More peer reviews point to benefits of AI support solution for mammography
Eight independent peer reviews have highlighted the benefits of ScreenPoint Medical’s Transpara dedicated breast AI decision support solution for 2D and 3D mammography.
Transpara powered by FusionAI is now in use in more than 20 countries with over 35 clinical studies looking at its role in screening workload reduction, reduction in interval cancers, helping with COVID-19 diagnostic delays and personalised risk.
According to marketing manager David Lee: “ScreenPoint Medical recognises that it is the endorsement of radiologists and breast clinicians that is most critical to the success of any new advance in the field, so we are happy to report what users say about Transpara.”
An independent study published in Radiology shows that Transpara breast AI can reduce radiologist workload up to 70 per cent for both 2D and 3D mammography. Results from this retrospective study of 15,987 mammograms demonstrated the AI reduced screening workload up to 70 per cent for either digital mammography or digital breast tomosynthesis-based screening programmes without reducing sensitivity by five per cent or more, and that using AI to transition from digital mammography screening to digital breast tomosynthesis screening would yield a reduction of 30 per cent in workload, a 25 per cent improvement in sensitivity, and a reduction of 27 per cent in recall rate.
Studies have also taken place in the USA, the Netherlands, Sweden, Germany, France and Spain. Radiologist Dr Alvarez Benito at Hospital Universitario Reina Sofia in Cordoba, Spain, said: “From the studies we did with Transpara, the conclusion was that it could help safely reduce screening workload. This could really help to address the radiology workforce issues that are faced in many countries.”
Picture: Dr Alvarez Benito’s team at Hospital Universitario Reina Sofia, Cordoba: Dr Marina Alvarez, Dr Sara Romero, Dr Esperanza Elias and Dr Jose Luis Ray.
Published on page 4 of the June 2021 issue of RAD Magazine.