Nav: Home

New AI tool developed by Dana-Farber identifies cancer outcomes using radiology reports

July 25, 2019

BOSTON - Scientists at Dana-Farber Cancer Institute have demonstrated that an artificial intelligence tool can perform as well as human reviewers - and much more rapidly - in extracting clinical information regarding changes in tumors from unstructured radiology reports for patients with lung cancer.

The AI tool performed comparably to trained human "curators" in detecting the presence of cancer; and whether it was responding to treatment interventions, stable or worsening.

The goal of the study, said corresponding author Kenneth Kehl, MD, MPH, a medical oncologist and faculty member of the Population Sciences Department at Dana-Faber, was to determine whether artificial intelligence tools can extract the most high-value cancer outcomes from radiology reports, which are a ubiquitous but unstructured data source.

Kehl noted that electronic health records now collect vast amounts of information on thousands of patients seen at a center like Dana-Farber. However, unless the patients are enrolled in clinical trials, information about their outcomes, such as whether their cancers grow or shrink in response to treatment, is recorded only in the text of the medical record. Historically, this unstructured information is not amenable to computational analysis and therefore could not be used for research into the effectiveness of treatment.

Because of studies like the Profile initiative at Dana-Farber/Brigham and Women's Cancer Center, which analyzes patient tumor samples and creates profiles that reveal genomic variants that may predict responsiveness to treatments, Dana-Farber researchers have accumulated a wealth of molecular information about patients' cancers. "But it can be difficult to apply this information to understand what molecular patterns predict benefit from treatments without intensive review of patients' medical records to measure their outcomes. This is a critical barrier to realizing the full potential of precision medicine," said Kehl.

For the current study, Kehl and colleagues obtained over 14,000 imaging reports for 1,112 patients and manually reviewed records using the "PRISSMM" framework. PRISSMM is a phenomic data standard developed at Dana-Farber that takes unstructured data from text reports in electronic health records and structures them so that they can be readily analyzed. PRISSMM structures data pertaining to a patient's pathology, radiology/imaging, signs/symptoms, molecular markers, and a medical oncologist's assessment to create a portrait of the cancer patient journey.

Human reviewers analyzed the imaging text reports and noted whether cancer was present and, if so, whether it was worsening or improving, and if the cancer had spread to specific body sites. These reports were then used to train a computational "deep learning" model to recognize these outcomes from the text reports. "Our hypothesis was that deep learning algorithms could use routinely generated radiology text reports to identify the presence of cancer and changes in its extent over time," the authors wrote.

The researchers compared human and computer measurements of outcomes such as disease-free survival, progression-free survival, and time to improvement or response, and found that the AI algorithm could replicate human assessment of these outcomes. The deep learning algorithms were then applied to annotate another 15,000 reports for 1,294 patients whose records had not been manually reviewed. The authors found that computer outcome measurements among these patients predicted survival with similar accuracy to human assessments among the manually reviewed patients.

The human curators were able to annotate imaging reports for about three patients per hour, a rate at which one curator would need about six months to annotate all of the nearly 30,000 imaging reports for the patients in the cohort. By contrast, the artificial intelligence model that the researchers developed could annotate the imaging reports for the cohort in about 10 minutes, the researchers said in a report in JAMA Oncology.

"To create a true learning health system for oncology and to facilitate delivery of precision medicine at scale, methods are needed to accelerate curation of cancer-related outcomes from electronic health records," said the authors of the publication. If applied widely, the investigators said, "this technique could substantially accelerate efforts to use real-world data from all patients with cancer to generate evidence regarding effectiveness of treatment approaches." Next steps will include testing this approach on EHR data from other cancer centers and using the data to discover which treatments work best for which patients.
-end-
The senior author of study is Deborah Schrag, MD, MPH, Chief of Division of Population Sciences at Dana-Farber and a medical oncologist.

The research was supported by grants from the National Cancer Institute (K05CA169384, RO1CA203636, and U01CA209414); the Claude Dauphin Philanthropic Research Fund, and the Simeon J. Fortin Foundation.

Kenneth Kehl reports serving as a consultant to Aetion, Inc. Conflicts reported by Deborah Schrag include serving as a consultant/advisor to Cureus and Pfizer, receiving research funding from the American Association of Cancer Research; serving as an editor and receiving personal fees from JAMA outside the related work; and holding equity in mixed biotech funds including QQQ. She also reports holding a trademark and having financial interest in the PRISSMM phenomic data standard.

