We are seeking a Staff Consultant AI/ML, Medical Imaging with expertise in developing, fine-tuning, validating, and benchmarking AI models for medical imaging applications. The role involves optimizing pre-trained deep learning models for tumor segmentation and risk prediction, developing robust inference pipelines, and conducting rigorous evaluation against clinical ground truth and patient outcomes. The ideal candidate will have hands-on experience with medical imaging AI frameworks such as MONAI, nnU-Net, PyTorch, and strong knowledge of model validation, generalizability testing, and performance optimization in healthcare environments
Location: - Remote ( Willing to travel to client locations as needed) / CST work hours
Education: - Engineering Degree BE / ME / BTech / M Tech / B.Sc. / M.Sc.
Tentative Duration - 6 months +
Key Responsibilities
Model Development & Optimization
Fine-tune pre-trained deep learning models for medical image segmentation and predictive analytics applications.
Perform hyperparameter tuning and model optimization to improve accuracy, robustness, and computational efficiency.
Implement transfer learning strategies for imaging datasets with limited annotations.
Develop and enhance AI models using frameworks such as nnU-Net, MONAI, PyTorch, and related toolsets.
Support model retraining and optimization based on validation outcomes.
Segmentation Model Evaluation
Evaluate tumor and lesion segmentation models against expert clinician delineations.
Measure model performance using metrics such as:
o Dice Similarity Coefficient (DSC)
o Intersection over Union (IoU) o Hausdorff Distance
o Sensitivity and Specificity o Precision and Recall
Analyze segmentation errors and recommend model improvements.
Support clinical validation activities and performance reporting. Risk Prediction & Outcome Analytics
Develop and evaluate risk prediction models using imaging and clinical data.
Assess model predictions against patient outcome labels and clinical endpoints.
Support feature engineering and extraction from imaging datasets.
Perform statistical analysis of model outputs and predictive performance.
Evaluate model calibration, discrimination, and robustness. Inference Pipeline Development
Design and implement scalable inference pipelines for medical imaging workflows.
Optimize preprocessing, inference, and post-processing components.
Support deployment readiness and integration into research or clinical environments.
Improve inference efficiency, scalability, and reproducibility. Validation & Benchmarking
Design and execute evaluation frameworks for model testing and validation.
Conduct benchmarking studies against baseline models and industry standards.
Perform cross-validation and independent test set evaluations.
Create technical documentation and validation reports. Generalizability Testing
Assess model performance across multiple institutions, scanners, imaging protocols, and patient populations.
Perform external validation and domain adaptation assessments.
Identify sources of model bias and performance degradation.
Recommend tuning strategies to improve model generalizability and robustness. Collaboration & Delivery Support
Collaborate with data scientists, radiologists, clinicians, and technical teams.
Translate clinical requirements into AI model development tasks.
Contribute to project planning, technical reviews, and delivery milestones.
Present findings, evaluation results, and model performance insights to stakeholders
Required Skills
AI/ML & Deep Learning
Strong understanding of machine learning and deep learning concepts.
Hands-on experience with:
o PyTorch
o MONAI
o nnU-Net
o TensorFlow (preferred)
Experience with transfer learning and fine-tuning pre-trained models.
Hyperparameter optimization techniques.
Model evaluation and validation methodologies. Medical Imaging
Knowledge of: o CT o MRI o PET o Digital Pathology
Experience working with DICOM datasets.
Understanding of tumor segmentation workflows.
Familiarity with radiomics and imaging biomarkers. Model Validation & Analytics
Segmentation performance evaluation.
Predictive modeling and risk assessment.
Statistical analysis and model benchmarking.
Generalizability and external validation testing.
Explainable AI and model interpretability concepts. Programming & Tools
Python programming.
NumPy, Pandas, Scikit-learn.
Version control using Git.
Experience with cloud-based AI environments is preferred.
Experience Required
Required Experience
10+ years of overall experience in AI/ML, data science, or medical imaging analytics.
5+ years of hands-on experience in healthcare or life sciences AI applications.
Proven experience developing and deploying medical image segmentation solutions.
Experience leading technical delivery teams and client-facing engagements.
Never miss an opportunity! Create an alert based on the job you applied for.