How data can be a tool that saves lives: The story of Datlowe
There is a lot of data circulating in healthcare, but it is rarely used to its full potential. Imagine, however, a situation where that data could directly help patients. Radim Janda and Jakub Kozák from Datlowe focused on the problem of hospital infections and decided to use large amounts of unstructured data to create a tool that can literally save lives.
Disclaimer: This article was translated from the Czech language by AI.
In the podcast Delfíni na vlnách (Dolphins on the Waves), they describe how HAIDI, their hospital data analysis tool, has become key to preventing healthcare-associated infections. Why? Because infections that patients contract directly in hospital are still one of the main causes of complications and deaths.
How Datlowe came to be
Datlowe is not just another start-up that collects and evaluates data. It is a technology that has a real impact on patient health. From the outset, Jakub and Radim were convinced that healthcare had enormous potential to make better use of data.
Healthcare systems collect enormous amounts of data, but it is often in an unstructured form that is difficult to process. This was the challenge they decided to tackle. And the result? HAIDI, a tool that connects directly to hospital systems and reads both structured data and text in medical records and notes written by nurses.
Healthcare-associated infections: How big of a problem are they?
Infections that patients contract directly in hospitals are now one of the most common reasons for prolonged hospitalisation and treatment complications in developed countries. And although hospitals are trying to address this problem, new cases continue to emerge. More than 4 million patients are infected each year in the EU. Statistics show that approximately 7% of hospitalised patients contract an infection that they may have caught in the hospital. And that is precisely the challenge that Datlowe is responding to. The HAIDI tool can identify potential infections and their risk factors in a timely manner. How? By analysing all available data about the patient, not only structured data, but also text from clinical reports and notes from doctors and nurses.
How data analytics helps prevent hospital infections
The cornerstone of HAIDI is the processing of large amounts of data – both structured and unstructured. This is done with the help of machine learning and natural language processing (NLP). This technology is capable of extracting key information from free text that might otherwise be easily overlooked. Whether it’s a catheter, surgical procedures or symptoms, HAIDI can find patterns that indicate potential infections.
This means that instead of doctors and nurses having to go through dozens of pages of records looking for important information, HAIDI automatically provides them with relevant alerts about potential problems.
How HAIDI works: Implementation process
The implementation of HAIDI in hospitals begins with its integration into the hospital information system. The tool then automatically evaluates new data and continuously provides the professional team with an overview of risks. For hospitals, this means reduced administration and effective infection prevention. Datlowe is already working in dozens of Czech, Slovak and Austrian hospitals, both in acute and follow-up care. The benefits are clear: reduced costs for treating infections and greater safety for patients.
What does the analysis of unstructured health data involve?
Unstructured data refers to medical reports that are often difficult for the average person to read. But for a machine model, it is a gold mine. HAIDI can extract critical information from these notes, such as the presence of infectious risks, the course of operations, symptoms, and more. Its own natural language processing module ensures that the machine can recognise different formulations (e.g. ‘the patient has a fever’ vs. ‘the patient is febrile’).
Why does healthcare still only use a fraction of its data potential?
Healthcare still has a lot of room for improvement. Once we start using data to its full potential, we will reach a new level of prevention. Healthcare, whether at the hospital level or across the entire healthcare infrastructure, uses only a fraction of the available data potential. The reasons for this are technical, legal and organisational barriers. And this is where Datlowe can play a key role. It is a way to remove obstacles and use data where it has a real impact.
What is the future of data in healthcare?
Healthcare is slowly beginning to open up to new technologies that help analyse data. But as Jakub Kozák points out, the future lies in integrating these data tools directly into the everyday practice of doctors. More than ever, data analytics is as important to doctors as a scalpel. This is precisely Datlowe’s vision – not to replace healthcare professionals, but to provide effective assistance to doctors, saving them time and helping them focus on what matters most: their patients.
Conclusion
Datlowe transforms hospital data into a tool that not only makes work easier for healthcare professionals, but also helps patients. Data analysis, which has enormous potential, can save lives and significantly reduce the number of hospital infections. And although we are still in the early stages, the outlook is clear: data in healthcare will play an increasingly important role, not only in prevention, but also in diagnosis and treatment.
This approach shows that when data is used correctly, it can bring real and measurable benefits to patients, hospitals and the entire healthcare system.
