Senior Data Scientist - AI and ECG

About the role

The Division of Cardiovascular Medicine at the University of Michigan is expanding a nationally and internationally funded research program at the cutting edge of cardiovascular medicine. Led by Drs. Venkatesh Murthy and Sascha Goonewardena, whose work has appeared in leading journals including NEJM, AI, JAMA, and Circulation, the program fuses artificial intelligence, advanced cardiac imaging, multiomics (proteomics, metabolomics, and genomics), and cardiometabolic disease biology to develop precision diagnostics and identify novel therapeutic approaches, with a particular focus on coronary microvascular disease and other cardiovascular conditions that disproportionately affect women and remain poorly served by existing diagnostic tools.

Responsibilities

• Develop, train, and refine deep learning foundation models (including transformer and self-supervised architectures) for ECG-based detection and endotyping of cardiometabolic disease
• Assemble, harmonize, clean, and catalog large-scale, heterogeneous ECG datasets from multiple internal and external sources and formats
• Build and maintain reproducible data processing pipelines, data loaders, and PostgreSQL-based data management infrastructure
• Evaluate model performance, document methods, and prepare written and code-based analytical reports (Python)
• Contribute to manuscripts, grant reports, and presentations
• Collaborate with collaborative network partners and other research collaborators on code, computational workflows, and data harmonization

Requirements

• Masters or doctoral degree in computer science, electrical engineering, biomedical engineering, computational biology, statistics with an AI/ML focus, data science, or a closely related quantitative field
• Demonstrated experience developing and training deep learning models, preferably in a biomedical or physiological signal context
• Proficiency in Python and deep learning frameworks (PyTorch preferred; TensorFlow acceptable)
• Experience with large-scale, heterogeneous data harmonization and processing pipelines spanning multiple source formats
• Familiarity with self-supervised, semi-supervised, or foundation model architectures (e.g., transformers, masked autoencoders)
• Proficiency in SQL, preferably PostgreSQL, for data querying and management
• Experience with high-performance computing environments, including Slurm-based job scheduling
• Strong organizational skills and attention to detail
• Ability to prepare and present written and code-based (Python or R) analytical reports
• Strong scientific communication skills; ability to contribute to manuscripts and grant reports

Qualifications

• Experience with electrocardiographic (ECG) or other physiological waveform data
• Experience with multimodal data integration, particularly combining physiological signals with cardiac imaging data
• Familiarity with clinical data infrastructure (EHR, DICOM, HL7/FHIR)
• Experience with transfer learning or domain adaptation across heterogeneous datasets
• Track record of peer-reviewed publications or preprints in machine learning, AI, or biomedical informatics
• Familiarity with cardiovascular physiology or cardiology research
• Experience with Git and reproducible research practices
• Familiarity with cloud computing environments (AWS, GCP, or Azure)

Desired Qualifications

• Experience with electrocardiographic (ECG) or other physiological waveform data
• Experience with multimodal data integration, particularly combining physiological signals with cardiac imaging data
• Familiarity with clinical data infrastructure (EHR, DICOM, HL7/FHIR)
• Experience with transfer learning or domain adaptation across heterogeneous datasets
• Track record of peer-reviewed publications or preprints in machine learning, AI, or biomedical informatics
• Familiarity with cardiovascular physiology or cardiology research
• Experience with Git and reproducible research practices
• Familiarity with cloud computing environments (AWS, GCP, or Azure)

Pay

Commensurate with experience.

Schedule

Monday through Friday, standard business hours. This is an onsite position.