Skip to main content

“contextflow” showcases agility during global pandemic

MedUni Vienna startup develops software for radiologists powered by AI and is combating COVID-19 with new features and partnerships
All News
Bild: contextflow
Co-Founders (from left): Chief Scientist Georg Langs, Professor of Data Intelligence Allan Hanbury, CEO Markus Holzer, and CTO Rene Donner

(Vienna, 06 July 2020) Fresh off its latest investment round, “contextflow”, a spin-off of the Medical University of Vienna, Technical University of Vienna and European research project Khresmoi, has developed new features within weeks designed to assist radiologists fighting COVID-19.

CEO & Co-Founder Markus Holzer explains, “Because we take a very general approach to supporting disease detection, we were able to quickly adapt our software, working closely together with our partner hospitals to outline what information and features radiologists find useful for COVID-19 patients.”

The company was able to take radiologists’ feedback and quickly design new features (currently available for research use only) including: 1) triaging; 2) enhanced search for COVID-19 disease patterns; and 3) automated report generation for suspected COVID-19 patients.

contextflow TRIAGE is a prioritization tool that looks for disease patterns in 3D medical images as soon as a scan is performed. Under normal circumstances, it quickly and clearly draws radiologists’ attention to time-sensitive patients, highlighting healthy versus unhealthy regions. Now, a reconfigured version scans for COVID-19-related patterns.

Additionally, the company’s software suite already included a 3D image-based search engine (SEARCH) that could search disease patterns in lung CTs, including those present in COVID-19 patients. Similar to TRIAGE, a new version searches for COVID-19-related patterns and provides distribution heatmaps of these patterns as well as the percentage of the total lung volume that is affected.

According to Senior Consulting Radiologist from Vienna General Hospital (AKH) Helmut Prosch,” Lung CTs are a vital part of the COVID-19 diagnostic process in patients with moderate to severe symptoms. Understanding where and how a patient’s lungs are affected can help determine complications of the diseases and the potential need for mechanical ventilation or ICU admission. ”

After a patient scan is evaluated, a radiologist must write a report detailing any findings. contextflow thus developed an automatic PDF report that could be attached to the complete report based on the findings of its SEARCH software, potentially saving radiologists valuable time.

Because of “contextflow’s” systematic approach, the company has secured new partnerships for testing the COVID-19 features with hospitals in Italy, one of the countries hardest hit by the pandemic. According to Rome’s Sant'Andrea University Hospital Chairman of Radiology Andrea Laghi, “The last few months have been challenging beyond belief, and we were searching for tools that could potentially provide value to our radiologists in terms of COVID-19 support but also in the future. contextflow is an example of what can be achieved in a short amount of time when hospitals and innovative companies work together.”

About “contextflow”
“contextflow” is a spin-off of the Medical University of Vienna, Technical University of Vienna and European research project Khresmoi. The company was founded in 2016 by a team of AI and medical imaging experts. Most recently, contextflow was awarded a €1.2 million grant from the European Commission for further developing COVID-19 capabilities. contextflow.com