Profile Details

Create Project

★★★★★

☆☆☆☆☆

USD 75 /hr

★★★★★

☆☆☆☆☆

Hire Aondona I.

Nigeria

USD 75 /hr

Medical AI & Deep Learning Consultant | Medical Image Segmentation & Reconstruction | 19+ Publications

Profile Summary

I am a medical AI and imaging researcher specializing in deep learning for medical image analysis, with 19+ peer-reviewed publications and first-author work accepted at MICCAI, the leading international conference in the field. I help researchers, startups, and clinical teams turn imaging data into validated, publication-ready results. My core expertise covers medical image segmentation (brain tumour, breast ultrasound, chest X-ray), MRI reconstruction and low-field MRI, physics-informed deep learning, domain adaptation for low-resource clinical settings, self-supervised pretraining, and model compression (knowledge distillation and quantization) for deployment on limited hardware. I build end-to-end pipelines in PyTorch and train large models on HPC and GPU clusters. Services I provide include: designing and building deep learning models for classification, segmentation, and reconstruction; debugging and improving existing pipelines that fail to converge or generalize; strengthening the machine learning and methods sections of manuscripts and preparing figures and results for publication; reviewing experimental design, evaluation metrics, and reproducibility; and advising on model deployment in resource-constrained or clinical environments. I currently serve as Programme Manager of a pan-African medical AI training programme running across 15+ countries. I have mentored teams to first- and second-place finishes in international medical imaging challenges. I hold an MSc in Medical Physics and a BTech in Physics. I work independently, communicate clearly, and deliver on deadline.

Subject Matter Expertise

• ☆ Biostatistics
• ☆ Data Analysis
• ☆ Deep Learning
• ☆ Statistics
• ☆ Medical Statistics
• ☆ Computational Physics
• ☆ Applied Physics
• ☆ Python
• ☆ Matlab
• ☆ SQL
• ☆ Excel
• ☆ Tensorflow
• ☆ Scikit-learn
• ☆ Medical Devices
• ☆ Medical Device Development
• ☆ Medical Device Safety Testing
• ☆ Medical Imaging/Radiology
• ☆ Nuclear Medicine
• ☆ Magnetic Resonance Imaging (MRI)
• ☆ Medical X-Ray
• ☆ Computed Tomography (CT)
• ☆ Mammography
• ☆ Positron Emission Tomography (PET)
• ☆ Ultrasound

Services

• Writing
• Research
• Consulting
• Data & AI

Writing Medical Writing, Technical Writing, Creative Writing

Research Scientific and Technical Research, Systematic Literature Review

Consulting Healthcare Consulting, Scientific and Technical Consulting

Data & AI Predictive Modeling, Statistical Analysis, Image Processing, Image Analysis, Algorithm Design-ML, Data Visualization, Data Cleaning, Data Processing

Work Experience

Research Assistant

Medical Artificial Intelligence Laboratory (MAI Lab)

February 2025 - Present

Research Programme Manager

SPARK Academy (pan-African medical AI training programme)

February 2024 - Present

Education

MSc (Medical Physics)

University of Lagos

November 2024 - September 2025

B.Tech (Physics)

Federal University of Technology Minna

October 2016 - July 2023

Certifications

• Practical MR Physics for Clinicians

  ESMRMB (European Society for Magnetic Resonance in Medicine and Biology)

  September 2025 - Present

• MedICSS Summer School Medical Image Computing

  University College London (UCL)

  February 2024 - Present

• McMedHacks Medical Image Analysis

  McGill University

  August 2023 - Present

Publications

BOOK CHAPTER

Aondona Iorumbur, Lesly Tsoptio Fougang, Joseph Muthui Wacira, Amal Jlassi, Dong Zhang, Confidence Raymond (2026). A Self-Supervised Framework for Glioma Segmentation Using Swin UNETR .

Abba Mohammed, Zulyadaini Muhammad Aminu, Ummulkhairi Ibrahim, Amina Suleiman Damo, Theodore Barfoot, Alexander Hammers, Raymond Confidence, Aondona Iorumbar, Abdulrazaq Zubair, Mubaraq Yakubu (2026). BRAIN-CATS: Brain Tumour Reliability-Aware Imaging with Neural Networks Using Calibration-Aware Training and Segmentation .

Aondona Iorumbur, Willem P. E. Boonzaier, Farhana Moosa, Kagiso Lebang, Hanifa Jabaar, Dong Zhang, Confidence Raymond (2026). Domain Adaptation for Adult Glioma Segmentation in Sub-Saharan Africa: An Ensemble of nnU-Net v2 and MedNeXt .

Aondona Iorumbur, Freedmore Sidume, Nkuebe Clement Moleko, Botsile Gorata Masalela, Preference Mangwayana, Lame Kaisara, Refilwe Goitsemang, Topo Lefika Rapula, Dong Zhang, Confidence Raymond (2026). LiMSA-UNet: A Lightweight Modality-Selective Attention ResUNet for Brain-Tumor Segmentation .

JOURNAL ARTICLE

Adaobi Emegoakor, Confidence Raymond, Yewande Gbadamosi, Richard Malumba, Charity Umoren, Diana Spence Betancourt, Aondona M. Iorumbur, Chinasa Kalaiwo, Abbas Rabiu Muhammad, Dennis Musinguzi, et al. (2026). Bridging the Gap: A Community Driven and AI-Enabled Approach to Early Breast Cancer Detection in Black African Women . Lecture Notes in Computer Science.

(2025). Implementation of Transfer Learning Approach for AI-Based Diagnosis of Brain Tumors in Low-Resource Settings . Molecular Imaging and Biology.

(2025). Performance Evaluation of Machine Learning Classification of Brain Tumors with WEKA and Python Programming . Molecular Imaging and Biology.

PREPRINT

Ayomide Oladele, Raymond Confidence, Dong Zhang, Charity Umoren, Aondana M Iorumbur, Anu Gbadamosi, Farouk Dako, Maruf Adewole, Udunna Anazodo (2025). Lightweight Brain Tumor Segmentation on Low-Resource Systems: A Step-by-Step Guide with 3D U-Net v2 .

CONFERENCE PAPER

Ilerioluwakiiye Abolade, Aniekan Udo, Augustine Ojo, Abdulbasit Oyetunji, Hammed Ajigbotosho, Aondana Iorumbur, Confidence Raymond, Maruf Adewole(2025). Domain-Adaptive Transformer for Data-Efficient Glioma Segmentation in Sub-Saharan MRI . arXiv

Claudia Takyi Ankomah, Livingstone Eli Ayivor, Ireneaus Nyame, Leslie Wambo, Patrick Yeboah Bonsu, Aondona Moses Iorumbur, Raymond Confidence, Toufiq Musah(2025). How We Won BraTS-SSA 2025: Brain Tumor Segmentation in the Sub-Saharan African Population Using Segmentation-Aware Data Augmentation and Model Ensembling . arXiv