AI model predicts breast cancer risk without racial bias
IMAGE:
ABNORMAL MAMMOGRAM.
view moreCREDIT: RSNA
CHICAGO – A deep learning artificial intelligence (AI) model that was developed using only mammogram image biomarkers accurately predicted both ductal carcinoma in situ (DCIS) and invasive carcinoma, according to research being presented today at the annual meeting of the Radiological Society of North America (RSNA). Additionally, the model showed no bias across multiple races.
Traditional breast cancer risk assessment models use information obtained from patient questionnaires, such as medical and reproductive history, to calculate a patient’s future risk of developing breast cancer.
“In the domain of precision medicine, risk-based screening has been elusive because we have not been able to accurately evaluate a woman’s risk of developing breast cancer,” said study lead author Leslie R. Lamb, M.D., M.Sc., a breast radiologist at Massachusetts General Hospital (MGH) in Boston. “Even the best existing traditional risk models do not perform well on the individual level.”
Traditional risk models have also demonstrated poor performance across different patient races, most likely due to the data used to develop the model.
“Traditional models likely have racial biases due to the populations on which they were developed,” Dr. Lamb said. “Several of the commonly used models were developed on predominantly European Caucasian populations.”
According to the American Cancer Society, Black women demonstrate the lowest 5-year relative survival rate for breast cancer among all racial and ethnic groups. This translates to a persistent 6% to 8% disparity in 5-year survival rates between Black and white women across all breast cancer types.
To accurately determine breast cancer risk, foster early detection and improve patient survival rates, it is important that risk models are developed that are applicable across different populations.
A deep learning AI risk assessment model developed using mammographic images alone can outperform traditional risk assessment models in future breast cancer development while also mitigating the racial biases seen in traditional models.
In the first study of its kind, Dr. Lamb and colleagues sought to assess the performance of an image-based deep learning risk assessment model in predicting both future invasive breast cancer and DCIS across multiple races.
The model’s performance was assessed by comparing areas under the receiver operating characteristic curve (AUC) with the DeLong test. The AUC score measures the predictive rate of the model on a scale of from 0 to 1. Multiple prior studies have estimated traditional risk model performance measured by AUC in the range of 0.59-0.62 for white women, with much lower performance in women of other races.
The multisite study included 129,340 routine bilateral screening mammograms performed in 71,479 women between 2009 to 2018 with five-year follow-up data. Patient demographics were obtained from electronic medical records, and instances of cancer were identified from the regional tumor registry.
The racial makeup of the study group included white (106,839 exams), Black (6,154 exams), Asian (6,435 exams), self-reported other races (6,257 exams) and unknown (3,655 exams). The mean age of the women was 59 years old.
The deep learning model consistently outperformed traditional risk models in predicting a woman’s risk of developing DCIS, which is early-stage breast cancer, and invasive breast cancer, which is cancer that has potential to spread.
“The model is able to translate the full diversity of subtle imaging biomarkers in the mammogram, beyond what the naked eye can see, that can predict a woman’s future risk of both DCIS and invasive breast cancer,” Dr. Lamb said. “The deep learning image-only risk model can provide increased access to more accurate, equitable and less costly risk assessment.”
The predictive rate of both DCIS and invasive cancer was 0.71 across all races. The AUC in predicting DCIS was 0.77 in non-white patients and 0.71 in white patients. The AUC in predicting invasive cancer was 0.72 in non-white patients and 0.71 in white patients.
“This is a particularly exciting domain for AI, as it demonstrates the opportunity to apply ‘AI for good’—to reduce well-known racial disparities in risk assessment,” said senior author Constance D. Lehman, M.D., Ph.D., a breast radiologist at MGH. “We are now poised to translate these findings into improved clinical care for our patients.”
Additional co-authors are Sarah F. Mercaldo, Ph.D., and Andrew R. Carney, M.S.
###
Note: Copies of RSNA 2023 news releases and electronic images will be available online at RSNA.org/press23.
RSNA is an association of radiologists, radiation oncologists, medical physicists and related scientists promoting excellence in patient care and health care delivery through education, research and technologic innovation. The Society is based in Oak Brook, Illinois. (RSNA.org)
Editor’s note: The data in these releases may differ from those in the published abstract and those actually presented at the meeting, as researchers continue to update their data right up until the meeting. To ensure you are using the most up-to-date information, please call the RSNA Newsroom at 1-312-791-6610.
For patient-friendly information on breast imaging, visit RadiologyInfo.org.
Regular screening mammograms significantly reduce breast cancer deaths
IMAGE:
ABNORMAL MAMMOGRAM.
view moreCREDIT: RSNA/RADIOLOGY
CHICAGO – Breast cancer mortality is significantly reduced when women regularly attend screening mammograms, according to research being presented today at the annual meeting of the Radiological Society of North America (RSNA).
Early detection of breast cancer, before symptoms are present, is key to survivability. According to the American Cancer Society, women between the ages of 45 and 54 should get mammograms every year. Women who are 55 years and older can switch to every other year or continue with annual mammograms. Skipping just one scheduled mammogram could result in a more advanced breast cancer diagnosis, significantly impacting a patient’s chance of survival.
