about

I am a medical doctor and independent researcher at ML Collective, working at the intersection of medical imaging, artificial intelligence, and computational medicine. My work focuses on building data-efficient and clinically useful learning systems for imaging, especially in resource-constrained environments.

My research interests include medical imaging AI, MRI and low-field MRI, neuroradiology, nuclear medicine, representation learning, foundation models, self-supervised learning, and imaging in underserved populations. In the long term, I hope to contribute to equitable AI systems for medical imaging and precision medicine. See my publications or projects to explore this work.

publications

Region-Aligned Training and Target-Specific 2D--3D nnU-Net Fusion for BraTS 2026 Brain Metastasis Segmentation

A. Mojtahedzadeh, O.I. Akinmuleya, J.O.I. Daniel, M. Adinavi

Improved Neuroanatomy Segmentation and Quality Assurance in Pediatric Ultra-Low-Field MRI

O.I. Akinmuleya, T. Musah

MYRA: Multi-domain Hybrid MRI Representation Architecture for Self-Supervised Low-Field MRI

O.I. Akinmuleya, I.K. Koomsan, C.N. Meschack-Augustin, L.N. Shambangu

NeuroAIHub: An AI-Driven Framework for Automated Curation and Discovery of Neuroradiology Datasets

B. Gheiji, S. Moradi, M. Vatanparast, S. Shokrani, M.A. Bagherzadeh, D. Elyassirad, O.I. Akinmuleya, S. Goodarzi, M. Heidari-Foroozan, M. Moassefi, J.D. Rudie, E. Calabrese, R. Jain, S. Faghani

  • American Journal of Neuroradiology
  • Accepted Manuscript
  • 2026

68Ga-PSMA PET/CT/MRI in the Initial Diagnosis and Staging of Prostate Cancer: A Review

O.I. Akinmuleya, P.F. Cohen, K. Kairemo

  • Advances in Radiotherapy & Nuclear Medicine, 2(4), 4590
  • Published
  • 2024

presentations

Quantifying Amyloid Burden in Alzheimer's Disease: A Reproducible Centiloid Pipeline for Analysis in Resource-Constrained Settings

Alzheimer's Association International Conference (AAIC) · Abstract ID: #4997

  • Poster Presentation
  • July 2026

Training 105 African Clinicians and Students in Open Neuroimaging Analysis: CAMERA's Dimensional Research Program

International Society for Magnetic Resonance in Medicine (ISMRM) Annual Meeting

  • Poster Presentation
  • May 2026

NeuroAI: An AI-Driven Framework for Automated Curation and Discovery of Neuroradiology Datasets

American Society of Neuroradiology (ASNR) Annual Meeting, Philadelphia, USA

  • Oral / Poster Presentation
  • 2025

The Role of 68Ga-PSMA PET/CT and PET/MRI in the Diagnosis of Primary Prostate Cancer and TNM Staging

International Conference of Hybrid Imaging, International Atomic Energy Agency (IAEA), Vienna

  • Poster Presentation
  • 2024

projects

In-hospital Mortality Prediction

End-to-end ML pipeline predicting in-hospital mortality, with a Streamlit app deployed on Google Cloud, CI/CD via GitHub Actions, and live model monitoring using Evidently AI and Grafana.

  • MLflow
  • Mage
  • Docker
  • GCP
  • Evidently AI
  • Grafana
  • Streamlit

Computer Vision Projects

A collection of computer vision work: a real-time face recognition system (MTCNN + InceptionResnetV1) and a medical imaging model for polyp detection in gastrointestinal images, fine-tuned on YOLOv11 to assist early diagnosis.

  • PyTorch
  • OpenCV
  • YOLOv11
  • MTCNN
  • InceptionResnetV1

Logos

An end-to-end RAG application for scripture-based Q&A, with an evaluation pipeline comparing keyword, vector, and hybrid search using hit rate and MRR for retrieval optimization.

  • RAG
  • Elasticsearch
  • MiniSearch
  • OpenAI API

Data Analyst AI Agent

An AI-powered data analyst agent that performs automated data exploration, visualization, and analysis from natural-language queries, built with LangGraph to interpret requests and generate insights.

  • LangGraph
  • OpenAI API
  • Agents

LLM Finetuning

Fine-tuned Meta's LLaMA 3 on a medical QA dataset for stronger clinical and diagnostic accuracy, using Unsloth to accelerate training and LoRA / PEFT to cut compute and memory cost.

  • LLaMA 3
  • Unsloth
  • LoRA
  • PEFT

news

September 2026 Our paper, Evaluating the Generalization of Neuroimaging Foundation Models to African Clinical MRI, won the Best Paper Award at AFRICAI 2026, a MICCAI satellite event. Congratulations to the team!
August 2026 Our paper, MYRA: Multi-domain Hybrid MRI Representation Architecture for Self-Supervised Low-Field MRI, was accepted at the MIRASOL 2026 MICCAI satellite event. Congratulations to the team!
August 2026 Our paper, Evaluating the Generalization of Neuroimaging Foundation Models to African Clinical MRI, was accepted at AFRICAI 2026, a MICCAI satellite event. Congratulations to the team!
August 2026 Our paper, Improved Neuroanatomy Segmentation and Quality Assurance in Pediatric Ultra-Low-Field MRI, was accepted at the MICCAI LISA Challenge 2026. Congratulations to the team!
August 2026 A paper I contributed to, Region-Aligned Training and Target-Specific 2D–3D nnU-Net Fusion for BraTS 2026 Brain Metastasis Segmentation, was accepted at the BraTS 2026 workshop. Congratulations to the team!
July 2026 Our work, Quantifying Amyloid Burden in Alzheimer’s Disease: A Reproducible Centiloid Pipeline for Analysis in Resource-Constrained Settings, was presented at the Alzheimer’s Association International Conference (AAIC).
May 2026 Welcome to my personal website. I will share updates, publications, and selected work here.
May 2026 Our work, Training 105 African Clinicians and Students in Open Neuroimaging Analysis: CAMERA’s Dimensional Research Program, was presented at the International Society for Magnetic Resonance in Medicine (ISMRM) Annual Meeting.

contact

I’m always open to conversations and collaborations around medical imaging, artificial intelligence, and computational medicine. Whether you’re interested in discussing research, exploring a potential collaboration, or simply exchanging ideas at the intersection of technology and healthcare, feel free to get in touch.