We have built an AI-powered assistant for HAI surveillance.

Empowering hospitals to optimize care through
advanced data analytics

main picture

Using natural language processing and advanced data analytics, we unleash the power of hospitals' clinical data and change the way hospitals use information hidden inside electronic health records.

Datlowe develops solutions based on advanced data analytics which help hospitals to solve their problems. Our solutions can automatically read all medical records from hospital information system, extract relevant pieces of information for the specific problem and provide support for decision making. For example, our solution HAIDi can find all potential healthcare-associated infections from electronic medical records and help to manage them.

Our vision

We believe in the power of information hidden in mostly unstructured electronic health records in hospitals. Helping hospitals to optimize the care and deliver it efficiently is our motivation. We are sure everybody may benefit from a data-driven approach to healthcare.


We use natural language processing (NLP) to automate reading of unstructured texts and information extraction. We combine extracted information with structured data and utilize advanced methods of machine learning (ML) to provide valuable insights. We create expert systems as well as advanced predictive models.

NLP and IE

We have extensive experience with natural language processing (NLP), which allows the computers to read and understand free text automatically. Information extraction (IE) builds upon that and provides information from unstructured texts in a structured format. Results are further used in predictive models.


Mastering machine learning (ML), which gives machines the ability to learn from data without being explicitly programmed, is essential for building solutions with predictive functions. Gradient boosted machines, random forests, and neural networks are our daily bread.

Domain knowledge

You cannot work in a business unless you know it. We use standardized terminologies and models such as SNOMED CT and FHIR.

Show more



We provide services and solutions based on advanced data analytics which allow to optimize care in hospitals. We developed HAIDi, an AI-powered assistant for surveillance of healthcare-associated infections (HAI), which changes the approach to HAI surveillance. HAIDi can automatically detect HAIs from electronic health records and help hospitals to concentrate on prevention based on carefully collected data.
Our team
Data science, natural language processing, machine learning, software development. Our team is built on experts, trust and friendly atmosphere. We believe in what we do.
Co-founder of Datlowe and data analytics enthusiast. Jakub manages the company and designs smart solutions for hospitals.

Jakub Kozák

CEO, co-founder
Software developer and researcher with a PhD. in semantic technologies and information extraction. Jan mostly develops smart algorithms.

Jan Dědek

Chief Technology Officer
Experienced data analyst with the background in business intelligence. Radim manages all data models and data transformations in our solutions.

Radim Janda

Chief of Data Management
Software developer with a broad range of expertise. Besides that, Ivana is proficient with both NLP and machine learning.

Ivana Sixtová

SW Developer, Data Scientist
Machine learning is the core expertise of Anastasia. She develops our models and builds upon her experience in bioinformatics.

Anastasia Lebedeva

Data Scientist
Data scientist by heart and soul with the background in statistics. Lenka supervises development of all predictive models and information extraction.

Lenka Vraná

Chief Data Scientist

Hospital Jihlava

The first implementation of HAIDi which runs on hospital-wide data and helps with HAI surveillance. We closely cooperate with Hospital Jihlava in the Czech Republic.

Hospital Kyjov

HAIDi pilot project for identification of HAIs with a validation phase.

Havlíčkův Brod

HAIDi pilot project for identification of HAIs with a validation phase.
latest blog posts

6 Reasons Why Keywords…

Listening to the following words may become a little tiresome: "I've heard you have a solution which can read unstructured texts. Can you tell me which keywords do you use?" It was a reason why we…
latest blog posts

Highlights from recent…

Results of an European survey on healthcare-associated infections (HAIs) has been published recently. In 2016 and 2017, the European Centre for Disease Prevention and Control (ECDC) conducted two…
latest blog posts

Launching new website

We have prepared a fresh new website! It is simple, reflects our current activities in healthcare analytics, and introduces our product HAIDi, which brings NLP and machine learning into surveillance…
Let’s talk

Do you want to meet us, talk to us or see HAIDi?
Send us a request!