“The purpose of mammography is to detect breast cancer during the few years it can be seen on a mammogram, but before symptoms are apparent,” said study author Robert A. Smith, Ph.D., senior vice president and director of the American Cancer Society Center for Cancer Screening in Atlanta, Georgia. “If a woman unknowingly has breast cancer and misses or postpones her mammogram during this time when she has no symptoms, but her breast cancer is growing and perhaps spreading, then the window for early detection will be lost.”
Even though regular mammograms are an important factor in early breast cancer detection, there are still many barriers that restrict women from receiving this preventative care, including access and work or family obligations.
“It is challenging to keep track of your schedule, and in the U.S., many women do not receive reminders. Further, for all of us, the obligations of work and family compete with our scheduled health care,” Dr. Smith said.
Dr. Smith and colleagues sought to identify the exact impact of missing even one mammogram.
The researchers obtained women’s screening history from oncology centers throughout Sweden for a period from 1992 to 2016. A total of 36,079 breast cancer patients were included in the study.
Using data from the Swedish Cause of Death Register, the researchers identified 4,564 breast cancer deaths among the patients included in the study.
The researchers then tracked all of the women’s participation in five or fewer most recent invitations for breast cancer screening prior to cancer diagnosis.
Women who attended all their invited screening mammograms had a survivability rate of over 80%. Women who didn’t participate in any screenings had a survival rate that ranged from 59.1% to 77.6%.
Women who attended all five screening mammograms saw a 72% reduction in the risk of dying from breast cancer compared to women who didn’t participate in any screening mammograms. Even after conservative adjustment for potential self-selection factors, there was a highly significant 66% reduction in the risk of breast cancer death.
“Women who attended all five previous mammography examinations prior to a diagnosis of breast cancer were nearly three times less likely to die from breast cancer compared with women who had not attended any examinations, and each additional examination attended among the five previous examinations conferred an additive protective effect against dying from breast cancer,” Dr. Smith said.
The researchers stressed that imaging facilities should prioritize getting patients in for screening at the earliest opportunity. This is especially important when women have to cancel their appointments. Facilities should reschedule these screening mammograms for the next earliest available appointment.
“These findings show that as much as possible, adherence to regular mammography screening is the very best insurance a woman has against being diagnosed with an advanced breast cancer that could be life-threatening,” Dr. Smith said.
Co-authors are Stephen W. Duffy, M.Sc., Amy Ming-Fang Yen, Ph.D., László Tabár, M.D., Abbie Ting-Yu, Ph.D., Sam Li-Sheng Chen, Ph.D., Chen-Yant Hsu, M.D., Peter B. Dean, M.D., and Tony Hsiu-His Chen, Ph.D.
###
Note: Copies of RSNA 2023 news releases and electronic images will be available online at RSNA.org/press23.
RSNA is an association of radiologists, radiation oncologists, medical physicists and related scientists promoting excellence in patient care and health care delivery through education, research and technologic innovation. The Society is based in Oak Brook, Illinois. (RSNA.org)
Editor’s note: The data in these releases may differ from those in the published abstract and those actually presented at the meeting, as researchers continue to update their data right up until the meeting. To ensure you are using the most up-to-date information, please call the RSNA Newsroom at 1-312-791-6610.
For patient-friendly information on breast cancer screening with mammography, visit RadiologyInfo.org.
Artificial intelligence paves way for new medicines
Researchers have developed an AI model that can predict where a drug molecule can be chemically altered.
A team of researchers from LMU, ETH Zurich, and Roche Pharma Research and Early Development (pRED) Basel has used artificial intelligence (AI) to develop an innovative method that predicts the optimal method for synthesizing drug molecules. “This method has the potential to significantly reduce the number of required lab experiments, thereby increasing both the efficiency and sustainability of chemical synthesis,” says David Nippa, lead author of the corresponding paper, which has been published in the journal Nature Chemistry. Nippa is a doctoral student in Dr. David Konrad’s research group at the Faculty of Chemistry and Pharmacy at LMU and at Roche.
Active pharmaceutical ingredients typically consist of a framework to which functional groups are attached. These groups enable a specific biological function. To achieve new or improved medical effects, functional groups are altered and added to new positions in the framework. However, this process is particularly challenging in chemistry, as the frameworks, which mainly consist of carbon and hydrogen atoms, are hardly reactive themselves. One method of activating the framework is the so-called borylation reaction. In this process, a chemical group containing the element boron is attached to a carbon atom of the framework. This boron group can then be replaced by a variety of medically effective groups. Although borylation has great potential, it is difficult to control in the lab.
Together with Kenneth Atz, a doctoral student at ETH Zurich, David Nippa developed an AI model that was trained on data from trustworthy scientific works and experiments from an automated lab at Roche. It can successfully predict the position of borylation for any molecule and provides the optimal conditions for the chemical transformation. “Interestingly, the predictions improved when the three-dimensional information of the starting materials were taken into account, not just their two-dimensional chemical formulas,” says Atz.
