From Stroke Detection to Cancer Screening: AI Tools Advancing Imaging Efficiency

Mass General Brigham has launched several initiatives to help address the overwhelming influx of artificial intelligence (AI) tools in healthcare, particularly radiology.

Subscribe to the latest updates from Neuroscience Advances in Motion

Error: Please enter a valid email address.

Email Address Submit

"There's a big gap between developing and training a model and an AI tool during research and developing a device that's ultimately going to be approved by the FDA, commercialized, and used by healthcare providers in patient care," says Bernardo Bizzo, MD, PhD, MSc , a radiologist at Massachusetts General Hospital, and director of the senior director of the Digital Clinical Research Organization at Mass General Brigham AI. "It's really hard to get it to market and get it used by hospitals."

Efforts such as the Mass General Brigham AI Governance Committee, Healthcare AI Challenge, and AI Clinical Research Organization aim to tackle the complex challenges of developing and implementing AI tools in clinical practice.

Training a Convolutional Neural Network to Improve Detection of Acute Infarct

"The technology we use to develop AI models in the industry today is based on convolutional neural networks (CNNs)," Dr. Bizzo explains. CNNs are a type of deep learning algorithm that uses data to learn patterns and perform tasks. They are often used for object and image recognition, classification, and detection, which makes them particularly useful in developing tools for radiology. "They are what we now call narrow AI tools, as they are focused on one specific task for one specific modality, one specific body part."

One example is a tool he and colleagues recently developed to detect acute infarcts on head computed tomography (CT) scans. CT is the imaging technology most commonly accessible in average hospital and emergency department settings, but the test is insensitive for this medical emergency, particularly in its early phases. However, the finding is easily seen with magnetic resonance imaging (MRI), which is not always available.

Dr. Bizzo and colleagues used a unique data set of patients at Mass General Brigham who had acute infarct with both CT and MRI findings. Using the data, they trained a CNN to detect acute infarct not yet visible to the human eye using CT. It outperformed expert neuroradiologists in detecting acute infarcts, with a sensitivity of 96% versus 61% to 66%. The groundbreaking findings were published in Scientific Reports in 2023. The researchers' industry partner is currently preparing to submit the tool for FDA review.

Challenges Facing AI in Radiology

That innovative solution is considered to be a success from a research perspective, and it could be a very useful tool in radiology practice, Dr. Bizzo says. However, it is a drop in a large bucket, and there are several hurdles to implementing such a tool into clinical practice.

"That work represents one single finding of a radiologist. Each one of those models, each one of those products, is just doing one task. But when I look at a head CT, there are hundreds of findings that I have to look at to make a diagnosis," he explains. "There are about 4,000 findings that would need to be addressed by these narrow AI models to do the work of a radiologist. Although there are thousands of such devices, many are doing the same task, looking for brain bleeds, for stroke, for lung nodules. There are only 42 unique use cases that are being covered by the existing AI products in the market for radiology."

Clearly, there's an enormous gap between what radiologists need and what narrow AI tools are offering today. Furthermore, there are many challenges to implementing AI tools into clinical workflow, such as:

• Addressing technical challenges in connecting these tools, which are often standalone solutions with their own interfaces and technical requirements, with clinical systems
• Assessing cost and contracting with many individual vendors that provide individual solutions
• Conducting internal validation to ensure that AI tools perform well in clinical practice and continue to do so over time
• Getting results into the clinical workflow
• Making sure there is a framework for cybersecurity risk assessments
• Understanding how many physicians will use an individual tool and how many patients it will be applicable to

AI Governance Committee

Dr. Bizzo and colleagues reviewed some of the above challenges in a recent article in Data Science, describing Mass General Brigham's AI Governance Committee, established to address challenges and implement solutions. The group is led by the radiology department with representation of other groups across the enterprise, including legal, ethics, and other clinical departments based on use cases.

"Any physician who wants to use an AI tool inside Mass General Brigham needs to go through that AI governance so that the tool can be vetted before it is used clinically. The group assesses requests, addresses challenges, and helps facilitate implementation," Dr. Bizzo says. "And if a tool is approved and implemented, the group continues to monitor its use. We perform independent validations, testing whether it lives up to expectations with local patients and local data."

Healthcare AI Challenge

Another strategy is the Healthcare AI Challenge, a new, multi-institutional, virtual, interactive series of crowdsourcing events designed to allow healthcare professionals to explore and assess the latest AI technologies in real-world healthcare scenarios.

The goal is to help clinicians determine the effectiveness of AI innovations, as well as assess their safety, effectiveness, and value. Healthcare professionals who participate will access simulations of AI tools for specific medical tasks, such as medical image interpretation. They can provide feedback and ratings on the tools' performance and utility. The public and industry can follow the initiative's progress online.

Radiology will be the first in the series, given the rapid proliferation of AI tools in the field. The initiative is partnering with the American College of Radiology as well as several other institutions across the country. (See also the "Keeping Pace With the Rise of AI" section in this article.)

Partnering With Industry to Develop AI Innovations

Dr. Bizzo also leads a group that partners with industry to develop and validate AI tools that can help improve patient care. The AI Clinical Research Organization was instrumental in his infarct study. The team works with companies around the world who want to introduce AI tools in the U.S. market.

"Our group generates clinically relevant evidence and medical expertise to inform the key steps — from concept to market adoption — before an AI tool reaches physicians and patients. We can really show the promise these tools hold in and the impact that can have in helping with increased demands for radiologists and other physicians, improving patient care."

Learn more about Mass General Brigham AI

Explore artificial intelligence research at Mass General