Please use this identifier to cite or link to this item:
http://dr.iiserpune.ac.in:8080/xmlui/handle/123456789/6075
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Kulkarni, Viraj | en_US |
dc.contributor.author | GUPTA, PAWAN | en_US |
dc.date.accessioned | 2021-07-13T05:48:36Z | - |
dc.date.available | 2021-07-13T05:48:36Z | - |
dc.date.issued | 2021-07 | - |
dc.identifier.citation | 42 | en_US |
dc.identifier.uri | http://dr.iiserpune.ac.in:8080/xmlui/handle/123456789/6075 | - |
dc.description.abstract | After recovery, many Covid Patients have a post-corona effect, and the most common is a lung infection (Pneumonia) in treatment response of this the first things that are done are taking X-rays and creating Reports. Medical images like X-rays, CT scans, and MRIs possess a lot of information in them, like the possible presence of some pathologies or the changes in the normal functioning of the organs. The information present in this data get transcribed by Radiologists in plain text reports, which summarise the crucial findings from the images. We are trying to create the medical report generation using artificial intelligence, more precisely using LSTM and Convolutional Neural Network. We want to propose the best method for generating text medical reports by comparing the hierarchical encoder-decoder network with and without attention mechanism. We have approached Models computationally and doesn’t get into depth of Mathematics. | en_US |
dc.language.iso | en | en_US |
dc.subject | Medical Report Generation Model | en_US |
dc.title | Medical Report Generation Model | en_US |
dc.type | Thesis | en_US |
dc.type.degree | BS-MS | en_US |
dc.contributor.department | Dept. of Data Science | en_US |
dc.contributor.registration | 20161038 | en_US |
Appears in Collections: | MS THESES |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
MS Thesis.pdf | 987.17 kB | Adobe PDF | View/Open Request a copy |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.