The idea behind Medical Search Technologies was born in 2010 when Jan Sevcik, CEO and founder, found himself facing a rare medical condition. Sevcik became increasingly frustrated at the lack of information available to him.
While many would settle for a Google search, Sevcik utilized his wealth of knowledge in software development and data analysis to find and eventually develop a better solution.
Sevcik already had a background in tech entrepreneurship. His first venture, focused on developing predictive pricing software, sold to Innerworkings in 2007. He then served as the Chief Information Officer at the public company, directing all IT functions encompassing software development, patent development, and infrastructure.
Sevcik’s technical and business acumen enabled him to take tools he developed while searching medical journals for specific treatment options for his own condition, and turn them into Medical Search Technologies. The software uses machine learning to process unstructured electronic health record text — things like doctor’s notes and complex medical literature — to provide alerts, enable follow-up tracking, display population health analytics, find relevant clinical research, and query complex medical parameters.
In other words, it can give patients a lot more information about the prognosis of their condition.
The technology can also be used to improve processes within provider and pharmaceutical operations, which Sevcik believes gives it the potential to make a major impact on the industry.
Founded in 2012, the Chattanooga, TN-based startup is still in its growth stage, employing 12. They recently graduated CO.LAB’s HealthTech Accelerator, backed by Erlanger Health Systems and Unum, which helped them deepen their connections to the healthcare industry.
Below, Sevcik shares more about their technology, market, and next steps.
Tell us about Medical Search Technologies
Medical Search Technologies allows real-time analysis of unstructured text in electronic medical records, which can be used to improve processes within provider and pharma operations. We are a SaaS model based on data volume and complexity of the request which determines the level of machine learning that is applied to solve the problem.
What problem are you solving?
Over 80 percent of all healthcare data is unstructured text, such as patient progress notes, radiology reports, and medical literature. Medical Search Technologies makes this data usable.
What has been your biggest obstacle thus far?
In the early years, our biggest challenge was amassing enough data to train our algorithms. Healthcare data is messy, so it requires a much different level of processing than other types of text. Another challenge is that healthcare as an industry is slow to adopt new technology.
Who are your ideal enterprise clients?
We are looking for innovative and forward-thinking healthcare provider organizations that are looking to solve both clinical and revenue cycle management challenges.
We are also seeking pharmaceutical partners looking to improve research and clinical trial processes. We have found that the most innovative companies freely discuss their challenges and are willing to collaborate to solve these challenges.
What are your next steps?
We are actively working on several transactions and continuing to discover new applications for our technology. Lastly, we are currently hiring Java engineers.
