Medicine Identification in Handwritten Prescriptions using Machine Learning

Abstract

"This study focuses on the development of an automated system for identifying medicine names in handwritten prescriptions that employs Optical Character Recognition (OCR) and Convolutional Recurrent Neural Network (CRNN) technologies. The project tackles the challenge of accurately transcribing various handwriting styles and potential distortions in medical prescriptions, which is critical for improving patient safety and healthcare efficiency. During the initial implementation phase, the author chose the IAM dataset, which consists of handwritten text images created specifically for handwritten text recognition (HTR) tasks. This dataset includes a wide variety of handwriting styles and formats commonly found in medical prescriptions, making it ideal for training and benchmarking OCR and CRNN models. This study focuses on the development of an automated system for identifying medicine names in handwritten prescriptions that employs Optical Character Recognition (OCR) and Convolutional Recurrent Neural Network (CRNN) technologies. The project tackles the challenge of accurately transcribing various handwriting styles and potential distortions in medical prescriptions, which is critical for improving patient safety and healthcare efficiency. During the initial implementation phase, the author chose the IAM dataset, which consists of handwritten text images created specifically for handwritten text recognition (HTR) tasks. This dataset includes a wide variety of handwriting styles and formats commonly found in medical prescriptions, making it ideal for training and benchmarking OCR and CRNN models. "