Dr. Tarek Barakat

Lead Author

Dr. Tarek Barakat

دكتور طارق بركات

CEO & Founder · PhD Researcher, AI Medical Imaging · Software Development Lead

Dr. Tarek Barakat founded Fractify (Databoost Sdn Bhd) to close the global radiology gap through accessible, infrastructure-light AI diagnostics. His PhD research centres on deep learning architectures for medical image analysis — specifically the multi-modal fusion models powering Fractify's X-Ray, CT, MRI, and dental screening engines. He writes directly from the development trenches: translating clinical AI research into production decisions that hospital teams can actually use.

Medical Imaging AI Deep Learning Brain MRI Chest X-Ray AI DICOM Clinical Decision Support

97.9%

Brain MRI accuracy

18+

X-Ray pathologies

<3s

Analysis time

Expert Review & Advisory Panel

All articles are peer-reviewed by clinical and AI specialists before publication.

Mohd Rizam

Mohd Rizam

محمد رزام

Co-Founder & CTO

Oversees Fractify's engineering architecture — multi-tenant infrastructure, DICOM pipelines, and AI model deployment at scale across hospital networks.

Dr. Ammar Bathich

Dr. Ammar Bathich

دكتور عمار بطحيش

Distinguished Advisor — AI & Digital Transformation

Reviews all articles on AI strategy, clinical implementation, and digital transformation. Ensures content reflects real-world deployment constraints facing healthcare AI.

Dr. Safaa Mahmoud Naes

Dr. Safaa Mahmoud Naes

دكتورة صفاء ناعس

Distinguished Advisor — Medical Sciences & Molecular Medicine

Provides clinical oversight on diagnostic accuracy claims, patient safety implications, and medical science accuracy across all published content.

Recent Articles by Dr. Tarek Barakat

clinical

Radiology Turnaround Time Benchmarks

Every published radiology turnaround time figure measures one of four different intervals, and almost nobody says which. This is how to define your segments, pull the timestamps, report percentiles instead of averages, and write a target you can defend.

Read →

imaging

DICOM Router: How Studies Reach an AI Engine

A study never arrives at an AI engine by accident — something decided it should. This is what makes that decision, what it knows at the moment it makes it, and what happens when it gets it wrong.

Read →

enterprise

Vendor Neutral Archive and AI Radiology

Most VNA writing explains, at length, that a vendor neutral archive is vendor neutral. This article skips that and asks the question the brochures avoid — what changes for an AI engine when the images of record live in the archive instead of the PACS.

Read →

enterprise

AI Radiology Implementation Timeline: Week-by-Week Hospital Deployment Plan

An AI radiology deployment timeline is the structured, week-by-week operational roadmap from hospital kickoff to full clinical production use of an AI-assisted diagnostic system. It maps the parallel workstreams—IT infrastructure, dicom/pacs integration, model validation, clinici…

Read →

clinical

AI Radiology Workflows: From Clinical Validation to Confidence and Hospital Adoption

Radiologists spend 40% of clinical time on routine detection tasks—yet 73% of hospital systems cite adoption barriers unrelated to model accuracy. What separates systems that radiologists actually use daily from those that sit unused in pilot programs?

Read →

ai

AI Confidence Score 97.9%: What That Number Actually Means for Your Hospital

A 97.9% confidence score is not a probability that the AI is correct. It measures how certain the neural network is in its decision boundary during inference. That is dramatically different from accuracy, calibration, sensitivity, or specificity—yet these terms collide constantly…

Read →
All Articles