Handwriting Recognition & Prescription Scanner

Prescription errors due to illegible handwriting and misinterpretation pose a serious threat to patient safety, often leading to incorrect medication dispensation, adverse drug interactions, and treatment complications. This project introduces an innovative Prescription Scanner built using Python to streamline prescription interpretation and reduce human error. The system integrates Optical Character Recognition (OCR) and Natural Language Processing (NLP) to accurately digitize both handwritten and printed prescriptions. Tesseract OCR extracts text from printed prescriptions, while a deep learning-based handwriting recognition model enhances the accuracy of handwritten script interpretation. Once extracted, the text undergoes NLP processing to identify key prescription elements, such as drug names, dosages, and usage guidelines. Additionally, real-time integration with a pharmaceutical database enables automated validation, helping to detect potential medication errors, incorrect dosages, and dangerous drug interactions. Experimental evaluation on a diverse dataset demonstrated high accuracy, achieving over 85% for handwritten text and more than 95% for printed prescriptions. This system provides a reliable solution to enhance patient safety, reduce pharmacists' workload, and improve prescription documentation. Future developments could focus on expanding dataset diversity, refining recognition accuracy, and integrating with electronic health records for real-time use.

• Publication year

  2025

• Venue

  International journal on advanced electrical and computer engineering

• Publication date

  2025-05-23

• Fields of study

  Not labeled

• Identifiers
  DOI 10.65521/ijaece.v14i1.407
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Handwritten Text Recognition using Deep Learning with TensorFlow

  2020cited by this paper
• A set of benchmarks for Handwritten Text Recognition on historical documents

  2019cited by this paper
• Robust Encoder-Decoder Learning Framework towards Offline Handwritten Mathematical Expression Recognition Based on Multi-Scale Deep Neural Network

  2019cited by this paper
• Convolve, Attend and Spell: An Attention-based Sequence-to-Sequence Model for Handwritten Word Recognition

  2018cited by this paper
• Attention is All you Need

  2017cited by this paper
• Are Multidimensional Recurrent Layers Really Necessary for Handwritten Text Recognition?

  2017cited by this paper
• The IAM-database: an English sentence database for offline handwriting recognition

  2002cited by this paper
• Long Short-Term Memory

  1997cited by this paper

• No citing papers are available for this paper.