Platform delivers AI-powered workflow efficiency and accuracy to improve disease detection, diagnosis, follow-up and treatment
November 25, 2018 — At the Radiological Society of North America’s (RSNA) 104th Scientific Assembly and Annual Meeting, Nov. 25-30 in Chicago, Nuance Communications Inc. introduced PowerScribe One, the next generation of the company’s radiology reporting platform. Harnessing Nuance’s advanced natural language understanding capabilities and cloud technology, the PowerScribe One platform integrates access to artificial intelligence (AI)-powered diagnostic and decision-support tools to measurably improve workflow efficiency and accuracy.
Purpose-built to address inefficiencies, inconsistent recommendations and failed follow-up, PowerScribe One benefits from the latest advances in AI such as natural language understanding, image characterization, speech recognition, intelligent exchange of structured data and cloud technology. This enables radiologists to:
- Extract structured data continuously while dictating;
- Apply structured data to aid in creation of accurate, clinically appropriate and consistent diagnostic reports;
- Share clinically relevant structured data between systems — including electronic health records (EHRs), picture archiving and communication systems (PACS) and others — enabling care teams to improve follow-up, monitoring and outcomes;
- Populate reports dynamically with content from integrated systems to minimize redundancy and reduce errors;
- Integrate AI image characterization directly into the reporting workflow via Nuance’s AI Marketplace;
- Make standardized reporting initiatives easier to launch, administer, promote and adopt; and
- Harness the cloud to easily access and use evolving technology and innovation.
“PowerScribe One transforms my dictations on-the-fly into relevant, structured findings so that it’s not just my words, but their meaning in the context of my patient’s clinical situation. Then, by incorporating the data into the decision support workflow, I can become both more effective and more efficient,” said Tarik Alkasab, M.D., Ph.D., service chief for information technology (IT)/informatics and operations at Massachusetts General Hospital. “PowerScribe One is more than a voice-enabled text report creation tool, it’s becoming an integration platform that empowers radiologists to capture, share, and take action on important patient information more quickly and easily.”
PowerScribe One leverages a cloud-powered, continuously learning and context-aware language understanding platform to convert unstructured speech-to-text input into structured data. The platform also integrates access to AI algorithms within the radiologist’s normal workflow, giving radiologists powerful solutions to improve efficiency and throughput, increase diagnostic accuracy and specificity, and ensure consistent, evidence-based follow-up recommendations. The integration of structured narrative data, workflow-embedded AI, clinical guidance, assisted diagnosis, quality-checking and report automation tools elevates the accuracy, completeness and value of radiology reporting.
PowerScribe One also enables enhanced sharing of concrete and actionable information with care teams and across the care pathway. It puts essential follow-up, timeframe and other information that was previously stranded within unstructured narrative report text directly into the hands of clinical users. Converting unstructured text to structured data opens a path to expanded interoperability and powers a new generation of data-driven applications. It also complements standardization work on Common Data Elements by the American College of Radiology (ACR), American Society of Neuroradiology (ASNR), RSNA, and other governing bodies to improve information exchange and advance radiologic practice, research and performance measurement.
“Imaging AI cannot be deployed in a silo – integrating the outputs of AI into the radiologist’s existing clinical workflow enables it to effectively augment the capabilities of the radiologist, while preserving as well as enhancing efficiency and quality,” said Warren B. Gefter, MD, professor of radiology at the Hospital of the University of Pennsylvania. “AI in radiology is rapidly evolving; the challenge is to make it usable and useful.”