PhD students

1. PhD students are welcome to write to Dr. Shah directly for thesis research on; a) interpreting and developing novel generative and predictive deep neural networks for interrogating proximity relationships between biological pathways and image features to discover disease sub-types and real-world clinical deployment; b) formulating novel unsupervised, on, and off-policy AI learning methods for clinical data. **The standard process for all interested applicants is to first formally apply here: **https://apply.grad.uci.edu/apply/.
2. External students interested in computer science Ph.D. graduate studies can select Dr. Shah as the faculty PI and apply to the Electrical Engineering and Computer Science (https://engineering.uci.edu/dept/eecs/academics/graduate) program. Engineering students can apply to the Biomedical Engineering (https://engineering.uci.edu/dept/bme/graduate/prospective/admission-requirements) PhD program. Students interested in interdisciplinary studies in computation + wet lab research can apply to the Gateway PhD Graduate Program in Cellular & Molecular Biosciences (CMB).

MD-PhD students

Physician scientist (MD-PhD) students and applicants to the Medical Scientist Training Program interested in clinical informatics + molecular research are welcome to write to Dr. Shah directly to determine appropriate PhD component projects.

1. Undergraduate students interested in quarterly deep learning research methods or an UROP project are welcome to contact Dr. Shah directly and/or apply either to the https://undergraduate.bio.uci.edu/bio-197-198-199/ or the https://urop.uci.edu/urop-opportunities/ programs and indicate Prof. Shah as faculty advisor.
2. Interns and visiting scientists, please propose a topic and research plan that is relevant to the group.

Postdoctoral Scholar in Deep Learning & Medical Imaging- Position 1

The computational medicine lab led by Dr. Pratik Shah at UCI invites applications for a Postdoctoral Scholar position from outstanding Ph.D., MD or foreign equivalent applicants with strong academic computer science and informatics research backgrounds. The lab is interested in developing novel AI-based medical imaging technology for digital biopsies.

Applicants experienced in solving critical challenges in medical data validation methods, training AI models for enabling diagnostic tests, their replicability, and proof of their clinical utility will be given preference. The scholar will be trained to develop and interpret generative deep neural networks (DNNs) for automating tissue-based analyses, interrogating basic biological features and proximity relationships between tissue constituents for digital biopsies. Responsibilities include managing, processing and visualizing high-parameter medical images (e.g., pathology, MRI, CT, RGB etc.), associated clinical labels, and molecular profiles for training, validation, and interpretations of interpretable DNNs.

Opportunities to publish in leading biomedical journals and machine learning conferences, networking with government funding agencies, industry partners, foundations, and academic experts. Training in fellowship writing, teaching/mentoring, oral presentations and review of manuscripts will be provided.

Qualifications

Basic qualifications
• Appointment as a Postdoctoral Scholar requires a doctoral degree (e.g., Ph.D., MD) or the foreign equivalent in computer science, biomedical engineering or statistics

Preferred qualifications (at time of application)

• In-depth knowledge of theory and applications of computer vision, and statistical machine learning in image processing
• Previous experience in leveraging deep learning libraries (e.g., OpenCV, Theano, Caffe, Keras and TensorFlow) and machine learning datasets and model inference from AlexNet, ImageNet, MNIST, etc.
• Experience in writing software in a team-oriented environment with version control, issue tracking, and code review with Python (Scikit-Learn Numpy, and Pandas), MATLAB and C++
• Track record of writing and publishing research papers in peer-reviewed journals or top machine learning conferences
• Strong analytical and organizational skills; detail-oriented
• Committed to mentoring others

A reasonable estimate for this position is $60,000-$71,952

Substantive inquiries about the position should be directed to:
Pratik Shah PhD
pratik.shah@uci.edu

Interested candidates should complete the application profile and provide a CV, cover letter including career motivation and why you are interested in joining the lab.

