"Data Architect Senior"

The Machine Learning in Neurosurgery (MLiNS) Laboratory in the Department of Neurosurgery at the University of Michigan studies how machine learning and artificial intelligence can be used to improve the diagnosis and treatment of patients with neurosurgical diseases. We work at the intersection of AI, biomedical research, and clinical medicine. The applications of AI in medicine have grown tremendously over the previous decade. AI systems are now used for early cancer detection, computer-aided diagnosis, automated tumor localization, and personalized treatment recommendations. Our team aims to translate the latest advances in AI, machine learning, and computer science to improve the diagnosis and treatment of patients with neurological and neurosurgical diseases.

The MLiNS lab has an immediate opening for a Data Scientist/Machine Learning Engineer who will be responsible for data management, machine learning pipelines, and software development to investigate the role of AI-based computer vision methods on optical and radiologic images. Our laboratory focuses on (1) intraoperative imaging with label-free optical microscopy and (2) neuro-imaging such as computed tomography (CT) and magnetic resonance imaging (MRI). The position will include training medical foundations models using vision-language modeling and self-supervision to solve a wide range of downstream diagnostic and predictive tasks. These models will be tested in a clinical setting to improve the diagnosis and treatment of brain tumor patients. The candidate will be an essential part of our team due to the increasing importance of high-quality data for developing ML models and making new scientific breakthroughs. The candidate will work with a team of data scientists, postdoctoral/doctoral students, medical students, and research scientists to generate and publish high-impact machine learning research.

• Collection, curation, and maintenance of medical imaging datasets
• Preprocessing optical and radiologic imaging data
• SQL-based database development for clinical, pathologic, and radiologic data
• Data quality assurance
• Training computer vision and large language models on biomedical datasets
• Validation/testing of models on multi-institutional datasets
• Communicate results via publications, conference proceedings, and academic conferences

M.S. or higher in Computer Science, Data Science, Computer Engineering, Biomedical Engineering, Bioinformatics or related fields.

• Expertise in Python programming
• Proficiency in Linux, Bash, SQL, Git
• Proficiency with high-performance computing, including SLURM
• Proven track record of executing data science projects
• Ability to work effectively with a team of scientists and students

Experience with machine learning/deep learning and computer vision techniques

• Experience in medical image analysis of 2D or 3D image data
• Publications in top AI conferences (CVPR, NeurIPS) or biomedical journals (Nature, Nature Medicine)
• Knowledge of the UM computing environment

Days / 40 hours

Details of our laboratory and research can be found at our lab website here.