scholarly journals Mice tracking using the YOLO algorithm

2019 ◽  
Author(s):  
Helton Maia Peixoto ◽  
Richardson Santiago Teles ◽  
John Victor Alves Luiz ◽  
Aron Miranda Henriques-Alves ◽  
Rossana Moreno Santa Cruz
Keyword(s):  
New Technologies ◽  
Good Alternative ◽  
Behavioral Neuroscience ◽  
Experimental Designs ◽  
Training Set ◽  
Computational Tools ◽  
Common System ◽  
Neuroscience Research ◽  
The Mean ◽  
System Errors

The development of computational tools is essential for the development of new technologies, including experimental designs needed for behavioral neuroscience research. The computational tool developed in this study is based on the convolutional neural networks and the You Only Look Once (YOLO) algorithm for detecting and tracking mice in videos recorded during behavioral neuroscience experiments. The task of mice detection consists of determining the location in the image where the animals are present, for each frame acquired. In this work, we propose mice tracking using the YOLO algorithm, running on an NVIDIA GeForce GTX 1060 GPU. We analyzed a set of data composed of 13622 images, made up of behavioral videos of three important researches in this area. The training set used 50% of the images, 25% for validation and 25% for the tests. The results show that the mean Average Precision (mAP) reached by the developed system was 90.79% and 90.75% for the Full and Tiny versions of YOLO, respectively. It has also been found that the use of the Tiny version is a good alternative for experimental designs that require real-time response. Considering the high accuracy of the results, the developed work allows the experimentalists to perform mice tracking in a reliable and non-evasive way, avoiding common system errors that require delimitations of regions of interest (ROI) or even evasive luminous identifiers such as LED for tracking the animals.

scholarly journals Mice tracking using the YOLO algorithm

2019 ◽  
Author(s):  
Helton Maia Peixoto ◽  
Richardson Santiago Teles ◽  
John Victor Alves Luiz ◽  
Aron Miranda Henriques-Alves ◽  
Rossana Moreno Santa Cruz
Keyword(s):  
New Technologies ◽  
Good Alternative ◽  
Behavioral Neuroscience ◽  
Experimental Designs ◽  
Training Set ◽  
Computational Tools ◽  
Common System ◽  
Neuroscience Research ◽  
The Mean ◽  
System Errors

The development of computational tools is essential for the development of new technologies, including experimental designs needed for behavioral neuroscience research. The computational tool developed in this study is based on the convolutional neural networks and the You Only Look Once (YOLO) algorithm for detecting and tracking mice in videos recorded during behavioral neuroscience experiments. The task of mice detection consists of determining the location in the image where the animals are present, for each frame acquired. In this work, we propose mice tracking using the YOLO algorithm, running on an NVIDIA GeForce GTX 1060 GPU. We analyzed a set of data composed of 13622 images, made up of behavioral videos of three important researches in this area. The training set used 50% of the images, 25% for validation and 25% for the tests. The results show that the mean Average Precision (mAP) reached by the developed system was 90.79% and 90.75% for the Full and Tiny versions of YOLO, respectively. It has also been found that the use of the Tiny version is a good alternative for experimental designs that require real-time response. Considering the high accuracy of the results, the developed work allows the experimentalists to perform mice tracking in a reliable and non-evasive way, avoiding common system errors that require delimitations of regions of interest (ROI) or even evasive luminous identifiers such as LED for tracking the animals.

scholarly journals Mice Tracking Using The YOLO Algorithm

2020 ◽  
Author(s):  
Richardson Santiago Teles De Menezes ◽  
John Victor Alves Luiz ◽  
Aron Miranda Henrique-Alves ◽  
Rossana Moreno Santa Cruz ◽  
Helton Maia
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Behavioral Neuroscience ◽  
High Accuracy ◽  
Mean Average Precision ◽  
Computational Tool ◽  
Training Set ◽  
Average Precision ◽  
The Mean

The computational tool developed in this study is based on convolutional neural networks and the You Only Look Once (YOLO) algorithm for detecting and tracking mice in videos recorded during behavioral neuroscience experiments. We analyzed a set of data composed of 13622 images, made up of behavioral videos of three important researches in this area. The training set used 50% of the images, 25% for validation, and 25% for the tests. The results show that the mean Average Precision (mAP) reached by the developed system was 90.79% and 90.75% for the Full and Tiny versions of YOLO, respectively. Considering the high accuracy of the results, the developed work allows the experimentalists to perform mice tracking in a reliable and non-evasive way.

scholarly journals Geometric property-based convolutional neural network for indoor object detection

