Arrumar a Home page do ethowatcher Os

[jmarcolan]

Precisamos rever o texto da home page do EthoWatcher Os

[jmarinoneto-github]

Então, eu posso começar a ajudar nestes textos.

[jmarinoneto-github]

que tal este novo texto para o inicio da página (não achei o botaozinho de editar que vc mostrou). Vai como abaixo: Introduction INTRODUCTION The EthoWatcher OS is a further development of The EthoWatcher (https://ethowatcher.paginas.ufsc.br/ ; Crispim-Junior et al, 2012) developed to support “manual” categorical (or ethographic) behavioral transcriptions (or annotations), tracking and extraction of morphologic (body size, proportions, orientation) and kinematic (displacement, velocity, angle variations) variables from digital video files of lab animals. Mounting users’ demands, suggestions, and collaboration proposals, associated to the rapid development of new machine vision algorithms and advanced collaborative platforms for open-source software development, allowed for extensive remodeling of the EthoWatcher structure and functions. Furthermore, the also increasing concerns with reproducibility of scientific results, especially those arising from models usual in behavioral preclinical research labs (Open Science Collaboration, 2015; Spruijt et al., 2014; Macleod and Mohan, 2019) imposed that any accessible and comprehensive behavioral recording software should allow also for the adoption of data quality assessment procedures and good lab practices in behavioral studies. Here we present a new version of this software, the EthoWatcher-OpenSource (OS), that adds new tools and procedures. This software was registered (COLOQUE AQUI OS DADOS DO REGISTRO NO INPE, COMO EU FIZ COM O ETHOWATCHER ANTIGO), and is licenced as a ...................... Licence........

Tools developed 1) easy and safe use of blind-to-condition categorical transcriptions, that are simultaneous with tracking and are facilitated by using mouse-triggered buttons to enter behavioral categories 2) routines to assess intra- and multiple inter- and intra-rater reliability throughout observers training and data collection, using Cohen´s Kappa (K), Fleiss’s K statistics and some corollary measures of agreement (bias, prevalence, Kmax , etc.), as well as resources for producing video clips or samples of behavioral categories, useful for observer training purposes, as well as to build databases devised to analysis by machine-learning algorithms.
3) resulting data reports (.csv and .html files), where behavioral categories are associated frame-by-frame to tracking, morphologic and kinematic attributes of animal’s image (extracted using machine vision algorithms, with advanced velocity controls and corrections for noise and artifacts of tracking), allowing to build databanks suitable for processing by machine learning algorithms 4) extended capabilities of temporal (multiple temporal bins for grouping results) and spatial (multiple areas and formats) segmentation of categorical, morphological, and kinematic data extracted from video, as well as offer outputs suitable for sequential analysis (e.g., semi-Markov chain-based statistics for behavioral sequences). 5) bulk video-file management, permitting the use of a large range of prerecorded video-file characteristics, origins, and codecs, 6) and, as the most encouraging and exciting news, the present version allows for continued collaborative development (e.g., addition of data extraction and data analysis algorithms), being available through an open-source platform cooperative environment. These new characteristics are expected to promote higher inter-experiment and multi-center standardization of behavioral recording, to increase and inform on the reliability of data and reproducibility of results derived from direct observation of behavior or from approaches using automatic activity/tracking analysis, fostering transparency in methodological aspects of the behavioral studies.

How to run You will need to install K-Lite Mega Codec Pack to correct open your videos.