Speech recognition helps King’s meet cancer target

  • 28 September 2005

Speech recognition technology has helped produce a 50% reduction in diagnostic reporting times according to a report of a recent cancer treatment trial at King’s College Hospital, London.

The trial, conducted by the South East London Cancer Network, was designed to find a solution that would reduce report creation times to meet NHS waiting time targets for cancer diagnosis and treatment.

The software used was Dragon NaturallySpeaking which converts speech into text at up to 160 words per minute, according to supplier, ScanSoft, which worked on the project with UK Dragon integration partners, Hands Free Computing.

Using speech recognition helped to automate the clinical documentation process and eliminate the high cost and long turnaround time associated with the manual transcription of patient notes. The pilot at King’s College Hospital audited the difference between conventional reporting using dictation to tape or written notes that are then typed compared to using speech recognition for the same task.

During the trial, the histopathology department reduced reporting times from six days to three.

Simon Howard, ScanSoft’s professional and healthcare solutions manager, EMEA, for Dragon NaturallySpeaking, said: “With sensitive patient data, the medical environment – where accuracy is absolutely critical – is the ideal proving ground for speech recognition technology. This project shows that it is being used successfully in such a demanding environment, and is a reflection on how far the technology has progressed, and how accurate it now is.”

The South East London Cancer Network has secured funding to complete the roll-out of the software in the histopathology and radiology department and to extend the project to another five trusts. The network plans to purchase in excess of 100 additional licences for radiology and histopathology services across South-east London.

The NHS cancer waiting times target is to treat people diagnosed with cancer within 62 days of a GP referral and within 31 days for urgent cases. The target requires the diagnostic phase of the patient journey be as quick as possible.

Subscribe To Our Newsletters

Subscribe to our newsletter

Subscribe To Our Newsletter

Related News

Penny Kechagioglou: ‘We’ve become too attached to AI’

Penny Kechagioglou: ‘We’ve become too attached to AI’

Ahead of Digital Health Summer Schools, Dr Penny Kechagioglou discusses clinical leadership and health inequalities.
Google AI outperforms human doctors in detecting breast cancer

Google AI outperforms human doctors in detecting breast cancer

An NHS study on the use of AI in breast cancer screening has found that Google AI was better at detecting cases of breast cancer…
AI tool predicts repeat heart attack risk for cancer patients

AI tool predicts repeat heart attack risk for cancer patients

Researchers have developed an AI tool that predicts the risk of further heart attacks in people with cancer.