A new technique picks up relevantly
clinical tumors not detected by
ultrasound in prostate cancer scans
Surgeons should rely on visual interpretations and software for tumor detection, says study
December 14, 2018
by John R. Fischer
, Senior Reporter
Researchers at University College London are recommending surgeons rely on both their visual interpretations and a software product used to identify clinically relevant prostate tumors not picked up by conventional ultrasound for biopsy guidance, based on the findings of a new study.
Deployed through a system called SmartTarget, the program enhances the technique of MR-targeted biopsies by relaying tumor information collected from the scans onto ultrasound images to identify malignancies and guide surgeons in removing them during biopsy procedures.
"The software provides them with a clear target," Dr. Dean Barratt, professor of medical physics and biomedical engineering at the UCL Centre for Medical Image Computing, and the inventor of the SmartTarget system, said in a statement. "As MR-targeted biopsies require a very high degree of expertise and experience, we hope that the imagery displayed by SmartTarget will help to bring high accuracy prostate cancer diagnosis to a much wider range of patients and hospitals."
The use of MR scans in detecting the location of tumors prior to biopsies has boosted detection rates from 50 to 90 percent in the last five years.
The SmartTarget system leverages the approach by utilizing advanced image processing and machine learning to create a 3D model of the prostate and cancer from the MR scan of a patient, and then fuses it with ultrasound images during a biopsy. The combination highlights the area of concern, which otherwise would not appear in the ultrasound images.
Researchers applied the technology to 129 patients with suspected prostate cancer. Each underwent a biopsy that utilized SmartTarget and another in which the surgeon visually reviewed the scan, with each detecting 93 clinically significant prostate cancers and identifying up to 80 of the malignancies. Each approach missed 13 that were picked up by the other.
Based on their findings, the developers of the system recommend using it in conjunction with visual interpretations, predicting both together reducing the number of biopsies required, as well as unnecessary surgeries caused by over-diagnosis of less harmful cancers.
"With this study we now have hard data showing that SmartTarget is as good as a group of experts in targeting tumours in the prostate, and have a glimpse of how clinicians and computers will be working together in the future for the good of the patient," said co-senior author, Professor Mark Emberton, dean of UCL Medical Sciences, in a statement.
The SmartTarget software has been commercialized by SmartTarget, a company spun off of UCL's commercialization company, UCL Business PLC, and is in use at hospitals throughout the U.S. and the U.K.
The study was funded by the U.K. Department of Health and Social Care and Wellcome Health Innovation Challenge Funding.
The findings were published in the journal, European Urology.