Be You. Be Bold. Choose Duke.

Be You.

Be Bold.

What You'll Do

Data Acquisition & Management

• Organize, curate, and maintain large-scale ophthalmic imaging datasets, including OCT, OCTA, and ultra-widefield fundus imaging.
• Interface with imaging platforms and data export systems to support efficient data acquisition processes.
• Maintain dataset structure, file naming standards, metadata integrity, version control practices, and audit documentation.
• Develop and manage databases that integrate imaging, clinical, and research information.
• Perform data cleaning, quality assurance, and standardization activities to promote reproducible research.

Image Processing & Annotation

• Preprocess ophthalmic imaging data for machine learning workflows through normalization, resizing, artifact reduction, and quality assessment.
• Coordinate imaging annotation and segmentation efforts with clinicians and research personnel.
• Develop documentation, annotation guidelines, and quality control procedures.
• Maintain consistency and accuracy across imaging datasets and labeling processes.

Machine Learning & Model Development

• Design, develop, optimize, and support machine learning and deep learning models.
• Prepare datasets for training, validation, and testing while ensuring data integrity and reproducibility.
• Execute model training workflows and monitor performance metrics.
• Perform hyperparameter optimization and maintain experiment tracking processes.
• Collaborate with investigators and research teams to implement sound validation methodologies and statistical approaches.

Model Evaluation & Validation

• Evaluate model performance using industry-standard measures such as ROC curves, AUC, sensitivity, specificity, and precision-recall analyses.
• Assess model generalizability across diverse study populations and research cohorts.
• Identify potential sources of bias, overfitting, and data leakage.
• Create visualizations and summaries that communicate model performance to technical and clinical audiences.

Research Data Analysis & Clinical Integration

• Integrate imaging biomarkers with clinical, cognitive, and genetic datasets.
• Support exploratory analyses and statistical research projects.
• Generate tables, figures, reports, and analytical summaries for presentations and publications.
• Translate complex analytical findings into clinically meaningful insights.

Pipeline Development & Automation

• Develop scalable and reproducible workflows for data preprocessing, machine learning training, validation, and evaluation.
• Automate routine quality assurance and data management processes.
• Apply version control and software development best practices for research environments.
• Improve workflow efficiency and computational performance across research computing platforms.

Regulatory Compliance & Data Security

• Support compliance with institutional policies, research protocols, and privacy requirements.
• Assist with data de-identification and secure data management practices.
• Maintain documentation supporting audits, protocol adherence, and research reproducibility standards.

Collaboration & Scientific Communication

• Partner with multidisciplinary teams across clinical, scientific, and technical disciplines.
• Present research updates and analytical findings during laboratory meetings and project discussions.
• Document workflows, methodologies, and technical processes to support knowledge sharing and reproducibility.
• Mentor students, trainees, and new team members as appropriate.

Scholarly Contributions

• Assist with manuscripts, abstracts, scientific presentations, and grant-related research activities.
• Create scientific figures, model visualizations, and technical documentation.
• Contribute to methods development and reporting for peer-reviewed publications and conference submissions.

Preferred Advanced Experience

• Experience implementing emerging machine learning architectures and advanced AI methodologies.
• Knowledge of multimodal modeling approaches integrating imaging, genetic, and clinical datasets.
• Familiarity with explainable AI techniques, including Grad-CAM and saliency mapping.
• Ability to contribute to study design, hypothesis generation, and innovative research initiatives.

Why Duke?

Choose Duke.

Duke is an Equal Opportunity Employer committed to providing employment opportunity without regard to an individual's age, color, disability, gender, gender expression, gender identity, genetic information, national origin, race, religion, sex (including pregnancy and pregnancy related conditions), sexual orientation or military status.