About PolypVision

Empowering Early Detection with AI

Welcome to PolypVision, an AI-powered platform dedicated to supporting medical professionals with cutting-edge tools for the detection and classification of colorectal polyps.

Using advanced Convolutional Neural Networks (CNNs) and state-of-the-art deep learning techniques, PolypVision helps distinguish between adenomatous and hyperplastic polyps with high accuracy, enabling faster and more informed clinical decisions.

Our platform currently offers two powerful AI tools:

  • Polyp Classifier - Classifies polyps as adenomatous or hyperplastic
  • Polyp Segmentor - Precisely outlines polyp boundaries in colonoscopy images

We're committed to continuous improvement and have ambitious plans for future development:

  • Real-time polyp localization and tracking
  • Polyposis risk analysis and prediction
  • Clinical decision support systems for surgeons
  • Integration with hospital information systems
  • Multi-modal data analysis (imaging + patient history)

PolypVision is built by a passionate team of AI researchers, software engineers, and healthcare collaborators working toward a smarter, faster, and more accessible medical future.

AI-powered polyp detection illustration

Our Mission

To democratize access to advanced AI-powered polyp detection tools, empowering healthcare professionals worldwide to detect colorectal cancer earlier and save lives.

Our Vision

A future where AI-assisted diagnostics are standard in every gastroenterology clinic, reducing missed detections and improving patient outcomes globally.

Meet the Team

Dr.Hamidreza Bolhassani

Lead AI Researcher

test text

DR.Mojgan Frooten

Medical Advisor

test text

Amirmohammad Akbari

Lead Software Engineer

Full-stack developer specializing in AI deployment

Hamid Reza Rastad

AI engineer and Data scientist

Expert in deep learning and medical imaging datasets and ML vision models

10,000+

Medical Images Analyzed

94%

Classification Accuracy

500+

Active Users

24/7

AI Model Availability

Powered By

PyTorch TensorFlow Django FastAPI CNN Transfer Learning ResNet50 U-Net