About Dana-Farber Cancer Institute

Dana-Farber Cancer Institute is one of the world's leading centers of cancer research and treatment. It is the only center ranked in the top 4 of U.S. News and World Report's Best Hospitals for both adult and pediatric cancer care.

Dana-Farber's mission is to reduce the burden of cancer through scientific inquiry, clinical care, education, community engagement, and advocacy. We provide the latest in cancer for adults through Dana-Farber/Brigham and Women's Cancer Care and for children through Dana-Farber/Boston Children's Cancer and Blood Disorders Center. Dana-Farber is dedicated to a unique and equal balance between cancer research and care, translating the results of discovery into new treatments for patients locally and around the world.

Dana-Farber Cancer Institute

Related Cancer Articles:

Radiotherapy for invasive breast cancer increases the risk of second primary lung cancer
East Asian female breast cancer patients receiving radiotherapy have a higher risk of developing second primary lung cancer.
Cancer genomics continued: Triple negative breast cancer and cancer immunotherapy
Continuing PLOS Medicine's special issue on cancer genomics, Christos Hatzis of Yale University, New Haven, Conn., USA and colleagues describe a new subtype of triple negative breast cancer that may be more amenable to treatment than other cases of this difficult-to-treat disease.
Metabolite that promotes cancer cell transformation and colorectal cancer spread identified
Osaka University researchers revealed that the metabolite D-2-hydroxyglurate (D-2HG) promotes epithelial-mesenchymal transition of colorectal cancer cells, leading them to develop features of lower adherence to neighboring cells, increased invasiveness, and greater likelihood of metastatic spread.
UH Cancer Center researcher finds new driver of an aggressive form of brain cancer
University of Hawai'i Cancer Center researchers have identified an essential driver of tumor cell invasion in glioblastoma, the most aggressive form of brain cancer that can occur at any age.
UH Cancer Center researchers develop algorithm to find precise cancer treatments
University of Hawai'i Cancer Center researchers developed a computational algorithm to analyze 'Big Data' obtained from tumor samples to better understand and treat cancer.
New analytical technology to quantify anti-cancer drugs inside cancer cells
University of Oklahoma researchers will apply a new analytical technology that could ultimately provide a powerful tool for improved treatment of cancer patients in Oklahoma and beyond.
Radiotherapy for lung cancer patients is linked to increased risk of non-cancer deaths
Researchers have found that treating patients who have early stage non-small cell lung cancer with a type of radiotherapy called stereotactic body radiation therapy is associated with a small but increased risk of death from causes other than cancer.
Cancer expert says public health and prevention measures are key to defeating cancer
Is investment in research to develop new treatments the best approach to controlling cancer?
UI Cancer Center, Governors State to address cancer disparities in south suburbs
The University of Illinois Cancer Center and Governors State University have received a joint four-year, $1.5 million grant from the National Cancer Institute to help both institutions conduct community-based research to reduce cancer-related health disparities in Chicago's south suburbs.
Leading cancer research organizations to host international cancer immunotherapy conference
The Cancer Research Institute, the Association for Cancer Immunotherapy, the European Academy of Tumor Immunology, and the American Association for Cancer Research will join forces to sponsor the first International Cancer Immunotherapy Conference at the Sheraton New York Times Square Hotel in New York, Sept.

Related Cancer Reading:

Best Science Podcasts 2019

We have hand picked the best science podcasts for 2019. Sit back and enjoy new science podcasts updated daily from your favorite science news services and scientists.
Now Playing: TED Radio Hour

Anthropomorphic
Do animals grieve? Do they have language or consciousness? For a long time, scientists resisted the urge to look for human qualities in animals. This hour, TED speakers explore how that is changing. Guests include biological anthropologist Barbara King, dolphin researcher Denise Herzing, primatologist Frans de Waal, and ecologist Carl Safina.
Now Playing: Science for the People

#SB2 2019 Science Birthday Minisode: Mary Golda Ross
Our second annual Science Birthday is here, and this year we celebrate the wonderful Mary Golda Ross, born 9 August 1908. She died in 2008 at age 99, but left a lasting mark on the science of rocketry and space exploration as an early woman in engineering, and one of the first Native Americans in engineering. Join Rachelle and Bethany for this very special birthday minisode celebrating Mary and her achievements. Thanks to our Patreons who make this show possible! Read more about Mary G. Ross: Interview with Mary Ross on Lash Publications International, by Laurel Sheppard Meet Mary Golda...