Roundup: Health Systems Embracing AI Imaging Tools
Healthcare systems that embrace artificial intelligence tools can improve not only radiology operations and quality, but also patient follow-up, which can result in greater staff efficiency, higher patient completion rates, and better patient access and outcomes.
A collaboration between East Alabama Medical Center, Inflo Health and the American College of Radiology Learning Network began using machine language models and advanced natural language processing to extract data from radiology reports to improve follow-up in lung patients, while Stamford Health in Connecticut was able to to extend additional radiological measures to all cardiovascular diseases through automation.
Also of note this week, Lunit, a provider of cancer diagnostics and therapeutics, announced that two recent studies evaluating AI-powered mammography screening found that the technology could also estimate breast cancer progression up to six years before a positive diagnosis.
“If the scores of commercial AI algorithms developed for the immediate detection of cancer can also estimate future cancer risk, then more accurate and reliable near-term risk estimation could lead to tailor-made, personalized preventive measures, which may result in earlier detection of breast cancer and less aggressive measures. treatment,” European researchers said in a statement on Wednesday.
EAMC improves patient follow-up
The Alabama health organization announced Thursday that a partnership to track radiology follow-up with AI – and involving primary care physicians in acute care communications – has changed the follow-up rate of recommendations by 74%.
EAMC partnered with Inflo Health, which uses radiology-specific language models and advanced NLP, and the American College of Radiology to increase patient engagement and physician productivity.
The AI-powered software runs against measurement specifications set forth by ACR’s ImPower program – which helps organizations build improvement leadership skills and practices to achieve better outcomes – providing EAMC radiologists with additional imaging recommendations and identify actionable findings, as well as automate department workflows.
The goal of the collaboration with EAMC was to improve the consistent uptake of post-scan recommendations for incidentally detected lung nodules and also increase the percentage of studies that received timely follow-up, the organizations said in a statement.
EAMC also implemented the AI software’s eligibility measures, which automated the process of identifying incidental lung nodules that met inclusion criteria.
This effort has significantly streamlined EAMC’s processes, reduced manual efforts and increased staff efficiency, said Melinda Johnson, the organization’s director of radiology.
“This has also allowed us to expand the care navigator role into other clinical areas,” she said in a statement. “This partnership illustrates how the integration of cutting-edge technology with strategic collaboration can set new standards in radiology practice and operational excellence.”
The result was a reduction in manual tasks from five hours per week to just 15 minutes, representing a 95% improvement in efficiency, employees said.
To improve patient completion and communicate recommended imaging follow-ups, EAMC addressed operational barriers including inconsistent communication between acute care and primary care. As a bonus, that effort generated an estimated $9,000 per month in additional revenue
“Leveraging technology to standardize and optimize clinical workflows requires the collective efforts of organizations and their software vendors working together so that the solution is built by understanding the problem,” said Judy Burleson, ACR vice president of quality management programs.
“The quality improvement education and support provided by the ImPower program, coupled with EAMC’s commitment to improving patient outcomes, and Inflo Health’s willingness to adapt their product, made this progress possible,” she said.
Stamford Health is improving access
Stamford Health, a nonprofit organization serving Fairfield County, Connecticut, announced earlier this month a new automated cardiovascular screening that enables timely and personalized follow-up care for at-risk patients.
Stamford Health’s Heart & Vascular Institute said in a statement that the AI-powered cardiovascular screening tool significantly improves the early detection and treatment of cardiovascular disease in the patient population.
The institute uses Bunkerhill Health’s advanced algorithm to identify the presence of coronary calcium by calculating the total coronary calcium or Agatston score, an indicator of future risk of coronary artery disease in a predefined patient population.
CAC screening normally requires a special order from a physician, but the automated algorithm now runs in the background of all of the institute’s unmonitored chest CT scans, such as those used in lung cancer screenings.
“We are focused on providing the most advanced, cutting-edge care to our patients,” says Dr. Ronald Lee, chairman of Stamford Health’s radiology department.
Patients automatically receive a CAC score during each non-contrast chest CT scan. When an elevated CAC is identified, the patient’s primary care physician or cardiologist is informed of their score and risk.
“This tool increases our ability to detect early signs of cardiovascular disease and ensures patients receive the follow-up care they need to prevent serious health problems,” said Dr. David Hsi, chief of cardiology and co-director of the institute.
AI testing for predictive mammography
The accuracy of mammography screening has long been a challenge because radiology protocols often require duplicate scan measurements. AI algorithms can highlight areas of concern and provide breast and exam-level malignant neoplasm scores to assist radiologists in reading images.
Lunit said on Wednesday that researchers from the Cancer Registry of Norway and Odense University Hospital in Denmark who were already using the INSIGHT MMG tools have demonstrated the potential to also improve the predictive value of its national breast cancer screening programmes, ultimately leading to earlier diagnosis and treatment for women.
The Norwegian retrospective studyArtificial Intelligence Algorithm for Subclinical Breast Cancer Detection, completed in August and published earlier this month in the JAMA Network, analyzed imaging data from a cohort of 116,495 women aged 50 to 69 with no history of breast cancer.
The Norwegian Cancer Registry, which has a contract with Lunit for research using AI software, offers digital mammography screening every two years. The patients in the retrospective cohort study underwent at least three consecutive biennial screening examinations, conducted between September 13, 2004 and December 21, 2018, at nine of the country’s breast screening centers.
Researchers divided the cohort into three groups: women with screening-detected breast cancer during the third examination round, women with interval cancer diagnosed after the third examination round, and women without breast cancer diagnosed after three consecutive examinations and six years without a cancer diagnosis. – finding 1,265 screening-detected cancers and 342 interval cancers.
For those identified with breast cancer — defined as ductal carcinoma in situ or invasive breast carcinoma — mean absolute AI scores were higher for breasts that developed compared with those that did not develop cancer four to six years before their eventual detection. AI scores were also higher and increased more rapidly over the three consecutive rounds of screening for women with a diagnosis of screen-detected cancer versus an interval cancer.
“These findings suggest that commercial AI algorithms developed for breast cancer detection could identify women at high risk for future breast cancer and provide an avenue for personalized screening approaches that could lead to earlier cancer diagnosis,” researchers said.
Andrea Fox is editor-in-chief of Healthcare IT News.
Email: afox@himss.org
Healthcare IT News is a HIMSS Media publication.