Artificial Intelligence-Based Grading Quality of Bovine Blastocyst Digital Images: Direct Capture with Juxtaposed Lenses of Smartphone Camera and Stereomicroscope Ocular Lens
dc.contributor.author | Gouveia Nogueira, Marcelo Fábio [UNESP] | |
dc.contributor.author | Bertogna Guilherme, Vitória [UNESP] | |
dc.contributor.author | Pronunciate, Micheli [UNESP] | |
dc.contributor.author | Dos Santos, Priscila Helena [UNESP] | |
dc.contributor.author | Lima Bezerra da Silva, Diogo [UNESP] | |
dc.contributor.author | Rocha, José Celso [UNESP] | |
dc.contributor.institution | Universidade Estadual Paulista (Unesp) | |
dc.date.accessioned | 2019-10-06T16:58:15Z | |
dc.date.available | 2019-10-06T16:58:15Z | |
dc.date.issued | 2018-12-15 | |
dc.description.abstract | In this study, we developed an online graphical and intuitive interface connected to a server aiming to facilitate professional access worldwide to those facing problems with bovine blastocysts classification. The interface Blasto3Q, where 3Q refers to the three qualities of the blastocyst grading, contains a description of 24 variables that were extracted from the image of the blastocyst and analyzed by three Artificial Neural Networks (ANNs) that classify the same loaded image. The same embryo (i.e., the biological specimen) was submitted to digital image capture by the control group (inverted microscope with 40× magnification) and the experimental group (stereomicroscope with maximum of magnification plus 4× zoom from the cell phone camera). The images obtained from the control and experimental groups were uploaded on Blasto3Q. Each image from both sources was evaluated for segmentation and submitted (only if it could be properly or partially segmented) for automatic quality grade classification by the three ANNs of the Blasto3Q program. Adjustments on the software program through the use of scaling algorithm software were performed to ensure the proper search and segmentation of the embryo in the raw images when they were captured by the smartphone, since this source produced small embryo images compared with those from the inverted microscope. With this new program, 77.8% of the images from smartphones were successfully segmented and from those, 85.7% were evaluated by the Blasto3Q in agreement with the control group. | en |
dc.description.affiliation | Laboratory of Embryonic Micromanipulation Department of Biological Sciences School of Sciences and Languages São Paulo State University (UNESP) | |
dc.description.affiliation | Multiuser Facility (FitoFarmaTec) Department of Pharmacology Biosciences Institute São Paulo State University (UNESP) | |
dc.description.affiliation | Laboratory of Applied Mathematics Department of Biological Sciences School of Sciences and Languages São Paulo State University (UNESP) | |
dc.description.affiliationUnesp | Laboratory of Embryonic Micromanipulation Department of Biological Sciences School of Sciences and Languages São Paulo State University (UNESP) | |
dc.description.affiliationUnesp | Multiuser Facility (FitoFarmaTec) Department of Pharmacology Biosciences Institute São Paulo State University (UNESP) | |
dc.description.affiliationUnesp | Laboratory of Applied Mathematics Department of Biological Sciences School of Sciences and Languages São Paulo State University (UNESP) | |
dc.description.sponsorship | Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) | |
dc.description.sponsorshipId | FAPESP: 2012/50533-2 and 2017/19323-5 | |
dc.identifier | http://dx.doi.org/10.3390/s18124440 | |
dc.identifier.citation | Sensors (Basel, Switzerland), v. 18, n. 12, 2018. | |
dc.identifier.doi | 10.3390/s18124440 | |
dc.identifier.issn | 1424-8220 | |
dc.identifier.scopus | 2-s2.0-85058733884 | |
dc.identifier.uri | http://hdl.handle.net/11449/189974 | |
dc.language.iso | eng | |
dc.relation.ispartof | Sensors (Basel, Switzerland) | |
dc.rights.accessRights | Acesso aberto | |
dc.source | Scopus | |
dc.subject | artificial intelligence | |
dc.subject | artificial neural networks | |
dc.subject | bovine blastocyst | |
dc.subject | digital image capture | |
dc.subject | embryo grading | |
dc.subject | image processing | |
dc.subject | smartphone camera | |
dc.subject | software | |
dc.title | Artificial Intelligence-Based Grading Quality of Bovine Blastocyst Digital Images: Direct Capture with Juxtaposed Lenses of Smartphone Camera and Stereomicroscope Ocular Lens | en |
dc.type | Artigo | |
unesp.author.orcid | 0000-0002-2239-9652 0000-0002-2239-9652[1] | |
unesp.author.orcid | 0000-0002-2239-9652 0000-0002-2239-9652[4] | |
unesp.department | Ciências Biológicas - FCLAS | pt |