This interview is part of our new AI in Healthcare series, where we interview the world's top thought leaders on the front lines of the intersections between AI and healthcare.
In this interview, we speak with Kevin Harris, CEO and Director of CureMetrix, to understand how his company is using AI to transform healthcare, and what the future of the industry holds.
1. What’s the story behind your company? Why and how did you begin?
KH: Imagine that your wife, sister or daughter went in for her regular mammogram and the healthcare team missed that she had cancer. Early and accurate detection in breast cancer is key to improving cancer survival rates worldwide through the development of next-generation medical image analysis solutions. With 40 million mammograms a year in the U.S., only five cancers per 1,000 are being found at screening. That means that a significant number of cancers are missed.
CureMetrix was established in 2014 and created to address a mission-critical need in the industry — to help radiologists better detect breast cancer with the objective, data-driven answers they need to quickly support patients and their healthcare teams.
2. Please describe your use case and CureMetrix uses AI:
KH: This year CureMetrix received FDA clearance for its product cmTriage™. Here is how CureMetrix uses AI to solve a problem in reading mammography, which is the most complex in radiology.
The problem is: radiologists don’t have critical data they need:
- Today, radiologists are not able to identify the suspicious cases in their worklist. A radiologist typically reviews cases first in, first out. With hundreds of studies and images to review, the old way does not help the radiologist address the most suspicious cases first.
The solution is: cmTriage™
- The AI and algorithms in cmTriage help the radiologist prioritize the suspicious cases first and optimize overall workflow.
- Here is how it works: The patient comes into the radiology center and the clinical team takes her X-rays; the images are loaded into the radiologists PACS system. These images are organized, anonymized, de-identified and the data is encrypted, then sent to the CureMetrix cloud where the AI is applied. There, our algorithm analyzes the images and then sends results back to the radiologists PACS in a structure report format so they can view the prioritized worklist on their monitors. The whole process takes less than five minutes. And as part of our FDA clearance, cmTriage is cyber secure, HIPPA and DICOM compliant so no personal health information (PHI) leaves the facility.
- Additionally, cmTriage is different in the market:
- cmTriage™ Works across all breast densities with an extremely high area under the curve (or AUC) of .95 – this is groundbreaking because this reduces the number of false positives. This is very important for some ethnic groups; for example, in many Asian populations, women have smaller, higher density breasts so it is more difficult to detect anomalies. This is why the ability of cmTriage to work across breast densities is important to all types of patients.
- Works with both mass and calcifications and across lesion sizes – enabling potentially fewer false positives. Fewer false positives has the potential to
- Reduce the number of recalls
- Reduce the cost of breast cancer screening
- Improves flow and increases efficiency for the radiology practice
With Artificial Intelligence (AI), cmTriage
- Helps the radiologist to:
- Focus on the most suspicious cases first to improve clinical outcomes
- Improve clinical and practice efficiency – workflow
- Increases assurance for all types of patients that they are getting the highest standard of care available in mammography screening
- Helps the radiologist to:
3. Could you share a specific customer/user that benefits from what you offer? What has CureMetrix done for them?
KH: CureMetrix supports the radiologist, the patient, the employer, and the healthcare system by delivering AI-based technology that helps radiologists and healthcare systems:
- Catch cancers early
- Reduce false positives
- Reduce recall rates
- Reduce unnecessary biopsies for both mass and calcifications
- Increases efficiency to reduce healthcare costs
- Reduce treatment costs
- Increases patient confidence
We have collected over two million images to conduct our studies and we have collaborated with esteemed institutions including MD Anderson Cancer Center, Johns Hopkins University Medical Center, University of California, San Diego, and many others. Through this process, the radiologists, their patients, and their families are the beneficiaries of the CureMetrix AI.
A testament to our clinical efficacy and value in healthcare, CureMetrix and our collaborators have recently published several peer-reviewed studies in the AHA and in the Journal of Data Imaging that demonstrate the efficacy of our solutions in both improving cancer detection and reducing false-positives.
Recently Published Studies:
4. What other AI use cases in healthcare are you excited about?
KH: In addition to cmTriage, we are currently conducting studies for our product cmAssist™, which is a proprietary, investigational SaaS intended to identify, mark and score regions of interest on screening and diagnostic mammograms. So, in addition to prioritizing the radiologist worklist based on the most suspicious cases, we will be able to highlight for the radiologist where actual cancers are showing up in the images.
5. Where will CureMetrix be in 5 years?
KH: While the CureMetrix team could have focused on other types of images, the team courageously chose to work on mammography because it is the most complex in all of radiology. Solving this complexity puts us in position to expand our AI technology to other body regions. We are currently working with our esteemed clinical advisors across radiology specialties to help us understand where our technology can best significantly improve cancer survival rates worldwide. By working with the radiology community, in five years we aim to be the leader in leveraging artificial intelligence (AI) to support early detection across a variety of diseases, supporting early detection across the globe.