Smart Reporting presents its new product generation ("SmartReports"). The embedded medical content is designed to increase efficiency of reporting while at the same time improving the quality of data acquisition. For example, a spoken free-text can be converted into text modules within the structured report in real time. The Munich-based company has designed the software in such a way that it can be intuitively integrated into the physician's previous reporting habits.
This makes medical documentation more complete and enables quick comparability. The result is medical documentation that is machine-readable and can be processed further in different digital contexts. As an infrastructure provider, Smart Reporting enables doctors in hospitals, clinics and practices to exchange and analyze data across departments - the basis for further digitization and the use of AI in the healthcare sector. The software is therefore eligible for funding in accordance with the guidelines of the Hospital Future Act (KHZG).
The structured reporting solution of Smart Reporting includes guideline-based decision trees that are available to physicians from radiology, pathology and other medical areas on the Smart Reporting platform. These are the so-called Reporting Templates. With SmartReports, physicians can maintain a reporting workflow that is familiar to them and can increase their efficiency even in routine reporting procedures. For example, they have the option to choose between a spoken free-text or their own previously created autotexts (canned text) when building their integrated guideline-based report. In addition, fully structured reporting can be performed using the predefined decision trees.
SmartReports understands the doctor
The new proprietary software generation consists of three components: The reporting infrastructure (the ability to speak or write free text), the stored medical content that provides the structure, and the analytics capabilities through machine-readable reports. The intelligent speech processing algorithms not only recognize the spoken words but can also translate them into structured text modules in real-time, based on the stored medical content - a kind of "real-time natural language processing".
"Our approach goes far beyond classic speech recognition, because we want the software to immediately understand what the doctor actually means", Thomas Huber, Head of Product Development. SmartReports recognizes the clinical context of the spoken text, so the basic structure of the report remains the same. Which is important, "because in the end - regardless of which path the physician chooses within the reporting process - the result should be a structured, comparable report," Huber continues.
Standard diagnostic procedures become more productive and efficient
According to the manufacturers, this new software generation immediately makes both complex niche reporting as well as routine reporting more efficient. One example is the improved communication between radiologists and referring physicians through the uniform and consistent structure of the report. The report format also enables machine learning and thus makes the data accessible for research and development – a benefit that the University Hospital of Cologne has also seen:
"The trend is generally moving toward personalized medicine, which requires extensive evaluations of the collected data," explains Prof. Dr. David Maintz, Director of the Institute for Diagnostic and Interventional Radiology at the University Hospital Cologne. This in turn requires structured information, which can be collected through structured reports. It might be possible to structure free-text after writing it, but "as studies show, this is often not the case, which is why we think structured reporting is the right way to go," Maintz says.
The text above was published at the 18.01.2021 on medical design news (https://www.medical-design.news) in german language by the author Melanie Ehrhardt. Medical Design News belongs to WEKA Fachmedien GmbH. Smart Reporting did the translation to English.