Text recognition (OCR) for patient records digitization using CNN

Optical character recognition (OCR) is widely used to transcribe texts from images in computer vision. Although current OCR methods can accurately transcribe printed text (structured), they often fall short on unstructured or handwritten text recognition. This project proposed a text recognition met...

Full description

Bibliographic Details
Main Author: Ong, Zi Leong
Format: Final Year Project / Dissertation / Thesis
Published: 2022
Subjects:
Online Access:http://eprints.utar.edu.my/4662/
http://eprints.utar.edu.my/4662/1/fyp_CS_2022_OZL.pdf
_version_ 1848886209615495168
author Ong, Zi Leong
author_facet Ong, Zi Leong
author_sort Ong, Zi Leong
building UTAR Institutional Repository
collection Online Access
description Optical character recognition (OCR) is widely used to transcribe texts from images in computer vision. Although current OCR methods can accurately transcribe printed text (structured), they often fall short on unstructured or handwritten text recognition. This project proposed a text recognition method to recognize handwritten text on patients' clinical data using a convolutional neural network (CNN). We compiled custom handwriting datasets from MNIST 0-9 and Kaggle A-Z datasets to add more handwriting diversity in training a more robust OCR model. The CNN has 3-convolutional layers to learn high-level features and a dropout layer to prevent overfitting. The preliminary results showed that the proposed model achieved 93.75% classification accuracy while Tesseract (the state-of-the-art OCR) scored 69.79%. The data will be transformed from handwritten text to computer-readable text and then stored in files in xml form for further development.
first_indexed 2025-11-15T19:34:52Z
format Final Year Project / Dissertation / Thesis
id utar-4662
institution Universiti Tunku Abdul Rahman
institution_category Local University
last_indexed 2025-11-15T19:34:52Z
publishDate 2022
recordtype eprints
repository_type Digital Repository
spelling utar-46622023-01-15T13:28:40Z Text recognition (OCR) for patient records digitization using CNN Ong, Zi Leong Q Science (General) T Technology (General) Optical character recognition (OCR) is widely used to transcribe texts from images in computer vision. Although current OCR methods can accurately transcribe printed text (structured), they often fall short on unstructured or handwritten text recognition. This project proposed a text recognition method to recognize handwritten text on patients' clinical data using a convolutional neural network (CNN). We compiled custom handwriting datasets from MNIST 0-9 and Kaggle A-Z datasets to add more handwriting diversity in training a more robust OCR model. The CNN has 3-convolutional layers to learn high-level features and a dropout layer to prevent overfitting. The preliminary results showed that the proposed model achieved 93.75% classification accuracy while Tesseract (the state-of-the-art OCR) scored 69.79%. The data will be transformed from handwritten text to computer-readable text and then stored in files in xml form for further development. 2022-04-21 Final Year Project / Dissertation / Thesis NonPeerReviewed application/pdf http://eprints.utar.edu.my/4662/1/fyp_CS_2022_OZL.pdf Ong, Zi Leong (2022) Text recognition (OCR) for patient records digitization using CNN. Final Year Project, UTAR. http://eprints.utar.edu.my/4662/
spellingShingle Q Science (General)
T Technology (General)
Ong, Zi Leong
Text recognition (OCR) for patient records digitization using CNN
title Text recognition (OCR) for patient records digitization using CNN
title_full Text recognition (OCR) for patient records digitization using CNN
title_fullStr Text recognition (OCR) for patient records digitization using CNN
title_full_unstemmed Text recognition (OCR) for patient records digitization using CNN
title_short Text recognition (OCR) for patient records digitization using CNN
title_sort text recognition (ocr) for patient records digitization using cnn
topic Q Science (General)
T Technology (General)
url http://eprints.utar.edu.my/4662/
http://eprints.utar.edu.my/4662/1/fyp_CS_2022_OZL.pdf