Share this post:
Hospitals generate enormous amounts of medical information.
As much as 85 percent of it is unstructured, which means it can’t be analyzed or cross-referenced with other kinds of data, making it nearly impossible to use clinical data when conducting business analyses, which wastes enormous opportunities in terms of potential cost savings and efficiency gains. For example, performing comparative effectiveness research, which compares how cost-efficient and effective different treatments are for a given pathology using tens of thousands of cases as a sample, is nowadays hugely time consuming, while with complete, structured and interoperable data, it would take a couple of clicks.
McKinsey Global Institute asserted back in 2011 that big data is the next frontier for innovation, competition and productivity. Many economic sectors face enormous opportunities related to this new technological feature, but it’s healthcare that has the most at stake. In a context of financial pressures over healthcare systems worldwide, mainly due to population ageing and decrease of active labor force, generating value from data is more important than ever. For example, if US healthcare could make a comprehensive, massive use of clinical data, the sector could create more than $300 billion in value every year.
Thus, circumstances were serendipitous for IOMED Medical Solutions S.L. to come on the scene with its first product, Medical Language API. The solution uses natural language processing to process clinical unstructured text inputs and extract structured and codified data without requiring doctors to change the way they work. This data, moreover, is standardized using internationally recognized standards, such as ICD10, SNOMED and LOINC.
Medical Language API runs on IBM Cloud Virtual Servers and works with any electronic health record (EHR) software or hospital information system (HIS).
An idea takes shape
Starting a company was one of my life’s goals. I was completely convinced that at some point in my life I was going to start a company.
About three years ago, I was sharing my flat with a doctor, Gabriel, who later turned out to be one of my business partners. He came in one night after work, totally frustrated with the hospital information systems, explaining the huge potential that was lost due to the way information was managed in the hospital. We decided something had to be done to make that better.
We therefore started researching: I started with the business end and he started with the medical and technology part.
Very soon, we joined a biotechnologist, Álvaro, at a hackathon event, who became our third partner, and from that point, the project just kept growing. When it got pretty serious, we decided to leave the jobs we had at that time and get involved full time in the company.
Getting together with IBM
The team knew it needed the right partner to get Medical Language API launched. After hearing about several opportunities in the field of corporate partnership, it crossed paths with IBM first time at the Health 2.0 event in Barcelona. Once the team explained the project, IBM scheduled a conference call to explain how it works with e-health companies and the IBM Global Entrepreneur program.
Aside from using the IBM Cloud infrastructure, IBM has put IOMED in contact with key decision makers within relevant players in the market, both in terms of potential technology partners as well as advisors.
IOMED has found that people pay more attention and are more open to exploring potential collaborations when they see it is backed by a company like IBM. The company is still a startup, so it needs this kind of support to grow.
From cost savings to operational efficiencies
According to Forbes, beyond improving profits and cutting down on wasted overhead, big data in healthcare can be used to predict epidemics, cure disease, improve quality of life and avoid preventable deaths.
Medical Language API is the key to that potential. It generates the necessary data with the necessary structure and codification for this value to be unleashed. Combining clinical data with economic data could offer strong insights on where hospitals are using their resources: What is more efficient? What is not efficient at all? Where can things be more efficient?
Imagine a big hospital in a city with 1,000 patients a month or even more. This hospital might lose more than $12 million a year in readmissions. These readmissions could be completely avoidable if the hospital could analyze patterns of and predict readmissions, which would actually save those $12 million dollars per year. Additionally, reducing the readmission rate is a key performance indicator (KPI) that hospitals care about very much.
Another application is in scientific research. If doctors had a huge database with real knowledge and structured data about millions of patients, it could speed up research immensely, both in terms of clinical research as in terms of new product development.
Medical language API is currently being implemented in a large hospital in Barcelona, as well as piloting with several others in Spain and Colombia. These are the first steps in fulfilling the company’s mission of reinforcing medicine with evidence and data, thereby identifying opportunities to achieve cost savings and drive operating efficiencies.
Try Medical Language API here.