---

This position is available to enrolled Northeastern University students who are currently eligible for Co-op January - June 2025.

Job Summary

Are you a graduate student interested in image analysis using deep learning methods? The Computational Sciences department is seeking a motivated student intern who will work under the supervision of a Sr. Scientific Software Engineer. The intern will focus on analyzing biomedical images including but not limited to Whole Slide images (WSI), Fluorescence images, etc., using deep learning techniques.

The goal of this project is to apply state-of-the-art deep learning tools to detect/segment features in microscopy images, like finding features in the eyeball image or classifying aging in the kidney images, etc., and develop a scalable and reproducible image analysis framework that can handle large dataset and deliver reliable results. The intern will work closely with the computational science and imaging sciences teams to ensure that the project is a success.

The ideal start date for this position is January 6, 2025.

Key Responsibilities

• Work closely with researchers to annotate data to create a training dataset.
• Apply state-of-the-art deep learning techniques including transfer learning to create task-specific models. For example using Vision Transformers, YOLOV, UNet, etc.
• Scaling transfer learning to operate on WSI using parallelization.
• Work with a new or existing vision transformer methodology to train a new objection detection or segmentation model and deploy it to Jax Image tools.
• Develop a strategy to apply the probability output of deep learning networks to find the high probable regions of interest, for instance, the YOLO model identifying features in WSI:
  • Generate training set based on the CS/MS group's specifications, ideally within the capabilities of JAX Image Tools software
  • Use JAX HPC resources to retrain all weights of the full network
  • Investigate how finetuning the hyperparameters improve results
  • Use of CUDA technology.
  • Investigate running classification on low-power devices, such as RaspberryPi.

Desired Qualifications

JAX encourages applications from a diverse mix of educational backgrounds and experiences. Preferred qualifications include:

• Must be enrolled in a college or university program
• Pursuing a Master's degree in computer science, biology, electrical engineering, or related field
• Experience with PyTorch or other ML framework
• Experience with programming languages (i.e. Python, Java, Typescript)
• Interest and familiarity with ML and/or Image Analysis
• Familiarity with CUDA programming is a plus

The hourly pay range is $17.50 - $35.00.

About JAX:

The Jackson Laboratory is an independent, nonprofit biomedical research institution with a National Cancer Institute-designated Cancer Center and nearly 3,000 employees in locations across the United States (Maine, Connecticut, California), Japan and China. Its mission is to discover precise genomic solutions for disease and empower the global biomedical community in the shared quest to improve human health.

Founded in 1929, JAX applies over nine decades of expertise in genetics to increase understanding of human disease, advancing treatments and cures for cancer, neurological and immune disorders, diabetes, aging and heart disease. It models and interprets genomic complexity, integrates basic research with clinical application, educates current and future scientists, and provides critical data, tools and services to the global biomedical community. For more information, please visit www.jax.org .

EEO Statement:

The Jackson Laboratory provides equal employment opportunities to all employees and applicants for employment in all job classifications without regard to race, color, religion, age, mental disability, physical disability, medical condition, gender, sexual orientation, genetic information, ancestry, marital status, national origin, veteran status, and other classifications protected by applicable state and local non-discrimination laws.