AI-Based Prescription Interpretation with Healthcare Support

Abstract

The AI-Based Healthcare Prescription Interpretation with Healthcare Support Application is an intelligent mobile healthcare solution designed to simplify the understanding of medical prescriptions and enhance medication management. The application enables users to upload or capture images of handwritten or printed prescriptions and medical reports using their mobile devices. These inputs are processed using advanced image recognition and natural language processing techniques to accurately extract and interpret medication details and essential health information. An integrated AI-powered chatbot further enhances user interaction by providing simplified explanations of prescribed medicines, addressing user queries, recommending general precautions, and suggesting personalized dietary guidance based on medication usage. To ensure inclusivity and accessibility, the application supports multiple languages, including English, Telugu, and Hindi, making it suitable for a diverse user population. Additionally, the application securely stores prescription data and user interaction history, enabling users to access and review past records whenever needed.

Introduction

The text describes an AI-powered healthcare system designed to solve problems in interpreting handwritten prescriptions and improving medication management. Traditional prescription handling is manual and error-prone, often leading to misinterpretation, poor record-keeping, missed doses, and difficulty finding pharmacies.

To address this, the proposed system uses a web-based platform built with Django and integrates multimodal AI (Gemini 2.5 Flash) to extract and interpret prescription information from images. It combines image processing and natural language processing to convert handwritten prescriptions into structured digital data, reducing human error and eliminating manual entry.

The system includes key features such as:

• Automated prescription analysis using AI (OCR + structured extraction)
• A personalized dashboard for managing prescriptions and medication history
• A context-aware chatbot that explains medications and dosage instructions
• A geospatial pharmacy locator using the Haversine formula to find nearby pharmacies
• Medication reminders to improve adherence

A structured methodology is used where users upload prescription images, which are processed into JSON data, stored in a database (SQLite), and then used across system modules. The architecture is layered, including UI, application logic (Django), AI processing (Gemini), geospatial services (Geopy), and data storage.

Results show high accuracy in extracting handwritten prescriptions compared to traditional OCR tools, with fast processing time (3–5 seconds). The pharmacy locator efficiently identifies nearby pharmacies, and the chatbot improves patient understanding by providing personalized responses based on prescription history.

Conclusion

This research presents the successful design and implementation of an AI-powered healthcare management system that addresses the challenges associated with manual prescription handling. The proposed system effectively automates the digitization and interpretation of handwritten prescriptions by leveraging advanced large multimodal models integrated within a robust web-based framework. By combining artificial intelligence with geospatial services, the system significantly reduces the risk of medication errors, improves prescription clarity, and enhances overall patient autonomy in managing healthcare activities. The system demonstrates several key advantages. It enables automated extraction of unstructured data from handwritten prescription images, thereby eliminating the need for manual transcription. The integration of a context-aware AI chatbot provides personalized medical insights and supports user queries based on historical prescription data. Additionally, the built-in medication reminder system improves adherence to prescribed treatments, ensuring timely dosage intake. The inclusion of real-time pharmacy location services further streamlines access to medicines by allowing users to identify nearby pharmacies efficiently. Despite these contributions, there remains scope for further enhancement. Future work will focus on integrating real-time pharmacy inventory APIs to enable users to verify the availability of prescribed medicines before visiting a pharmacy. Moreover, the incorporation of blockchain-based solutions for maintaining prescription records can improve data security, integrity, and controlled sharing between healthcare providers and patients. These advancements will further strengthen the system’s reliability, scalability, and applicability in modern digital healthcare environments.

References

[1] Google DeepMind, “Gemini: A Generative AI Model for Multimodal Understanding,” 2024. [2] Django Software Foundation, “Django Documentation: The Web Framework for Perfectionists with Deadlines,” 2024. [3] Geopy Project, “Geopy: Python Geocoding and Distance Calculation,” 2023. [4] IEEE, “IEEE Standard for Health Informatics—Personal Health Record (PHR)—Architecture and Functional Requirements.” [5] J. Smith et al., “Deep Learning for Handwritten Medical Prescription Recognition,” IEEE Journal of Biomedical and Health Informatics, 2022. [6] L. Zhang, “Comparative Analysis of LLMs in Clinical Entity Recognition,” in Proceedings of the International Conference on AI in Medicine, 2023. [7] World Health Organization, “Medication Adherence: Challenges and Digital Solutions in Modern Healthcare,” 2021.

Copyright

Copyright © 2026 Ramisetti Naga Sowmya Sri, Polisetty Pravallika, Shaik Ramjanbi, Yerru Ruthika. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.