OCR based text extraction tool for Medical Text Analysis 2025 - 2026
MINOR PROJECT REVIEW - I
Project Title:
OCR based text extraction tool for Medical Text Analysis
Project Name: MediBrief
Submitted By:
Faikaf Ahmad – 2405112120047
Akanksha Gupta – 2405112120011
Anchal Maheshwari – 2405112120016
Under the Guidance of:
Prof C Sheeja
Department of Computer Applications
Parul Institute Of Computer Application
Session: 2025 - 2026
OCR based text extraction tool For Medical Text Analysis 1
OCR based text extraction tool for Medical Text Analysis 2025 - 2026
Project Title & Definition
2.1 Project Definition
Project Title: OCR based text extraction tool for Medical Text Analysis
Project Name: MediBrief
2.2 Project Description
MediBrief is an AI-powered system designed to digitize and simplify the processing of
medical prescriptions. The system automates the entire workflow—from uploading
prescription files to extracting text using Optical Character Recognition (OCR), cleaning the
data, and applying Natural Language Processing (NLP) for structured summarization.
This system addresses key inefficiencies in the healthcare domain, where manual data entry
is both time-consuming and error-prone. MediBrief enables healthcare professionals to
quickly upload PDF, DOCX, or image-based prescriptions. Through a combination of EasyOCR
and PyMuPDF, it extracts text, even from challenging handwriting and scanned documents.
OpenAI’s GPT-3.5 Turbo model is then used to interpret and summarize patient data,
medical conditions, and prescribed medicines into clean, structured formats.
The summarized output is available for download in PDF, DOCX, or JSON format, making it
easy to integrate with other hospital systems. In addition, the system monitors costs
associated with OpenAI API usage and ensures that it remains affordable for institutions.
With Django as the backend and PostgreSQL as the production database, the platform is
scalable and secure. MediBrief not only improves the efficiency of prescription handling but
also provides a foundation for advanced use cases such as multilingual processing and
medical analytics.
2.3 Existing System and/or Work Environment
1. ABBYY FineReader – Offers strong OCR but lacks AI summarization.
2. Gpt4.o API – Good for text recognition but expensive and doesn’t summarize or
structure content.
3. Manual Entry in Hospitals – Prone to human error, time-intensive, and
inconsistent.
Drawbacks:
No AI-based summarization
High API costs
OCR based text extraction tool For Medical Text Analysis 2
OCR based text extraction tool for Medical Text Analysis 2025 - 2026
Lack of integration with EMR systems
Manual processes are slow and error-prone
2.4 Problem Statements
Prescription analysis takes 10–15 minutes manually
Handwritten content leads to OCR errors
Scattered data causes poor record-keeping
₹5–10 per page OCR tools are unaffordable without summaries
2.5 Need for New System
Drastically reduce processing time and effort
Provide accurate AI-based prescription summaries
Save costs compared to traditional OCR services
Centralized and structured data for better decision-making
2.6 Proposed System & Features
Modules:
File Upload (PDF/DOCX/Image)
Text Extraction (OCR via EasyOCR, PyMuPDF)
Preprocessing (noise removal, image cleaning via OpenCV)
AI Summarization (GPT-3.5 Turbo for structured summaries)
Export Module (Download in DOCX, PDF, JSON)
Cost Tracker (API usage monitor)
Features:
Web-based interface for easy access
Secure login and role-based access
Multilingual readiness
Real-time processing and download
2.7 Project Scope
Where: Hospitals, clinics, pharmacies
By Whom: Doctors, healthcare staff, pharmacists
How: Via web portal and APIs
When: During patient visits, prescription reviews, and audits
Scope: Can expand to include EMR systems and multilingual NLP
2.8 Outcomes
OCR based text extraction tool For Medical Text Analysis 3
OCR based text extraction tool for Medical Text Analysis 2025 - 2026
90% reduction in manual workload
AI-generated structured prescriptions
30% cost savings
Higher efficiency and fewer errors
Easy integration with future healthcare tech (EMRs, telemedicine)
2.9 Tools & Technology Used
Tool/Tech Purpose Why Chosen
Django Backend/API Secure, scalable, Python-native
EasyOCR Image OCR Lightweight, multilingual support
PyMuPDF PDF Parsing Fast and efficient
OpenCV Image Preprocessing Boosts OCR accuracy
python-docx DOCX Export & Parsing Simplifies file operations
OpenAI GPT-3.5 Turbo Text Summarization Best-in-class language model
2.10 Project Plan & Timeline
Tabular Plan:
Task Start Date End Date Duration
Requirement Analysis 01-07-2025 12-07-2025 12 Days
Backend Development (Django) 12-07-2025 02-08-2025 22 Days
OCR Integration 02-08-2025 06-08-2025 5 Days
AI Summarization Module 06-08-2025 13-08-2025 8 Days
Frontend Development 13-08-2025 20-08-2025 8 Days
Testing & Debugging 20-08-2025 27-08-2025 8 Days
Deployment & Final Review 27-08-2025 31-08-2025 5 Days
OCR based text extraction tool For Medical Text Analysis 4