DocAssist: Signature Perception System using Deep Learning
Pooja Sharma, Ishika Miglani, Lucky Sharma · 2024
All In the realm of communication, individuals employ two primary modes: written and spoken language. Handwriting, in particular, serves as a powerful tool for conveying information and emotions across various contexts, including healthcare. Despite its widespread use, deciphering handwritten text remains a challenge due to variations in individual writing styles and the lack of standardization. This challenge is particularly pronounced in the medical field, where handwritten prescriptions can lead to serious consequences if misinterpreted. Our research addresses this issue by proposing DocAssist, a novel model that utilizes Convolutional Neural Networks to achieve an impressive 81% accuracy in recognizing doctors' handwriting. By leveraging machine learning and image recognition technologies, DocAssist aims to enhance prescription legibility, mitigate medication errors, and ultimately improve patient safety and healthcare outcomes. Through rigorous experimentation with the EMNIST dataset, our study demonstrates the effectiveness of DocAssist in accurately interpreting handwritten prescriptions, highlighting its potential to revolutionize prescription processing and alleviate the burden on healthcare professionals and pharmacists.