2021 ◽  
Vol 18 (1) ◽  
pp. 172988142199332
Author(s):  
Xintao Ding ◽  
Boquan Li ◽  
Jinbao Wang
Keyword(s):  
Neural Network ◽  
Object Detection ◽  
Convolutional Neural Network ◽  
Geometric Property ◽  
Ground Truth ◽  
Geometric Constraints ◽  
Depth Information ◽  
Training Set ◽  
Object Knowledge ◽  
The Mean

Indoor object detection is a very demanding and important task for robot applications. Object knowledge, such as two-dimensional (2D) shape and depth information, may be helpful for detection. In this article, we focus on region-based convolutional neural network (CNN) detector and propose a geometric property-based Faster R-CNN method (GP-Faster) for indoor object detection. GP-Faster incorporates geometric property in Faster R-CNN to improve the detection performance. In detail, we first use mesh grids that are the intersections of direct and inverse proportion functions to generate appropriate anchors for indoor objects. After the anchors are regressed to the regions of interest produced by a region proposal network (RPN-RoIs), we then use 2D geometric constraints to refine the RPN-RoIs, in which the 2D constraint of every classification is a convex hull region enclosing the width and height coordinates of the ground-truth boxes on the training set. Comparison experiments are implemented on two indoor datasets SUN2012 and NYUv2. Since the depth information is available in NYUv2, we involve depth constraints in GP-Faster and propose 3D geometric property-based Faster R-CNN (DGP-Faster) on NYUv2. The experimental results show that both GP-Faster and DGP-Faster increase the performance of the mean average precision.

scholarly journals Comparative study between fascia lata and temporalis fascia in myringoplasty

2020 ◽  
Vol 36 (1) ◽  
Author(s):  
Ayman Ali Abdel Fattah ◽  
Abdel Hay Rashad Elasy ◽  
Ahmed Helmy Hoseini ◽  
Tarek Abdel Rahman Abdel Hafez
Keyword(s):  
Tympanic Membrane ◽  
Good Alternative ◽  
Fascia Lata ◽  
The Body ◽  
Temporalis Fascia ◽  
Fascia Graft ◽  
Fascia Lata Graft ◽  
Group A ◽  
The Mean ◽  
Group B

Abstract Background Repair of a perforated tympanic membrane (myringoplasty) can facilitate normal middle ear function, resist infection, and help re-establish normal hearing. Autogenous graft materials are the most popular graft materials used in myringoplasty because of their easy acceptability by the body. This study is conducted to compare between temporalis fascia graft and fascia lata graft in myringoplasty for patients with tubo-tympanic dry perforation. Results A total of 60 patients with persistent dry tympanic membrane perforation were included in our study during the period from January 2018 to May 2020. Patients underwent myringoplasty with temporalis fascia (30 patients as group A) or fascia lata (30 patients as group B). Patients were scheduled for follow-up visits concerning graft status, ear discharge, and audiograms. The mean postoperative air-bone gap in group A was 17.5 ± 4 after 1 month and 8.6 ± 6.9 after 3 months, while in group B, the mean postoperative air-bone gap was 17.6 ± 4.9 after 1 month and 9.4 ± 7.5 after 3 months. There was 90% success in graft uptake in group A, while there was 80% success in group B. Conclusion Using temporalis fascia is still the best and most trustworthy technique of myringoplasty compared to fascia lata graft. However, fascia lata can be a good alternative to temporalis fascia especially in cases of revision myringoplasty, ears having large perforation, or near-total perforation where the chances of residual perforation are high because of the limited margin of remnant tympanic membrane overlapping the graft.

scholarly journals Comparison of Virtual and Actual Education Models on the Learning of Internal Interns During the Pandemic of COVID-19

2021 ◽  
Author(s):  
Azadeh Mottaghi ◽  
Nazanin Alibeik ◽  
Shokoufeh Savaj ◽  
Behnam Shakiba ◽  
Raheleh Alimoradzadeh ◽  
...  
Keyword(s):  
Medical Students ◽  
Good Alternative ◽  
Educational Process ◽  
Virtual Education ◽  
P Value ◽  
Educational Models ◽  
Significant Difference ◽  
Post Test ◽  
E Learning ◽  
The Mean

Abstract Background: Following the outbreak of coronavirus and its impact on the educational process of medical students, attention was paid to e-learning due to importance of education and research. The aim of present study is to evaluation of virtual and actual education models on the learning of internal interns during the pandemic of COVID-19. Results: The mean±SD age of the participants was 24.9±2.3 years, 46% of participants were men and 54% of them live in dormitory. Results of study shows that difference of pre-test and post-test score with actual and virtual education models in the male and female is significantly different (P value=0.020). Virtual education for men and actual education for women have resulted in higher difference scores in pre-test and post-test. Actual and virtual education models were no different in academic achievements of participants. Thus, the scores obtained in 9 areas did not make a significant difference between the two types of educational models.Discussion: Virtual and actual education methods for internal interns were examined and the results showed that there is no significant difference between these two methods in student output. Given the conditions that have created by COVID-19 disease in the world, it seems that the use of virtual education can be a good alternative to educating medical students so that education does not stop.

