CoDiagnosis
AI-Powered Medical Diagnosis & Radiology Education Platform
V2F Solutions collaborated with experts from the University of Houston and The University of Texas MD Anderson Cancer Center to build CoDiagnosis, an AI-powered medical diagnosis and radiology education platform. It gives medical trainees hands-on access to clinical-grade DICOM imaging, structured case assignments, expert mentorship, and AI-assisted analysis, all within one connected learning environment.
Project Overview
CoDiagnosis is a web-based platform that bridges traditional medical education and real-world diagnostic practice. Students analyze real imaging cases, annotate findings, compare their work against expert data, and receive measurable feedback, while instructors assign cases and track progress.
Our Engagement
We worked closely with academic and medical experts to understand the educational requirements, clinical workflows, and AI capabilities needed, then designed and developed CoDiagnosis as a complete digital learning ecosystem. The platform combines a modern LMS, a browser-based clinical imaging environment, DICOM infrastructure, and integrated AI models for detection, segmentation, and automated evaluation.
Timeline
4 Months
Discovery → Launch
June 2024 – September 2024
Target User
Medical Trainees, Educators
Radiology Students and Mentors
Students, Instructors, Medical Experts
Goal
Practical Diagnostic Training
Hands-on imaging with measurable feedback
AI-assisted learning & evaluation
- The Challenge
Radiology education depends on analyzing complex medical images and comparing findings with expert interpretations. Traditional learning environments struggle to offer consistent access to clinical-grade imaging, structured assignments, expert feedback, and measurable performance data in one place.
• Limited hands-on access to real-world DICOM imaging cases.
• No structured, case-based learning environment for radiology
• Difficulty for instructors to assign cases and monitor progress
• No reliable way to compare student findings against expert data
• Delayed, inconsistent, and hard-to-measure feedback on diagnostic performance

• The client needed one platform bringing together imaging, education, mentorship, AI-assisted diagnosis, and performance analytics, not just an LMS or an imaging viewer.
- How V2F Solved It
An integrated web platform combining structured learning, clinical-grade imaging, and AI-assisted diagnosis.
Delivers a structured radiology LMS with courses, case-based learning, targeted assignments, and progress tracking for students and instructors.
Integrates the OHIF Viewer so students can view, measure, and annotate DICOM images directly in the browser using professional diagnostic tools.
Embeds AI models into the workflow for abnormality detection, image segmentation, and automated comparison of student annotations with expert data.
Provides teacher and expert mentorship tools plus analytics to review submissions, give feedback, and measure diagnostic accuracy over time.
How It Works
A Complete Diagnostic Learning Cycle for Trainees and Educators
Assign Case: Instructors assign targeted imaging cases to students or groups.
Analyze Images: Students review DICOM images using clinical-grade viewer tools.
Annotate & Submit: Students add findings, use AI-assisted analysis, and submit their work.
Review & Improve: AI and mentors evaluate submissions against expert data with measurable feedback.
Powerful Features
Imaging, AI, mentorship, and analytics brought together in one learning platform.
Clinical-Grade Imaging
Browser-based DICOM viewing with zoom, pan, measurements, multi-planar reconstruction, and annotation tools.
AI Diagnostic Assistance
Computer-aided detection, automated segmentation, and structure outlining built directly into case workflows.
Automated Evaluation
Compares student annotations with expert-established data to grade diagnostic accuracy instantly.
AI-Powered Radiology Education & Diagnostic Training
A secure, interactive platform for case-based learning, clinical imaging analysis, and expert-led mentorship.
Tech Stack
Built with modern, enterprise-grade technologies for performance, scalability, and security
Frontend
React
Next.js
Tailwind CSS
Medical Imaging
OHIF Viewer

Multi-Planar Reconstruction

Annotation Tools
DICOM & PACS
Orthanc

DICOMweb APIs

Secure Image Streaming
AI / ML
Anomaly Detection (CAD)
Image Segmentation

Automated Evaluation
LMS & Analytics
Courses & Assignments
Assessments

Performance Tracking
Backend & Security
Python

Next.js
Agile Development
Iterative approach with regular client feedback
Thoroughly Tested
Comprehensive testing for reliability
Well Documented
Clear documentation for easy maintenance
Client’s Feedback
What the client said after launch
Results & Impact
Transforming scattered historical information into a centralized digital resource that preserves Mary Hamilton’s story and makes her legacy accessible to a wider audience.
Hands-On Learning
Clinical-Grade Access: Students work directly with real DICOM cases in the browser.
Instant Feedback
AI-Driven Evaluation: Automated comparison against expert data speeds up grading.
Connected Mentorship
3 Core Roles: Students, teachers, and medical experts in one workflow.
Measurable Progress
Data-Backed Insights: Tracks accuracy, completion, and outcomes over time.
- Project Deliverables
- Radiology Learning Management System
- Course & Curriculum Management
- Case-Based Learning & Targeted Assignments
- AI Anomaly Detection (CAD)
- Automated Image Segmentation
- Automated Grading & Expert Comparison
Academic Partnership
Built with University of Houston and MD Anderson experts
Scalable Solution
Architecture ready for more courses, cases, and institutions
Easy Maintenance
Clean code with comprehensive documentation
Let's Build What's Next
Whether you’re developing a new digital product, modernizing enterprise systems, adopting AI, or accelerating cloud transformation, V2F has the expertise to help you move forward with confidence.
