Joan U.

Key Skills

Software

Other

Top Subject Matter

Healthcare diagnostic imaging data quality and evaluation

Healthcare multi-modal dataset curation

annotation QA

Top Data Types

Image

Text

Document

Top Task Types

Diagnosis

Classification

Freelancer Overview

Radiographer (clinical imaging procedures supporting evaluation-grade datasets) - Federal Medical Centre, Lagos. Brings 8+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal, Proprietary Tooling, and N. Education includes Bachelor of Science, University of Ibadan (2024). AI-training focus includes data types such as Medical and DICOM and labeling workflows including Evaluation, Rating, and Diagnosis.

Labeling Experience

AI-Assisted Chest X-ray Analysis and Radioprotection in Radiotherapy (research) - Ongoing

Diagnosis

Conducting AI-assisted diagnostic evaluation work on chest X-ray interpretation to benchmark clinical accuracy metrics. Applying rigorous clinical evaluation methodology to measure sensitivity and specificity targets for reliable AI performance in radiology settings. Integrating clinical radioprotection considerations with model evaluation to support safer, more effective healthcare AI deployment. • Targets 95%+ sensitivity and specificity benchmarks for chest X-ray AI diagnostics • Evaluates AI diagnostic accuracy within clinical radiology workflows • Investigates natural antioxidant approaches for radioprotection in radiotherapy • Supports research aimed at production reliability through structured measurement

2026 - Present

AI Evaluation and Quality Management Specialist - Remote

Engineered enterprise-ready quality assurance and annotation validation operations for large-scale multi-modal medical imaging datasets. Delivered 99.2% annotation accuracy across extensive validation workflows while improving data integrity to protect downstream ML model performance. Established replicable QA frameworks and annotation protocols to scale reliable training datasets across diverse project specifications. • Curated, processed, and validated 7,000+ multi-modal assets for AI training readiness • Implemented QA/validation workflows achieving 99.2% annotation accuracy over 500+ validations • Improved dataset consistency leading to 12% quality uplift across multi-domain projects • Built reusable, production-grade QA/annotation protocols for enterprise scaling

2024 - Present

Radiographer (clinical imaging procedures supporting evaluation-grade datasets) - Federal Medical Centre, Lagos

Built and executed AI/data-quality evaluation workflows for diagnostic imaging outcomes across X-ray and CT modalities. Maintained high-quality standards via protocol adherence, image quality assessment, and safety-focused checks across clinical datasets. Drove measurable performance improvements by pairing structured evaluation with clinical throughput and workflow optimization. • Evaluated image quality using quantitative scoring (average 9.2/10) • Performed data-quality and compliance checks supporting reliable model-ready imaging inputs • Achieved 98.5% procedure accuracy across X-ray/CT activities • Coordinated with radiologists to support large-scale case processing for downstream AI use

2024 - Present

AfterQuery Medical Data Pipeline Optimization - Medical Imaging Project

Other

Optimized an end-to-end medical imaging data processing and annotation review pipeline to improve dataset validation efficiency. Rebuilt QA checkpoints and review systems to increase labeling accuracy to 99.5% while accelerating data delivery cycles for ML training. Enabled scalable ingestion and seamless integration of multi-modal assets across pipeline components. • Rewrote critical AfterQuery medical imaging processing pipeline • Improved validation efficiency by 35% and reduced processing errors by 42% • Redesigned annotation review system achieving 99.5% accuracy • Automated quality checkpoints accelerating delivery by 19 days per cycle

2024 - 2024

Clinical Diagnostic Imaging QA Initiative - Medical Imaging Project

Classification

Led data labeling quality audit and correction for diagnostic imaging records to reduce annotation errors. Identified critical annotation issues and implemented corrective measures to raise dataset accuracy and reduce variability in downstream diagnostic interpretation. Improved reporting throughput by standardizing data and optimizing labeling-aligned workflows. • Audited 2,700+ diagnostic imaging records for labeling accuracy • Identified and corrected 247 critical annotation errors • Achieved 99.2% data accuracy and reduced interpretation variability by 18% • Accelerated radiologist report generation time by 25% via data standardization

2024 - 2024

Show more

Education

Work History