UCLA to conduct a large -scale randomized AI test during breast cancer screening

UCLA and UC DAVIS will conduct a newly financed, multinestitutional clinical trial in order to assess whether artificial intelligence (AI) can help radiologists interpret mammograms more precisely to improve breast cancer cancer screening and reduce unnecessary feedback and anxiety phrases in patients.

The study, known as a prismatic study (pragmatic randomized study of artificial intelligence in the field of mammography), is supported by a $ 16 million prize from the Institute of Research Research Focusing 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 studied AI support tool will be allocated by the Screepoint Medical with the integration of the clinical flow of work provided by the AIOS platform.

“This is the first randomized AI study on a large scale Screening tests of breast cancer In the United States – said Dr. Joann G. Elmore, double main researcher and head of the administrative coordinating site and a professor of medicine at David Geffen School of Medicine in Ucla and health policy and management at UCLA Fielding School of Public Health. “We look carefully and objectively at whether AI helps or hinders – and for whom. Expert radiologists remain in the driver’s place for all interpretations.”

Artistic question about breast cancer care

Breast cancer remains one of the main causes of cancer death among women in the US, while mammographic routine screening reduces mortality by early detection, it also has disadvantages – including false positive, which can lead to unnecessary testing, anxiety and costs and skipping cancer.

“AI has a great promise, but also raises real questions,” he said Elmore, who is also a researcher at Ucla Health Jonsson Comprehensive Cancer Center and acts as the director of the UCLA NATIONAL CLINICIAN SCHOLARS program. “We want to know if artificial intelligence helps radiologists find more cancer, or simply flags more exams that ultimately turn out to be normal.”

Focused on the patient by the project

What distinguishes the prism is emphasis on tests focused on the patient. The study was developed in close cooperation with patients supporters, clinicists, leaders of the healthcare system and decision -makers.

Each participating object will continue routine screening as usual, without changes in the patient’s experience. Mammograms will be randomly assigned to interpret by the radiologist on its own or with the help of the AI ​​support tool cleaned by the FDA. In all cases, the radiologist reads the exam and makes the final decision.

A test of this range has never been carried out, in which he looks at artificial intelligence during screening of breast cancer in the USA. The results will help inform not only clinical practice, but also insurance protection, technology and patient communication. ”

Dr. Hannah Milch, Co -registration researcher and UCLA Pi and assistant professor radiology in ucla

“There are many hope that AI will improve care, but very little strict attempts assessed its real effects,” Elmore said. “This is our opportunity to generate trustworthy evidence from the perspective of patients and the center.”

In addition to the analysis of cancer detection and withdrawal indicators, the study will include focus groups and surveys to capture how patients and radiologists perceive and trust AI’s care.

Joint efforts in six states

The test of the prism combines seven leading academic medical centers:

• Ucla (Administrative coordinating party managed by Dual-Percipal investigator, dr. Joann Elmore; Competition and PI Pi, dr Hannah Milch)
• UC Davis (Data coordinating center run by a double -level researcher, Dr. Diana Miglitretti)
• Boston Medical Center (Main researcher, Dr. Clare B. Ponton)
• Uc san diego health (Main investigator, Dr. Haydeee Ojeda-Fournier)
• University of Miami (Main researcher, dr Jose M. Net)
• University of Washington – Fred Hutchinson Cancer Center (Main researcher of the page, dr Jan M. Lee)
• University of Wisconsin – Madison (Cooperative researcher, dr Christoph I. Lee; main researchers, Dr. Mai Elepaba and Dr. Ryan Woods)

Hitting the right balance

Elmore emphasized that the goal is not to replace human knowledge, but to understand how AI can complement it. “Our experts will continue to make a final connection. AI can be a useful second pilot – but it is the radiologist who holds the wheel.”

The study is expected to inform future political decisions, the best practices in the field of screening and integrate new technologies with patient care as effectively.