Interpreting head CT studies can be challenging for many reasons, including inconsistencies caused by suboptimal patient positioning. Radiologists are trained to read through these inconsistencies, but they do so at the potential expense of increased interpretation time and intra- and inter-observer variability. CT CoPilot® by HealthLytix leverages AI and machine learning to address some of these issues.
Mark Wittenberg, Vice President of Business Development at HealthLytix, shares his thoughts about the impetus behind the HealthLytix software and the potential impact their AI and machine learning-based solutions will have on the speed and accuracy of MR and CT imaging and reporting.
Jonathon Dreyer: Tell us about your business – when and how you started and your development journey.
Mark Wittenberg: HealthLytix is a leading healthcare technology company focused on helping radiologists improve their diagnostic accuracy, reading efficiency, and clinical confidence by leveraging machine learning and AI to develop advanced imaging software for both MRI and CT.
JD: What AI models do you have and what do they do?
MW: CT CoPilot is FDA-cleared post-processing software for head CT exams that improves radiologist reading efficiency, measurement accuracy, and clinical confidence. It addresses exam inconsistencies created by sub-optimal patient positioning while providing insights into key changes over time. CT CoPilot deploys an intelligent atlas-based registration algorithm to automatically generate consistently aligned multi-planar reformats on every study for more efficient viewing. In addition, the software increases conspicuity of change between exams by identifying the most recent prior study and generating a subtraction series that quantifies density changes over time. CT CoPilot also improves measurement of change over time by quantifying changes in lateral ventricular volume and midline shift between exams.
JD: What’s the big “Aha” moment when you first show users what your AI models(s) can do for them?
MW: We are solving a well-known problem in radiology using a novel approach with no change to radiology workflow and no manual intervention. Radiologists understand the value of consistently aligned studies and quantitative change metrics, but given their significant workloads, they need these features to be automated in their workflow, which is where CT CoPilot comes in. The value of CT CoPilot was confirmed with rapid acceptance and adoption at several top-tier, high-volume hospitals with Level 1 trauma centers. Since launch, our customers have reported being able to read head CT scans faster and with more confidence with improved diagnostic accuracy on difficult cases, such as detecting subtle density changes due to early infarcts.
JD: What challenges or needs did you see that drove you to focus on this?
MW: We created CT CoPilot in response to the increasing clinical demands being placed on radiologists. Generating correctly and consistently aligned head CT images and subtraction series allows for better, faster reads, especially of complex postoperative or trauma exams. Additionally, we’re seeing greater demands being placed on technologists. CT CoPilot eliminates the need for technologists to perform manual realignment, which allows them to focus instead on patient comfort to improve the patient experience.
JD: What’s the number one benefit you offer?
MW: CT CoPilot improves radiologist productivity without any change to their existing workflow.
JD: Are there any stories you can share about how your AI model(s) drove measurable patient care outcomes?
MW: According to Dr. Nikdokht Farid, a board-certified neuroradiologist at UC San Diego Health and early adopter of CT CoPilot, “CT CoPilot quickly and automatically provides correctly aligned images, which enables me to read head CT scans faster and more confidently. It has increased my productivity and has become indispensable in my interpretation of head CTs.” To learn more about her experience, read the story here.
JD: What benefits does Nuance and its AI Marketplace for Diagnostic Imaging bring to your users? What problems does the marketplace and integration into Nuance’s workflow solve?
MW: The Nuance AI Marketplace integrates our applications directly into the radiology workflow. It offers our solution as a single API to connect CT CoPilot to radiologists at more than 7,000 healthcare facilities and streamlines the hospital IT implementation process.
JD: What has your experience been working with the Nuance team?
MW: The Nuance team has been incredible across the board. From executive leadership to the product development teams to the commercial sales teams, Nuance’s support through the product life cycle of CT CoPilot has been impressive.
JD: What is your vision for how your solution(s) will evolve over the next five years?
MW: Our team is constantly working to enhance the current features of CT CoPilot, as well as collect market feedback from clinical experts to inform which new features we should add based on what will provide the most clinical value across imaging modalities. A natural next step is to incorporate additional predictive analytics and AI for triaging cases and extending the technology outside of the head.
JD: In one sentence, tell us what you think the future of medicine will look like.
MW: We foresee medicine moving toward a more integrative approach where imaging, artificial intelligence, and other clinical information, such as genetics, are combined in a single platform to help clinicians predict, detect, diagnose, and treat disease more accurately and efficiently.
To learn more about Nuance AI Marketplace for Diagnostic Imaging, please visit https://www.nuance.com/healthcare/diagnostics-solutions/ai-marketplace.html
Intelligence at Work is a blog series by Jonathon Dreyer, Vice President, Solutions Marketing, Nuance Communications. Intelligence at Work showcases projects and applications that demonstrate how Nuance technologies extend the value, use and performance of integration and development partner offerings. This blog series focuses on inspiring the healthcare developer community to think beyond their current state and take their innovations to new heights by tapping into the latest in artificial intelligence.