The AI tool automated tasks such as saving and measuring the images, freeing the sonographer to spend more time interpreting the scans
This study demonstrates the feasibility of translating artificial intelligence from the biomedical research domain to a real world clinical application that could have significant benefits for patients and staff in antenatal screening and diagnostic services.” Jacqueline Matthew, NIHR clinical doctoral research fellow, King’s College London
Using artificial intelligence (AI) techniques for routine elements of an ultrasound scan can save up to seven-and-a-half minutes per scan, research at Guy’s and St Thomas’ NHS Foundation Trust and King’s College London has found.
The sonographers who took part in the study also felt that the use of AI techniques could free them to focus more on interpreting the scans.
The study’s participants were 23 pregnant women having their mid-pregnancy scan, which screens for anomalies and check the baby’s development. Each patient received a standard scan and an AI-assisted scan, performed by separate sonographers. A third sonographer then compared the images to assess the quality of each scan.
No significant differences were detected in the fetal measurements recorded by the two techniques. In terms of the quality of the saved image, the AI-assisted scan achieved a satisfactory image in 93% of cases, compared to 98% for the conventional scan.
Jacqueline Matthew, NIHR clinical doctoral research fellow at King’s College London and the study’s lead author, said: “This study demonstrates the feasibility of translating artificial intelligence from the biomedical research domain to a real world clinical application that could have significant benefits for patients and staff in antenatal screening and diagnostic services.”
The figures demonstrating the seven-and-a-half minute time saving were based on busy sonographers performing up to 12 scans a day. The types of jobs that could be carried out by the AI tool included saving and measuring the images, creating automatic reports and other manually repetitive tasks.
The sonographers reported that significant amounts of time were saved by having to switch tasks less often during the scans. The results also demonstrate the potential for using AI to reduce human error and variations when taking measurements.
Professor John Simpson, professor of paediatric and fetal cardiology at Evelina London, said that using AI to help sonographers performing fetal anomaly scans had “huge potential to improve workflow and quality.” He added: “This study is an important step, which we are confident will be followed by further advances, such as automated recognition of anomalies to further assist the sonographer.
“The intelligent fetal imaging and diagnosis (iFIND) project team are continuing the development of the AI tools evaluated in this paper. In addition to recognising standard image views and foetal measurements, work is also ongoing to develop tools that can recognise specific anomalies before birth including serious heart conditions. The aim is to further support ultrasound specialists and improve the accuracy of these important examinations.”