Position Overview

The Department of Dermatology at the Johns Hopkins University School of Medicine is seeking a highly motivated postdoctoral fellow with expertise in machine learning, computer vision, and image analysis to join a multidisciplinary research program focused on developing and applying computational approaches to digital pathology and cancer research.

The successful candidate will work at the intersection of machine learning, computational imaging, and pathology, developing novel algorithms to analyze high-dimensional tissue images and extract quantitative biomarkers of disease response and tumor biology. The position offers an opportunity to work closely with pathologists, computational scientists, and clinical investigators and to contribute to projects spanning translational research and clinical applications.

Research Focus

The fellow will develop and apply machine learning and image-analysis approaches to large-scale digital pathology datasets, with particular emphasis on:

• Development of supervised and deep learning algorithms for whole-slide image (WSI) analysis, particularly on hematoxylin and eosin-stained (H&E) slides
• Automated assessment and quantification of tumor characteristics and tissue classes
• Segmentation, classification, and spatial analysis of tumor and tissue compartments
• Integration of H&E, immunohistochemistry (IHC), multiplex immunofluorescence, and other spatial imaging data
• Development of quantitative tissue-based biomarkers for precision oncology
• Application of machine learning to translational and clinical research datasets
• Validation and optimization of computational biomarkers across independent patient cohorts
• Exploration of multimodal approaches integrating pathology images with clinical and molecular data

The fellow will have opportunities to develop independent research projects, publish in high-impact scientific journals, present work at national and international conferences, and collaborate with investigators across Johns Hopkins and external academic and industry partners.

Responsibilities

The successful candidate will:

• Develop, implement, and validate machine learning models for image analysis
• Design computational pipelines for processing and analyzing digital pathology datasets
• Perform image segmentation, feature extraction, classification, and spatial analysis
• Develop reproducible computational workflows for large-scale image analysis
• Collaborate closely with pathologists to translate biological and clinical questions into computational approaches
• Analyze and interpret complex imaging and clinical datasets
• Contribute to manuscript preparation, grant applications, and scientific presentations
• Mentor graduate students and research staff as appropriate

Research Environment

This position offers an exceptional opportunity to work within the highly collaborative research environment of the Johns Hopkins University School of Medicine, with access to extensive clinical, pathology, imaging, and translational research resources.

The fellow will work closely with expert pathologists and clinical investigators while collaborating with computational scientists and researchers across disciplines. The position is particularly well suited to an individual interested in developing clinically meaningful AI and computational pathology approaches that can ultimately improve cancer diagnosis, treatment assessment, and personalized patient care.

Application

Applicants should submit:

• Curriculum vitae
• Cover letter describing research interests and relevant experience
• Brief statement of research interests and career goals
• Names and contact information for three references

Applicants are encouraged to include links or examples of relevant publications, software, GitHub repositories, or other computational work.

Required Qualifications

• PhD in Machine Learning, Computer Science, Computational Biology, or a closely related field
• Strong training in machine learning and deep learning, with hands-on experience training and fine-tuning deep neural networks (not just using pretrained models)
• Demonstrated experience with image classification and/or segmentation, ideally in a digital pathology or biomedical imaging context
• Strong programming skills in Python
• Experience with deep learning frameworks such as PyTorch (preferred), TensorFlow, or equivalent
• Experience with whole-slide image (WSI) handling (e.g., OpenSlide, QuPath, ASAP) and gigapixel-scale image processing
• Experience with convolutional architectures (e.g., ResNet, EfficientNet, ConvNeXt) and evaluating/selecting backbones for downstream tasks such as segmentation
• Experience working with large image datasets and developing reproducible, version-controlled computational pipelines
• Strong quantitative, analytical, and problem-solving skills
• Excellent written and verbal communication skills
• Demonstrated ability to work independently and collaboratively in a multidisciplinary research environment

Preferred Qualifications

Experience in one or more of the following areas is highly desirable:

• Digital pathology or whole-slide image analysis
• Histopathology or biomedical image analysis
• Cell and tissue segmentation
• Multiplex immunofluorescence or spatial biology
• Weakly supervised, self-supervised, or foundation-model approaches
• Vision transformers or other modern computer vision architectures
• Image registration and multimodal image analysis
• Computational oncology
• Experience with high-performance computing or cloud-based analysis
• Experience publishing machine learning or computational imaging research in peer-reviewed journals