UCLA and UC Davis will realize a recently funded, multi-institutional clinical trial to evaluate whether artificial intelligence (AI) can help radiologists in interpreting mammography more accurately, with the aim of improving their sorting and cancer.
The study, known as a prism test (realistic randomized artificial intelligence test for sorting mammography), is backed by the $ 16 million award by the Institute of Research patients focused on the patient (PCORI). The study will include hundreds of thousands of mammograms interpreted in academic medical centers and breast imaging facilities in California, Florida, Massachusetts, Washington and Wisconsin. The AI support tool to be studied will be transpara per screenpoint Medical with clinical flow flow of integration provided by the Aidoc Aios platform.
“This is the first randomized AI’s large -scale test test in the control of breast cancer in the United States,” said Dr. Joann G. Elmore, Director and Guide to Administrative Coordination and Professor of Medicine at the David School of Medicine Health. “We look carefully and objectively if AI helps or hinder – and for whom. Specialized radiologists remain in the driver’s position for all interpretations.”
A pressing question in the care of breast cancer
Breast cancer remains one of the main causes of cancer death among women in the US, while the usual mammography sorting reduces mortality through early detection, also has disadvantages – including false positives that can lead to unnecessary tests and costs.
“AI has a great promise, but it also raises real questions,” he said Elmore, who is also a researcher at UCLA Health Jonsson Contrekens Cancer Center and serves as director of the UCLA National Clinicer program program. “We want to know if AI helps radiologists find more cancers or just mean more tests that ultimately prove to be normal.”
Centered on the patient from the design
What makes Prism different is the emphasis on the patient’s focus. The test was developed in close cooperation with patient supporters, clinical doctors, health system leaders and policymakers.
Each participation will continue the usual examination as usual, without changes in the patient’s experience. The mammograms will be assigned by chance to be interpreted either by a radiologist on their own or with the help of an AI support tool that has been cleared with FDA. In all cases, an radiologist reads the examination and makes the final decision.
There has never been a test of this field of application examined by AI in the control of breast cancer in the US, the results will help inform not only clinical practice but also of insurance coverage, adoption of technology and patient communication. ”
Dr. Hannah Milch, the co-star researcher and the UCLA site PI and Assistant Professor of Radiology at UCLA
“There are many hopes that AI will take care better, but very few strict tests have really evaluated the results of the real world,” Elmore said. “This is our opportunity to create reliable elements, with the perspective and center of the patient’s perspective.”
In addition to analyzing cancer detection and recall rates, the study will include focusing groups and research to capture how patients and radiologists perceive and trust AI care.
Cooperative effort in six states
The Prism Test gathers seven top academic medical centers:
- Ucla .
- Uc davis (Data Coordination Center led by a double mainstream researcher Dr. Diana Miglioretti)
- Boston Medical Center (Main location researcher, Dr. Clare B. Poynton)
- UC San Diego Health (Main location researcher, Dr. Haydee Ojeda-Fournier)
- Miami University (Head of site researcher, Dr. Jose M. Net)
- University of Washington – Fred Hutchinson Cancer Center (Main site researcher, Dr. Janie M. Lee)
- University of Wisconsin -madison (Co -chair of researcher, Dr. Christoph I. Lee, Director of Researchers, Dr. Mai Elezaby and Dr. Ryan Woods)
Hitting the correct balance
Elmore stressed that the goal is not to replace human know -how, but to understand how AI can supplement. “Our experts Radiologists will continue to make the final call.
The trial is expected to inform future policy decisions, best practices in control and how they will more effectively integrate emerging technologies into patients’ care.