The method has already been successfully used to identify positions in existing active ingredients where additional active groups can be introduced. This helps researchers develop new and more effective variants of known drug active ingredients more quickly.
JOURNAL
Nature
ARTICLE TITLE
Enabling late-stage drug diversification by high-throughput experimentation with geometric deep learning
AI may aid in diagnosing adolescents with ADHD
IMAGE:
USING ARTIFICIAL INTELLIGENCE (AI) TO ANALYZE SPECIALIZED BRAIN MRI SCANS OF ADOLESCENTS WITH AND WITHOUT ATTENTION-DEFICIT/HYPERACTIVITY DISORDER (ADHD), RESEARCHERS FOUND SIGNIFICANT DIFFERENCES IN NINE BRAIN WHITE MATTER TRACTS IN INDIVIDUALS WITH ADHD.
view moreCREDIT: RSNA
CHICAGO – Using artificial intelligence (AI) to analyze specialized brain MRI scans of adolescents with and without attention-deficit/hyperactivity disorder (ADHD), researchers found significant differences in nine brain white matter tracts in individuals with ADHD. Results of the study will be presented today at the annual meeting of the Radiological Society of North America (RSNA).
ADHD is a common disorder often diagnosed in childhood and continuing into adulthood, according to the Centers for Disease Control and Prevention. In the U.S., an estimated 5.7 million children and adolescents between the ages of 6 and 17 have been diagnosed with ADHD.
“ADHD often manifests at an early age and can have a massive impact on someone’s quality of life and ability to function in society,” said study co-author Justin Huynh, M.S., a research specialist in the Department of Neuroradiology at the University of California, San Francisco, and medical student at the Carle Illinois College of Medicine at Urbana-Champaign. “It is also becoming increasingly prevalent in society among today’s youth, with the influx of smartphones and other distracting devices readily accessible.”
Children with ADHD may have trouble paying attention, controlling impulsive behaviors or regulating activity. Early diagnosis and intervention are key to managing the condition.
“ADHD is extremely difficult to diagnose and relies on subjective self-reported surveys,” Huynh said. “There is definitely an unmet need for more objective metrics for diagnosis. That’s the gap we are trying to fill.”
Huynh said this is the first study to apply deep learning, a type of AI, to identify markers of ADHD in the multi-institutional Adolescent Brain Cognitive Development (ABCD) Study, which includes brain imaging, clinical surveys and other data on over 11,000 adolescents from 21 research sites in the U.S. The brain imaging data included a specialized type of MRI called diffusion-weighted imaging (DWI).
“Prior research studies using AI to detect ADHD have not been successful due to a small sample size and the complexity of the disorder,” Huynh said.
The research team selected a group of 1,704 individuals from the ABCD dataset, including adolescents with and without ADHD. Using DWI scans, the researchers extracted fractional anisotropy (FA) measurements along 30 major white matter tracts in the brain. FA is a measure of how water molecules move along the fibers of white matter tracts.
The FA values from 1,371 individuals were used as input for training a deep-learning AI model, which was then tested on 333 patients, including 193 diagnosed with ADHD and 140 without. ADHD diagnoses were determined by the Brief Problem Monitor assessment, a rating tool used for monitoring a child’s functioning and their responses to interventions.
With the help of AI, the researchers discovered that in patients with ADHD, FA values were significantly elevated in nine white matter tracts.
“These differences in MRI signatures in individuals with ADHD have never been seen before at this level of detail,” Huynh said. “In general, the abnormalities seen in the nine white matter tracts coincide with the symptoms of ADHD.”
The researchers intend to continue obtaining data from the rest of the individuals in the ABCD dataset, comparing the performance of additional AI models.
“Many people feel that they have ADHD, but it is undiagnosed due to the subjective nature of the available diagnostic tests,” Huynh said. “This method provides a promising step towards finding imaging biomarkers that can be used to diagnose ADHD in a quantitative, objective diagnostic framework,” Huynh said.
Co-authors are Pierre F. Nedelec, M.S., M.T.M., Samuel Lashof-Regas, Michael Romano, M.D., Ph.D., Leo P. Sugrue, M.D., Ph.D., and Andreas M. Rauschecker, M.D., Ph.D.
###
Note: Copies of RSNA 2023 news releases and electronic images will be available online at RSNA.org/press23.
RSNA is an association of radiologists, radiation oncologists, medical physicists and related scientists promoting excellence in patient care and health care delivery through education, research and technologic innovation. The Society is based in Oak Brook, Illinois. (RSNA.org)
Editor’s note: The data in these releases may differ from those in the published abstract and those actually presented at the meeting, as researchers continue to update their data right up until the meeting. To ensure you are using the most up-to-date information, please call the RSNA Newsroom at 1-312-791-6610.
For patient-friendly information on brain MRI, visit RadiologyInfo.org.
No comments:
Post a Comment