Medicovate — AI for Surgical Training

The Problem

Surgical training is broken.

Every year, millions of cataract surgeries are performed worldwide. Yet training the next generation of surgeons still relies on outdated, subjective, and expensive methods.

Years to Mastery

Phacoemulsification alone requires years of supervised practice. There's no shortcut — unless you fundamentally change the feedback loop.

Too Few Mentors

Expert ophthalmologists are scarce. Most trainees get feedback that is sporadic, inconsistent, and highly subjective.

Subjective Feedback

Two experts watching the same surgery may give completely different evaluations. There's no standardized, objective baseline.

Expensive Simulators

Wet labs, dry labs, and surgical simulators cost hundreds of thousands — yet they still can't replicate real operative conditions.

"The surgical training pipeline hasn't fundamentally changed in decades. We think AI can."

— Medicovate Team

Our Approach

AI that watches surgery and actually understands it.

Medicovate is building a production SaaS platform that ingests cataract surgery videos and runs them through 5 specialized AI subsystems — in real-time and post-operatively — to generate objective, actionable feedback.

01

Phase Recognition

Identifies which of the 8 canonical stages the surgery is in — from Incision to Wound Hydration — in real time.

02

Instrument Detection

Detects and segments 22 surgical instruments using YOLO11s-seg. Knows what tool is being used and where.

03

Anatomy Segmentation

CaDIS (36 classes) + SAM3 provides pixel-level understanding of the surgical field's anatomical structures.

04

Error Detection

4 models (ResNet50, EfficientNet-B2, IOL rotation, pupil reaction) catch 18 types of surgical errors as they happen.

05

Post-op Reporting

A fine-tuned SmolVLM2-2.2B generates full surgical narratives and performance reports automatically.

The Pipeline

From raw video to structured insight.

Here's how Medicovate processes a single surgical video end-to-end.

1

Upload

Surgeon or hospital uploads the operative video to the platform.

2

Phase Detection

AI identifies all 8 surgical phases and timestamps each transition.

3

Detection & Tracking

Instruments are detected, anatomy is segmented, errors are flagged.

4

Reasoning Engine

A "ChatGPT for surgery" — asks frame-by-frame questions, links facts, verifies understanding, and builds context.

5

Performance Report

A structured report graded on OASIS, GRASIS, and ICO-OSCAR:phaco scales.

Assessment Metrics

Grounded in established medical scales.

Our AI doesn't invent its own grading. It maps to internationally recognized surgical proficiency frameworks.

OASIS

Objective Assessment of Skills in Intraocular Surgery

GRASIS

Global Rating Assessment of Skills in Intraocular Surgery

ICO-OSCAR:phaco

Ophthalmology Surgical Competency Assessment Rubric for Phacoemulsification

The Team

Built by engineers who care about surgery.

We're a focused team of AI engineers and medical technologists on a mission to make surgical training accessible, objective, and scalable.

Nada Hassab

CEO and AI Engineer

Youssef Abdelmaksod

AI Engineer

Allam Elshiekh

AI Engineer

FAQ

Frequently Asked Questions

Common questions about our surgical AI platform.

How does Medicovate handle patient privacy?

We prioritize data security and are building our platform with HIPAA and GDPR standards in mind. All surgical videos are de-identified before processing.

Does the AI work in real-time?

Yes, our platform supports both real-time intraoperative guidance and post-operative analysis to provide comprehensive surgical support.

How do I integrate it with my OR cameras?

Our system is camera-agnostic. It can ingest raw video feeds from standard surgical microscopes and endoscopic cameras without requiring proprietary hardware.

What surgical specialties do you support?

While our current focus is on cataract surgery (phacoemulsification), our core AI architecture is designed to be adaptable to other surgical domains.

Open for Collaboration

We're looking for hospitals who want to shape the future of surgical training.

Medicovate is in its MVP stage. We're actively seeking partnerships with teaching hospitals, ophthalmology departments, and surgical training programs to pilot our platform, co-develop features, and validate our AI in real clinical settings.

Pilot the platform at your facility

Co-develop and validate with real surgical data

Be part of a published research collaboration

Early access to all platform features