Skip to content

The power of artificial intelligence for MRI transformation

By Mirjam Blechner, Senior Industry Solution Executive Health, Driving Digital Transformation in Health with State of the Art Technology on April 18, 2017

Filed under Health

Two doctors looking at a scan of a brain on a computer

Addressing the Healthcare Challenge

The current healthcare industry must deal with extended life expectations, exponentially rising costs, and overworked staff. Furthermore, investment cycles for new, improved medical equipment are lengthy, often combined with public tenders. Also, remote geographical regions often suffer from a shortage of experienced radiologists, even when an MRI system itself can be afforded.

Balzano Informatik GmbH worked with experts from ETH Zürich, one of the world’s leading science universities, the teaching University Hospital Balgrist Zürich, and Microsoft to address those challenges using modern Artificial Intelligence and Cloud Services from Microsoft. After months of system development and deep learning, the “ScanDiags” service is now being tested in first clinical environments in Switzerland.

The “ScanDiags” Service Benefits

“ScanDiags” is an Artificial Intelligence-based MRI analytics solution which automates MRI interpretation and documentation. It reduces the evaluation time of MRI scans up to 20% per patient case and delivers an immediate, automated second opinion to the radiology expert. It also helps standardizing diagnosis and documentation methods with the highest quality to provide patients the best possible treatment. Thus, ScanDiags helps saving time and costs, while increasing quality. It also comes as a Web Service, delivering immediate return to clinics without any upfront investments and lengthy purchase cycles.

Building on standard cloud services, ScanDiags marks a quantum leap in time required to take a clinical AI solution to market, as it does not only collect and classify thousands of cases, but it also uses artificial intelligence, neural networks, and machine learning to speed up and continuously improves its performance and accuracy. Finally, it also significantly reduces time required for know-how sharing among radiologists, as it reduces the need for continual meetings.

Optimize Your Clinical and Operational Effectiveness and Transform the Care Continuum

When asked about his vision for ScanDiags, René Balzano, CEO and founder of Balzano Informatik GmbH, underlines the initial reduction in time per scan, the immediate availability of an automatically generated report containing the key medical findings as second opinion, and the cost savings achieved as some of the first drivers for the solution design and the overwhelmingly positive customer feedback received so far. Today, top of mind for every clinic is aggregating and analyzing robust patient data, and using predictive analytics to improve patient outcomes. ScanDiags addresses these imperatives nicely, at an unprecedented cost.

ScanDiags is designed as a global service. René Balzano says “ScanDiags has the ability to gather expertise from the best experts worldwide, continuously learn from them, improve outcomes, and share those with the world. It also gives the local healthcare industry, with its worldwide reputation for quality, a chance to “export” this know-how globally.”

The solution outline

ScanDiags runs as an Azure-based WebApp in a PaaS architecture that is managed by Microsoft. This provides inherent, fully-automated high-availability and scalability. Thus, no maintenance requirements are imposed, neither on customers nor on Balzano when the solution is provided to them. The ScanDiags WebApp queries and evaluates pre-trained cognitive toolkit models via a Node.js and Python implementation. Statistics are stored in an Azure SQL Database and presented through PowerBI dashboards.

The Microsoft Cloud Services used allow the solution to differentiate itself from other AI solutions in the market. René Balzano: “Differently than large AI systems like IBM Watson, the Azure Platform and the Microsoft Cognitive Toolkit allow for minimal time-to-market, and short implementation cycles at very low implementation and runtime costs… truly democratizing AI.”

Customer Statement (teaching University Hospital Balgrist Zürich – Prof. Dr. Christian Pfirrmann, Head of Radiology)

“A continuously learning, AI-based solution like “ScanDiags” exponentially improves the important knowledge-sharing process among radiologists, which is crucial for our patients, as it increases the diagnostic quality. At the same time, the optimization of this process substantially reduces the overall costs of healthcare.”

Balzano Informatik is one of our partners in the AI in Health Alliance, which aims to democratize healthcare for the greater good. You can learn more about this initiative here.

Useful Links

Contact Us