Developing a Benchmark Data Set for Optical Character Recognition of Drug Names in Medical Prescriptions
Article Sidebar
Main Article Content
Abstract
Interpreting handwritten medication names remains a global challenge that often causes medical errors and delays in drug dispensing. This issue threatens patient safety and demands effective solutions. To address it, a data set was developed of handwritten drug names to support the training of optical character recognition (OCR) systems. The goal is to improve prescription readability, enable digital archiving, and support electronic health records. The data set can also be used in smart health applications to help patients manage their medications. The data set was built using contributions from 250 physicians and medical students in southern region of Libya. Each participant wrote 20 predefined drug names in standardized forms, producing 9,225 handwriting samples. Image processing techniques were used, such as converting color images to grayscale, to reduce memory requirements by reducing the size to one-third of the original size and increasing processing speed by ~66%. In addition, input normalization was applied by converting pixel. These techniques accelerate model training, enhance accuracy, and address artifacts caused by different handwriting sizes. This resource aims to reduce prescription-related errors, enhance data accessibility through searchable archives, and support faster and safer medication processing. The paper also proposes a practical approach to integrating intelligent systems in healthcare and highlights the importance of collaboration between clinicians and researchers to improve patient safety and care quality.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
References
A. Kumar, I. Dangi, S. Chowdary, and K. K. Pandey, “Ideal drug prescription writting,” vol. 8, no. 3, pp. 634–654, 2019, doi: 10.20959/wjpps20193-12989.
R. Achkar, K. Ghayad, R. Haidar, S. Saleh, and R. Al Hajj, “Medical handwritten prescription recognition using CRNN,” CITS 2019 - Proceeding 2019 Int. Conf. Comput. Inf. Telecommun. Syst., pp. 1–5, 2019, doi: 10.1109/CITS.2019.8862004.
E. Hassan, H. Tarek, M. Hazem, and S. Bahnacy, “Medical Prescription Recognition using Machine Learning,” IEEE 11th Annu. Comput. Commun. Work. Conf. (CCWC, pp. 973–979, 2021.
S. Kumar Sarkar, “A New Approach to Information Retrieval based on Keyword Spotting from Handwritten Medical Prescriptions Arghya Mukhejee 1a , Arunit Halder 1b , Subhrapratim Nath 1c,” Adv. Ind. Eng. Manag., vol. 6, no. 2, pp. 90–96, 2017, doi: 10.7508/aiem.2017.02.006.
Y. Amol Rathod et al., “Handwritten Character Recognition Using CNN, KNN and SVM,” Int. J. Technol. Eng. Arts Math. Sci., vol. 1, no. 2, pp. 2583–1224, 2022, doi: 10.11591/eei.v9i3.xxxx.
E. Kamalanaban, M. Gopinath, and S. Premkumar, “Medicine box: Doctor’s prescription recognition using deep machine learning,” Int. J. Eng. Technol., vol. 7, no. 3.34 Special Issue 34, pp. 114–117, 2018, doi: 10.14419/ijet.v7i3.34.18785.
P. S. Dhande and R. Kharat, “Character Recognition for Cursive English Handwriting to Recognize Medicine Name from Doctor’s Prescription,” 2017 Int. Conf. Comput. Commun. Control Autom. ICCUBEA 2017, pp. 1–5, 2017, doi: 10.1109/ICCUBEA.2017.8463842.
H. Brits et al., “Illegible handwriting and other prescription errors on prescriptions at National District Hospital , Bloemfontein Illegible handwriting and other prescription errors on prescriptions at National District Hospital , Bloemfontein,” South African Fam. Pract., vol. 6190, pp. 1–4, 2017, doi: 10.1080/20786190.2016.1254932.
M. A.-A.-S. C. A. A. M. A. M. R. N. T. Mia, Abdur Rahim, “A Deep Neural Network Approach with Pioneering Local Dataset to Recognize Doctor’s Handwritten Prescription in Bangladesh,” IEEE Int. Conf. Adv. Comput. Commun. Electr. Smart Syst. ), Dhaka, Bangladesh, pp. 1–6, 2024, doi: 10.1109/iCACCESS61735.2024.10499631.
U. Ali, M. Nadeem, H. Ishfaq, and W. Ali, “Leveraging Deep Learning with Multi-Head Attention for Accurate Extraction of Medicine from Handwritten Prescriptions,” vol. 978-1–7281, 2024.
U. V. Marti and H. Bunke, “A full English sentence database for off-line handwriting recognition,” Proc. Int. Conf. Doc. Anal. Recognition, ICDAR, no. November 1999, pp. 709–712, 1999, doi: 10.1109/ICDAR.1999.791885.
R. E. Ferner and J. K. Aronson, “Nominal ISOMERs ( Incorrect Spellings Of Medicines Eluding Researchers )— variants in the spellings of drug names in PubMed : a database review,” 2016, doi: 10.1136/bmj.i4854.
L. J. Fajardo et al., “Doctor’s Cursive Handwriting Recognition System Using Deep Learning,” 2019 IEEE 11th Int. Conf. Humanoid, Nanotechnology, Inf. Technol. Commun. Control. Environ. Manag. HNICEM 2019, pp. 6–11, 2019, doi: 10.1109/HNICEM48295.2019.9073521.
S. R. Dhar D, Garain A, Singh P, “HP_DocPres: a method for classifying printed and handwritten texts in doctor’s prescription,” Multimed. Tools Appl., 2020.
N. P. and N. Sampath., “Detecting and extracting information of medicines from a medical prescription using deep learning and computer vision,” Int. Conf. Knowl. Eng. Commun. Syst., pp. 1–6, 2022.
K. Sehimi, “A Transfer Learning approach for handwritten drug names recognition,” 2023 IEEE Int. Conf. Networking, Sens. Control, vol. 1, pp. 1–6, 2023, doi: 10.1109/ICNSC58704.2023.10318974.