Application of variational autoencoders in image-based analysis of cellular response profiles
Cell images reconstruction from a subset of the MCF7 image repository is the primary goal of this work. It is implemented through a variational autoencoder: a generative, unsupervised learning paradigm whose architecture consists of an encoder that reduces the dimensionality of the input space by ob...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis de maestría |
| Fecha de publicación: | 2020 |
| País: | España |
| Institución: | Universitat Oberta de Catalunya (UOC) |
| Repositorio: | O2, repositorio institucional de la UOC |
| OAI Identifier: | oai:openaccess.uoc.edu:10609/121466 |
| Acceso en línea: | http://hdl.handle.net/10609/121466 |
| Access Level: | acceso abierto |
| Palabra clave: | variational autoencoders deep learning generative modelling autoencoders variacionals aprenentatge profund modelatge generatiu codificadores automáticos variacionales aprendizaje profundo modelado generativo Bioinformatics -- TFM Bioinformàtica -- TFM Bioinformática -- TFM |
| Sumario: | Cell images reconstruction from a subset of the MCF7 image repository is the primary goal of this work. It is implemented through a variational autoencoder: a generative, unsupervised learning paradigm whose architecture consists of an encoder that reduces the dimensionality of the input space by obtaining a distribution over the latent space and a decoder, that rebuilds the inputs from the encoding. This work is completed with the description of the activities associated with the primary goal, like image segmentation and processing and infrastructure setup, the latest driven by automation tools and performed in a cloud environment. |
|---|