  
Computed Tomography Angiography ◽  
Data Augmentation ◽  
Domain Adaptation ◽  
Synthetic Image ◽  
Wide Range ◽  
The Impact

AbstractDeep learning methods produce promising results when applied to a wide range of medical imaging tasks, including segmentation of artery lumen in computed tomography angiography (CTA) data. However, to perform sufficiently, neural networks have to be trained on large amounts of high quality annotated data. In the realm of medical imaging, annotations are not only quite scarce but also often not entirely reliable. To tackle both challenges, we developed a two-step approach for generating realistic synthetic CTA data for the purpose of data augmentation. In the first step moderately realistic images are generated in a purely numerical fashion. In the second step these images are improved by applying neural domain adaptation. We evaluated the impact of synthetic data on lumen segmentation via convolutional neural networks (CNNs) by comparing resulting performances. Improvements of up to 5% in terms of Dice coefficient and 20% for Hausdorff distance represent a proof of concept that the proposed augmentation procedure can be used to enhance deep learning-based segmentation for artery lumen in CTA images.

scholarly journals Uncertainty-Aware Deep Learning-Based Cardiac Arrhythmias Classification Model of Electrocardiogram Signals

Computers ◽  
2021 ◽  
Vol 10 (6) ◽  
pp. 82
Author(s):  
Ahmad O. Aseeri
Keyword(s):  
Deep Learning ◽  
Cardiac Arrhythmias ◽  
Large Scale ◽  
Clinical Decision Making ◽  
Probabilistic Approach ◽  
Classification Model ◽  
Gating Mechanism ◽  
Uncertainty Estimates ◽  
Wide Range

Deep Learning-based methods have emerged to be one of the most effective and practical solutions in a wide range of medical problems, including the diagnosis of cardiac arrhythmias. A critical step to a precocious diagnosis in many heart dysfunctions diseases starts with the accurate detection and classification of cardiac arrhythmias, which can be achieved via electrocardiograms (ECGs). Motivated by the desire to enhance conventional clinical methods in diagnosing cardiac arrhythmias, we introduce an uncertainty-aware deep learning-based predictive model design for accurate large-scale classification of cardiac arrhythmias successfully trained and evaluated using three benchmark medical datasets. In addition, considering that the quantification of uncertainty estimates is vital for clinical decision-making, our method incorporates a probabilistic approach to capture the model’s uncertainty using a Bayesian-based approximation method without introducing additional parameters or significant changes to the network’s architecture. Although many arrhythmias classification solutions with various ECG feature engineering techniques have been reported in the literature, the introduced AI-based probabilistic-enabled method in this paper outperforms the results of existing methods in outstanding multiclass classification results that manifest F1 scores of 98.62% and 96.73% with (MIT-BIH) dataset of 20 annotations, and 99.23% and 96.94% with (INCART) dataset of eight annotations, and 97.25% and 96.73% with (BIDMC) dataset of six annotations, for the deep ensemble and probabilistic mode, respectively. We demonstrate our method’s high-performing and statistical reliability results in numerical experiments on the language modeling using the gating mechanism of Recurrent Neural Networks.

scholarly journals 6D pose estimation with combined deep learning and 3D vision techniques for a fast and accurate object grasping

2021 ◽  
pp. 103775
Author(s):  
Tuan-Tang Le ◽  
Trung-Son Le ◽  
Yu-Ru Chen ◽  
Joel Vidal ◽  
Chyi-Yeu Lin
Keyword(s):  
Deep Learning ◽  
Pose Estimation ◽  
3D Vision
Sign in / Sign up

Export Citation Format

Share Document