TO APPLY: Please log onto UC Irvine’s RECRUIT located at: https://recruit.ap.uci.edu/JPF08836

Postdoctoral Scholar in Deep Learning & Uncertainty Estimations from Medical Data-Position-2

The computational medicine lab led by Dr. Pratik Shah at UCI invites applications for a Postdoctoral Scholar position from outstanding Ph.D, MD or foreign equivalent applicants with strong academic computer science, biomedical informatics or statistics backgrounds. The lab is interested in developing novel AI-based technology for quantification and real-world validation of clinical decision making.

Applicants experienced in in solving critical challenges for developing biologically informed statistical methods and uncertainty estimations for training deployable AI models that establish causal relationships in clinical data will be given preference. The scholar will be responsible for training deep generative reinforcement AI models for learning predictive and prescriptive clinical decision-making from time-varying electronic medical records. You will formulate novel statistical methods for uncertainty quantification of patients’ outcomes and use unsupervised learning to discover biologically relevant disease subtypes. Determine feature importance for validating predictions and estimations from fully trained deep neural network models for correlative and matrixed molecular sequencing data.

Opportunities to publish in leading journals and machine learning conferences, networking with government funding agencies, industry partners, foundations, and leading academic experts. Training in fellowship writing, teaching/mentoring, oral presentations and review of manuscripts will be provided.

Qualifications

Basic qualifications

• Appointment as a Postdoctoral Scholar requires a doctoral degree (e.g., Ph.D., MD) or the foreign equivalent in computer science, biomedical informatics or statistics

Preferred qualifications (required at time of application)

• Experience in applied statistics i.e., probabilistic models and Bayesian models for uncertainty quantification, causality estimates for explainable deep learning, reinforcement learning.
• Experience in open-source deep learning frameworks such as TensorFlow or PyTorch
• Experience using Python, C++, and Java, Linux
• Experience in writing software in a team-oriented environment with version control, issue tracking, and code review with Python (Scikit-Learn Numpy, and Pandas, MATLAB and C++)
• Familiarity with data processing techniques for text and time series clinical data, and knowledge of MySql, MongoDb, databases.
• Track record of writing and publishing research papers in peer-reviewed journals or top machine learning conferences
• Strong analytical and organizational skills; detail-oriented
• Committed to mentoring others.

A reasonable estimate for this position is $60,000-$71,952

Substantive inquiries about the position should be directed to:
Pratik Shah PhD
pratik.shah@uci.edu

Interested candidates should complete the application profile and provide a CV, cover letter including career motivation and why you are interested in joining the lab.

TO APPLY: Please log onto UC Irvine’s RECRUIT located at: https://recruit.ap.uci.edu/JPF08838

Emergency Medicine and Critical Care Collaborators:

Attending clinicians (MDs), residents and fellows are welcome to collaborate with us as we create new clinical models driven by artificial intelligence (AI) to analyze data from hundreds of thousands of de-identified EMR records from emergency and intensive care departments nationwide. Clinical collaborators will work with computer scientists to evaluate the data for early diagnosis of diseases, therapeutic choices, and evaluating confounders in treatment effects of medications. You will be able to work remotely to review the data via a secure online web portal, but are welcome spend time in lab. Clinicians may also collaborate as authors on research publications from this project and learn machine-learning and informatics skills. We welcome clinicians interested in developing cutting-edge health related AI technologies by using clinical principles.

Clinical and Anatomic Pathology Collaborators:

Attending clinicans (MDs), residents and fellows are welcome to collaborate with us as we create new clinical models driven by artificial intelligence (AI) to analyze data from hundreds of thousands of de-identified histopathology images and molecular panels from nationwide hospitals. Medical collaborators will work with computer scientists to clinically evaluate the images for different purposes such as early diagnosis of diseases, therapeutic choices, tumors, genomic profiles, and patient stratification. You will be able to work remotely to review the images via a secure online web portal, but are welcome spend time in lab. Clinicians may also collaborate as authors on research publications from this project and learn machine-learning and informatics skills. We welcome clinicians interested in developing cutting-edge health related AI technologies by using clinical principles.