Postdoc Fellow: AI for Multimodal Neuroimaging & Omics

University of Maryland, Baltimore (UMB); Baltimore (MD); On-site; USD 60,000 - 90,000; Full time

Job summary

The University of Maryland, Baltimore (UMB) is seeking a Postdoctoral Fellow to advance statistical methods for multimodal neuroimaging, multi-omics, EHR, and genetics data. The role involves applying machine learning and AI to uncover interactive relationships across aging and development, and implementing intensive algorithms on high-performance clusters.

Responsibilities include developing AI/ML-driven causal inference models for real-world evidence, integrating large datasets such as UKBB,

Qualifications

• Doctorate or terminal degree in the field of specialization.
• May require unrestricted license in the State of Maryland.

Responsibilities

• Develop statistical methods to model multimodal neuroimaging, multi-omics, EHR and genetics data.
• Apply machine learning and AI to identify interactive relationships between imaging modalities and omics during aging and development.
• Run computationally intensive algorithms on high-performance clusters.
• Develop AI/ML-based causal inference models for real-world evidence applications (e.g., target trial emulations).
• Develop data integration methods to combine large datasets (UKBB, HCP, ENIGMA, UMMS EHR).
• Draft and prepare manuscripts.
• Perform other duties as assigned.

Skills

Statistical methods

Machine learning

AI approaches

Data integration

Neuroimaging

Genomics & EHR

Education

Doctorate or terminal degree

Job description

The University of Maryland, Baltimore (UMB) is seeking a Postdoctoral Fellow to advance statistical methods for multimodal neuroimaging, multi-omics, EHR, and genetics data. The role involves applying machine learning and AI to uncover interactive relationships across aging and development, and implementing intensive algorithms on high-performance clusters.

Responsibilities include developing AI/ML-driven causal inference models for real-world evidence, integrating large datasets such as UKBB,