scholarly journals PR0P0F0L VS. THIOPENTAL NA

2009 ◽  
Vol 16 (04) ◽  
pp. 542-549
Author(s):  
FARNOUSH FARZI ◽  
Mona Oudi ◽  
MARZIEH MEHRAFZA ◽  
Zahra Mohammad Tabar ◽  
ALI MIRMANSOURI ◽  
...  
Keyword(s):  
Pregnancy Rate ◽  
Embryo Quality ◽  
Postoperative Nausea And Vomiting ◽  
Oocyte Retrieval ◽  
Good Alternative ◽  
Fertilization Rate ◽  
Nausea And Vomiting ◽  
Informed Consent Form ◽  
The Mean ◽  
Post Operative Nausea

O b j e c t i v e : The aim of this study was to compare the side effects, fertilization rate and pregnancy rate (PR) and duration ofrecovery between Propofol and Thiopental Na after ICSI-vaginal retrieval of oocyte in ART cycle. D e s i g n : This study was a prospective,randomized clinical trial Materials and M e t h o d s : Ninety eight ASA(American Society of Anesthesiologist) physical status I and II womenparticipating in an intracytoplasmic sperm injection) ICSI) program were assessed. All of the patients underwent general anesthesia inductionwith Propofol and Thiopental Na. The first group (49cycles) received 2-2.5mg/kg of Propofol, and the second group (49cycles) received 5mg/kgThiopental during transvaginal oocyte retrieval. An informed consent form was obtained for each patients treatment. Variables under studyincluded: female age, cause and duration of infertiIity^postoperative nausea and vomiting(PONV), heamodynamic changes, mean number ofoocyte retrieved, oocyte metaphase II, embryo cleaved, embryo transferred, embryo quality and pregnancy rate(PR)and duration recovery.Statistical analysis was carried out by using SPSS.10 software and statistical test of T-test and chi-square. R e s u l t s : The PR in Propofol groupwas 18(36.7%) and in Thiopental Na group was 19(38.8%) with no significant differences the mean duration of infertility and weight weren'tstatistically significant. The mean number of oocyte retrieved (metaphase II), embryo cleaved, embryo transferred and embryo quality weren'tsignificant between the two groups. The incidence of nausea in Propofol group in comparison with Thiopental Na group was lower withsignificant differences. The incidence of vomiting between two groups was statistically significant (46.9% vs.28.6% respectively)(P<0.05).between two groups. Duration of recovery in Propofol group was 15+/_3min and in Thiopental Na group was 25+/_5 min that was statisticallysignificant(P<0.05). C o n c l u s i o n s : Propofol offered lower incidence of post operative nausea and vomiting and a quick recovery from anesthesia without any adverse effect on pregnancy outcome. These findings showed that Propofol was a good alternative for Thiopental Na in short timeoperation, like ICSI -vaginal retrieval of oocytes.

scholarly journals Willingness to pay for improvements in rural sanitation: Evidence from a cross-sectional survey of three rural counties in Kenya

PLoS ONE ◽  
2021 ◽  
Vol 16 (5) ◽  
pp. e0248223
Author(s):  
Diana Mutuku Mulatya ◽  
Vincent Were ◽  
Joseph Olewe ◽  
Japheth Mbuvi
Keyword(s):  
Willingness To Pay ◽  
New Technologies ◽  
Contingent Valuation Method ◽  
Evaluation Research ◽  
Total Sample ◽  
Cross Sectional Survey ◽  
Cross Sectional ◽  
Poor Households ◽  
Life Years ◽  
The Mean

Poor sanitation worldwide leads to an annual loss of approximately $222.9 billion and is the second leading cause of Disability-Adjusted Life Years (DALY’s) lost due to diarrhoea. Yet in Kenya, the slow rate and levels at which the household’s access improved sanitation facilities remain a concern, and it is unknown if the cost of new technologies is a barrier to access. This study assessed the maximum willingness to pay (WTP) for SAFI and SATO sanitation products and identified those factors that affect the willingness to pay (WTP) valuation estimates by households in three counties in Kenya. It used quantitative economic evaluation research integrated within a cross-sectional survey. Contingent valuation method (CVM) was used to determine the maximum WTP for sanitation in households. We used the logistic regression model in data analysis. A total of 211 households were interviewed in each county, giving a total sample size of 633 households. The mean WTP for SAFI latrines was $153.39 per household, while the mean WTP for SATO pans and SATO stools was $11.49 and $14.77 respectively. For SAFI latrines, households in Kakamega were willing to pay $6.6 more than average while in Siaya, the households were willing to pay $5.1 less than the average. The main determinants of households WTP for the two sanitation products included household’s proximity to the toilet (p = 0.0001), household income (β = .2245741, p = 0.004), sanitation product (β = -2968.091; p = 0.004), socioeconomic status (β = -3305.728, p = 0.004) and a household’s satisfaction level with the current toilet (β = -4570.602; p = 0.0001). Increased proximity of households to the toilet, higher incomes, and providing loan facilities or subsidy to poor households could increase the demand for these sanitation technologies.

