iSoft’s RadCentre radiology information system will now offer real-time speech recognition, as part of its ongoing integration with Philips SpeechMagic technology.

The upgrade means that dictations will appear instantly as text on screen, speeding up the time to produce critical radiology reports and share patient diagnoses. SpeechMagic is currently an optional extra for RadCentre users.

An iSoft spokesperson told E-Health Insider: “Many customers are already benefiting from Philips’ speech recognition technology. Real-time speech recognition promises even greater time savings, which is especially vital in emergencies, plus greater speed, accuracy and security.”

The real-time speech recognition has been piloted at two early adopter sites in the UK – one in West Middlesex and the other in Yorkshire. iSoft says these sites are already witnessing improved workflows, a faster turnaround of reports, and reduced waiting times for patients and referring doctors.

The system is said to not only recognise words but also understand context, enabling it to ignore dialogue that is not part of the dictation, implement dictated corrections, and rephrase sentences.

The software can also be used to fill information into forms, and formats and organise text, for example adding section headings, numbering, and standard content to produce usable reports.

Gary Williams, partner programme manager for Philips Speech Recognition Systems, said: “Together with iSoft we can now further improve patients’ healthcare experience and potentially save lives by speeding-up documentation flows and ensuring the availability of patient information when needed, where needed.”

Paul Richards, managing director of iSoft UK and Ireland, added: “Real-time speech recognition promises even greater time savings, which is especially vital in emergencies.

“The speed, accuracy and security of radiology reports are improved markedly, with the clinical radiologist able to create and verify the report as a single, continuous process. Being able to dictate reports and see them instantly on screen, saves radiologists having to check, correct, and possibly re-send reports for typing.”

The RadCentre real-time speech recognition option is network-based, highly scalable, and can be administered and managed centrally. It is used by 56 NHS sites including eight in BT’s London cluster under NPfIT .





Joe Fernandez