Enlitic, a San Francisco-based medical deep learning company, announced it raised $15 million in Series B funding to utilize AI for streamlining medical imaging workflows for radiologists. This investment round was led by Marubeni. And existing investor Capitol Health also participated along with participation from several top investors in Australia.
About 10% of this funding round was set aside for individual radiologists and other physicians who have either trained or used the company’s medical AI solutions. And Enlitic is planning to use the funding to enhance its AI product portfolio, expand its engineering and data science teams, and focus on regulatory approval for clinical use in the U.S., Australia, Japan, Europe, Canada, and Brazil.
“Radiologists have one of the hardest jobs in the world. They need to be able to identify thousands of different abnormalities in hundreds of different types of images. Even a single mistake can mean life or death, and yet they’re asked to read under tremendous time pressure in an environment full of distractions,” said Enlitic CEO Kevin Lyman in a statement. “Our Series B is a huge step toward helping them relieve this pressure and improve patient outcomes. We are incredibly grateful for the support of our partners, including Marubeni and Capitol Health, and look forward to continuing our work with them to build meaningful diagnostic solutions.”
Enlitic works closely with hospitals and radiology providers around the world to develop algorithms in order to identify and analyze suspicious findings in medical images. Plus Enlitic has developed a comprehensive platform enabling the development, validation, and seamless integration of clinical AI at scale.
The early applications of Enlitic’s technology were proven to speed up radiologist’s interpretation by over 20% while improving true positive rates and reducing false positive rates by more than 10%. And other applications were shown to assist in the identification of malignancy up to two years earlier.
“Closing this round is a significant advancement to help Enlitic achieve its ambitious strategy to make meaningful clinical products for radiology and the associated healthcare markets,” explained Enlitic chief medical officer and radiologist Dr. Anthony Upton. “This will allow Enlitic to scale production towards achieving AI models that cover the entire body, which will offer a significant impact to daily clinical workflows.”
The first product made by Enlitic interprets chest x-rays, triaging normal from abnormal scans, and detecting and characterizing over 40 distinct abnormalities like cardiomegaly, lung nodules, and pneumothorax. Due to the difficulty in interpreting the image’s lack of specificity, chest x-rays are notoriously difficult to interpret. However, they are still the most common type of radiological examination, representing an estimated 45% of all studies globally.
Enlitic is working with a number of partners around the world for the relevant approvals to deploy this product in several countries, helping radiologists interpret chest x-rays faster and with higher accuracy.
Earlier this year, Enlitic launched a new data center with a GPU supercluster enabling model development at more than 10x the speed. And Enlitic also launched its global beta program in February — which enabled their models to be regionally tuned and tested in several countries simultaneously. And at the current pace, Enlitic is expecting to be able to interpret 95% of all types of x-rays by the end of 2019 and 95% of common CTs and MRIs by 2021.
