Read the First Edition of The Lens, a new magazine by LSI arrow-icon

Kirill Lopatin, xAID - AI-Powered Diagnosis of CT and MRI Scans | LSI Europe '24

xAID is an AI-powered tool that helps analyze scans, write reports, define patient priority, notify doctors, and manage follow-ups.
Speakers
Kirill Lopatin
Kirill Lopatin
Founder & CTO, xAID

Kirill Lopatin 00:02
Thank you. Happy to be here. So basically, we are developing an ultimate AI pilot for CT of chest, abdomen, and pelvic processing. So to start with, our company is about to tackle a huge, almost 40 billion euro market. Our founding team has successful previous experience in building profitable AI solutions in the healthcare industry, and now we are seeking $3 million for our seed round to expand in Europe, with 300k already committed. For our clients, we have already processed and analyzed more than 1 million CT and MRI scans. We developed more than 30 different solutions to detect different pathologies in body parts, starting from the head and ending with the pelvis. We have a nationwide contract in Kazakhstan, which is to be signed by the end of the year and is priced at $1.5 million per year. We processed, collected, and annotated a unique dataset of more than 100,000 different CT and MRI scans. Our CEO has successful previous experience in building profitable AI solutions in the Spanish market and made a profitable exit with it. I, as the CTO, was involved in a company developing an offline computer vision analytics product, which was successfully deployed with big clients such as McDonald's and PETRONAS. Our CPO is the head of AI at the biggest marketing platform for e-commerce in the CIS region. Our CBO is a professional with more than 20 years of experience in the healthcare industry. He has worked for such companies as Johnson & Johnson and Alcon. Our Medical Advisory Board is evenly spaced between the EU and the US, and one of our medical advisors is an ex-CIO of the AI healthcare marketplace Azimus, and the second one is an ex-president of the Northern California Society of Nuclear Medicine. There are a lot of problems that exist in radiology and emergency markets today. Nine out of 10 emergency departments face shortages of radiologists, especially at night. About 40% of CT scans of emergency type contain at least one missed supplementary pathology that is overlooked because it's not of an emergency type, and it reveals that because of that, missed pathology clinics lose more than 150 euros on each CT scan. So all these problems create a huge, almost 40 billion euro market, which our company is about to enter. Utilizing all distribution channels and covering all targeted geographies, we can obtain more than 1 billion from that market. So what we have already developed is a lot of different solutions covering the most examined body parts, such as head, chest, abdomen, and spine. We can detect more than 30 different pathologies in those body parts, and our solutions are not a black box. Here you can see how our chest CT, for example, works. We provide pathology detection and classification, different types of measurements, and segmentation. And what is also very important, we provide pre-populated medical reports for each scan. Thus, we can optimize radiologists' time. It reveals that there is an empty space in the AI healthcare emergency sector for chest, abdomen, and pelvic CT scan processing. Chest and abdomen have a combined share of 60% of the total flow of emergency CT scans. That's why we decided to focus on these solutions. Our vision is to make an emergency tool that will help radiologists process chest and abdomen CT scans faster and more accurately. The next step is to extend our tool to the ability of fully automated reporting of these scans, especially during the night shift when the radiologist is not available at all, and our vast range of pathologies that can be found already by our tools strongly aligns with the opportunistic screening guidelines for emergency examinations. That's why we can find supplementary pathologies that are usually not in the focus of radiologists because they are not of an emergency type. So up to 95% of CT scans of chest and abdomen, especially during night shifts, can be automated by our tool, and because we can find additional pathologies that can usually further be converted to additional sales inside the clinic, we can increase the revenue stream for the clinic by up to $150 per CT scan. And of course, we can bring time reduction for radiologists in a copilot mode during the full description of these CT scans. So there is already competition in the space, but none of them provide a complete and comprehensive solution for the emergency domain for chest and abdomen. There are prominent companies such as RapidAI, Visa AI, and AIDoc, which are primarily focused on head CT analysis and stroke detection. They barely have solutions for chest and abdomen and can find maybe separately one or two different pathologies for these body parts in the emergency sector. On the other hand, there are companies such as Infravision, Coraline, or Q.A., which are primarily focused on pulmonary analysis and lung detection. What stands apart is that we're already able to find more than 10 different pathologies for emergency chest and abdomen CT scans, and this list will be further extended to the maximum to fully automate reporting of these body parts. So until now, we are focusing on getting our regulatory approval for Latin America and Kazakhstan. It allows us to generate revenue from the current moment, like we will do in Kazakhstan by the end of the year. We have already started the process of getting the CE mark approval and expect to get it in the middle of next year, and we will start our sales pipeline next year in Europe, and in 2026 we will be ready to tackle the US market after we get our FDA approval. We already have pilots in several different regions, such as Brazil, Kazakhstan, and the Czech Republic in the European Union. Next year, we will focus on the European Union and expand our presence in Latin America. And in 2026, we'll be ready to tackle the US market. So our business model is simple. We sell annual licenses with different numbers of studies included. Our players are hospitals and imaging centers, and our end users are radiologists. Now we are seeking 2.8 million euros, with 275k committed for our seed round. We will use these funds to develop the next stages of our product, get our CE mark approval in Europe, and tackle markets in new regions. So to sum up, we are ready to tackle a very big emergency market, and we will be the first company to provide a comprehensive solution for processing emergency chest, abdomen, and pelvic CT scans. Thank you for your attention.

LSI USA ‘25 is filling fast. Secure your spot today to join Medtech and Healthtech leaders.

March 17-21, 2025 Waldorf Astoria, Monarch Beach | Dana Point, CA Register arrow