Joining the Applied Chest Imaging Lab at Brigham and Women’s

in General
Joining the Applied Chest Imaging Lab at Brigham and Women’s

Investigating pulmonary embolism trajectories using machine learning at the Applied Chest Imaging Laboratory.

“The intersection of medical imaging and artificial intelligence is where some of the most exciting healthcare innovations are happening right now.”

What Is the Applied Chest Imaging Laboratory?

The Applied Chest Imaging Laboratory at Brigham and Women’s Hospital is where cutting-edge imaging technology meets lung disease research. The lab is part of Harvard Medical School and focuses on using computational methods to better understand diseases like COPD, pulmonary vascular disease, and interstitial lung disease. What makes this lab special is its commitment to turning complex imaging data into meaningful insights that can actually improve patient care.

Lab Research Advanced research into the bloodflow within the pleural space.

Instead of just looking at CT scans the traditional way, the team here develops sophisticated algorithms and machine learning models to quantify what’s happening in patients’ lungs. It’s essentially teaching computers to see patterns in medical images that might help doctors make better decisions faster.

My Project: Pulmonary Embolism Trajectories

I’m working on a project investigating pulmonary embolism—or PE for short. A PE is a blood clot in the lungs, and it’s actually one of those diagnoses that gets missed or delayed more often than it should, even though it can be life-threatening. The challenge is that radiologists have to look through hundreds of CT images searching for clots, and it’s time-consuming and sometimes things get overlooked.

That’s where machine learning comes in. My work involves using deep learning algorithms to analyze CT pulmonary angiography scans and understand PE trajectories—basically, how these clots develop, change over time, and what patterns might help us predict outcomes for patients.

Why This Matters

The goal isn’t to replace radiologists—it’s to give them better tools. Imagine an AI system that can quickly flag scans that likely have a PE, helping prioritize which cases need urgent attention. Or a model that can predict which patients might have complications based on the characteristics of their clot. That’s the kind of impact this research could have.

What I find most exciting is that this work sits right at the intersection of computer science, medicine, and patient care. I’m writing code and training models, but ultimately, the goal is to help real people get diagnosed and treated faster.

What I’m Learning

Working in this lab has been eye-opening. I’m learning how to:

  1. Work with real medical imaging data and understand the clinical context behind it
  2. Develop and train deep learning models for image analysis
  3. Collaborate with radiologists and clinicians who provide crucial domain expertise
  4. Think critically about how AI tools can actually be integrated into clinical workflows
  5. Navigate the ethical considerations of using AI in healthcare

The best part? I’m getting hands-on experience with technology that’s actually being developed for clinical use, not just theoretical research projects.

Looking Ahead

This position at the Applied Chest Imaging Lab is giving me invaluable experience at the intersection of medicine and technology. As someone planning to go to medical school, understanding how AI can augment clinical decision-making feels like preparing for the future of healthcare. Machine learning and medical imaging aren’t replacing doctors—they’re becoming powerful tools that can help doctors be more effective.

I’m excited to continue this work and see where it leads, both for the research itself and for my own understanding of how technology is reshaping medicine.

Learn More About ACIL

Interested in learning more about the lab’s research? The Applied Chest Imaging Laboratory website can be found here.

Definitions

Pulmonary Embolism (PE)
A blood clot that travels to the lungs and blocks a pulmonary artery, which can be life-threatening and requires prompt diagnosis and treatment.
Deep Learning
A type of machine learning that uses neural networks to learn patterns from data, particularly effective for analyzing complex medical images like CT scans.

© 2023. All rights reserved.