Prediction of the Maximum Number of Repetitions and Repetitions in Reserve From Barbell Velocity

2018 ◽  
Vol 13 (3) ◽  
pp. 353-359 ◽  
Author(s):  
Amador García-Ramos ◽  
Alejandro Torrejón ◽  
Belén Feriche ◽  
Antonio J. Morales-Artacho ◽  
Alejandro Pérez-Castilla ◽  
...  
Keyword(s):  
Body Mass ◽  
General Equation ◽  
Bench Press ◽  
Body Height ◽  
Strong Relationship ◽  
Training Set ◽  
Mean Velocity ◽  
The Mean ◽  
Acceptable Reliability

Purpose: To provide 2 general equations to estimate the maximum possible number of repetitions (XRM) from the mean velocity (MV) of the barbell and the MV associated with a given number of repetitions in reserve, as well as to determine the between-sessions reliability of the MV associated with each XRM. Methods: After determination of the bench-press 1-repetition maximum (1RM; 1.15 ± 0.21 kg/kg body mass), 21 men (age 23.0 ± 2.7 y, body mass 72.7 ± 8.3 kg, body height 1.77 ± 0.07 m) completed 4 sets of as many repetitions as possible against relative loads of 60%1RM, 70%1RM, 80%1RM, and 90%1RM over 2 separate sessions. The different loads were tested in a randomized order with 10 min of rest between them. All repetitions were performed at the maximum intended velocity. Results: Both the general equation to predict the XRM from the fastest MV of the set (CV = 15.8–18.5%) and the general equation to predict MV associated with a given number of repetitions in reserve (CV = 14.6–28.8%) failed to provide data with acceptable between-subjects variability. However, a strong relationship (median r2 = .984) and acceptable reliability (CV < 10% and ICC > .85) were observed between the fastest MV of the set and the XRM when considering individual data. Conclusions: These results indicate that generalized group equations are not acceptable methods for estimating the XRM–MV relationship or the number of repetitions in reserve. When attempting to estimate the XRM–MV relationship, one must use individualized relationships to objectively estimate the exact number of repetitions that can be performed in a training set.

scholarly journals Residents' Self-Perceived Errors in Transitions of Care in the Emergency Department

2011 ◽  
Vol 3 (1) ◽  
pp. 37-40 ◽  
Author(s):  
Dustin Smith ◽  
J. Wayne Burris ◽  
Guisou Mahmoud ◽  
Gregory Guldner
Keyword(s):  
Emergency Department ◽  
Emergency Medicine ◽  
Resident Physician ◽  
Transitions Of Care ◽  
Transition Of Care ◽  
Standardized Protocol ◽  
The Mean ◽  
System Errors ◽  
The Impact ◽  
Research Assistants

Abstract Background The Accreditation Council for Graduate Medical Education requirements for systems-based practice state residents are expected to participate in identifying system errors and implementing potential systems solutions. The objective of this study was to determine the numbers of perceived errors occurring from patient pass offs between resident physicians in our emergency department. Methods Using a prospective observational study, we queried emergency medicine residents about perceived errors in the transition of care using trained research assistants and a standardized protocol. Transition of care was defined as the transfer of responsibility to evaluate and treat and disposition of a patient in the emergency department from 1 resident physician to a second oncoming emergency department resident physician. Mean resident-perceived errors per shift and per patient transfer of care were calculated. Additionally, the mean number of perceived errors impacting patients was calculated. Results Emergency medicine residents on 107 shifts reported receiving 713 patients in pass off with a mean of 7 patients per physician per shift, with 40% of patients passed off needing some intervention (mean of 2.8 patients per provider per shift). Nineteen of the 107 shifts (17.8%) during which a resident took patients from a prior provider had a perceived error in at least 1 patient signed off. Of the 713 patients transitioned, the receiving physician perceived an error related to the transition of care for 23. Two of the 23 errors were determined by reviewing emergency medicine attendings to not be errors, and for 9 the receiving physician perceived an impact on the patient. All were delays in care or disposition. Conclusion Our data suggest emergency medicine residents were able to perceive errors related to transitions of care, describe the types of pass-off errors, and, to a lesser degree, describe the impact these errors have on patients.

Sign in / Sign up

Export Citation Format

Share Document