Breaking News
January 16, 2019 - Is mindfulness worthy of all the hype?
January 16, 2019 - Physical Activity, Any Type or Amount, Cuts Health Risk from Sitting
January 16, 2019 - New understanding in the evolution of human feet
January 15, 2019 - AHA: New Cholesterol Guidelines Put Ethnicity in the Spotlight
January 15, 2019 - Different brain areas linked to smoking and drinking
January 15, 2019 - Henry Marsh shares insights into neurosurgery and more at Dean’s Lecture Series
January 15, 2019 - Want to Live Longer? For Just 30 Minutes a Day, Do Anything Else But Sit
January 15, 2019 - The Current issue of “The view from here” is concerned with Targets
January 15, 2019 - Plain packaging sparked tobacco price rises, new study finds
January 15, 2019 - Sedentary lifestyles can be unhealthy, physical activity can lower risk
January 15, 2019 - Gut microbiome may help prevent development of cow’s milk allergy
January 15, 2019 - Lesbian, gay and bisexual individuals more likely to suffer severe substance use disorders
January 15, 2019 - New England Journal of Medicine Publishes Positive Results of the Pivotal Trial of Cablivi (caplacizumab) for Rare Blood Clotting Disorder
January 15, 2019 - Levels of inflammatory marker (CRP) linked to housing type and tenure
January 15, 2019 - Three gifts I’m glad I gave myself in 2018
January 15, 2019 - Columbia’s Pediatrics Department Names New Vice Chairs, Expands Leadership
January 15, 2019 - US FDA Accepts Regulatory Submissions for Review of Tafamidis to Treat Transthyretin Amyloid Cardiomyopathy
January 15, 2019 - Staying fit can cut your risk of heart attack by half
January 15, 2019 - Vitamin D supplements are of no gain to those over 70, study shows
January 15, 2019 - Scientists create comprehensive new method to predict breast cancer risk
January 15, 2019 - Research shows connection between social media use and impaired risky decision-making
January 15, 2019 - FDA Approves Expanded Use of Adacel (Tdap) Vaccine for Repeat Vaccination
January 15, 2019 - Treating spinal pain with replacement discs made of ‘engineered living tissue’ moves closer to reality
January 15, 2019 - Providers Walk ‘Fine Line’ Between Informing And Scaring Immigrant Patients
January 15, 2019 - Outcomes Poorer for Medicaid Beneficiaries With STEMI
January 15, 2019 - Decorative Products on Foods Can Be Unsafe
January 15, 2019 - A dream of sustainable surgery in Uganda
January 15, 2019 - Study shows how herpes viruses and tumors have learned to manipulate the same ancient RNA
January 15, 2019 - Common Heart, Diabetes Meds May Help Ease Mental Illness
January 15, 2019 - Stress and trauma in earliest years linked to reduced hippocampal volume in adolescence
January 15, 2019 - Scientists identify endogenous activator of sigma-1 receptors in human cells
January 15, 2019 - MAR treatments unlikely to be cause of premature or low birth weight babies
January 15, 2019 - Parental CPTSD increases transmission of trauma to offspring of Tutsi genocide survivors
January 15, 2019 - High-fat diets shown to increase blood pressure
January 15, 2019 - New institute for food safety to be established in Netherlands
January 15, 2019 - Keele University researchers receive £2.4 million grant to help reduce overprescribing of opioids
January 15, 2019 - Synthetic compound reverses mutant p53 aggregate accumulation, study shows
January 15, 2019 - First elder care robot tested in a WSU smart home apartment
January 15, 2019 - Oxford researchers explore relationship between technology use and adolescent mental health
January 15, 2019 - From microbiome research to healthier and sustainable foods
January 15, 2019 - How coaching moms and dads improves infants’ language skills
January 15, 2019 - Precision health approach tapped to identify causes of poverty
January 14, 2019 - DNA origami can accurately measure how antibodies interact with several antigens
January 14, 2019 - Researchers identify multiple new subtypes of most common childhood cancer
January 14, 2019 - Total Fertility Rates Vary by State
January 14, 2019 - Elevated blood lead level in early childhood associated with increased risk of academic problems in school-aged children
January 14, 2019 - Superior technique identified that can block CRISPR gene editing
January 14, 2019 - Turning breast cancer cells into fat cells prevents the formation of metastases
January 14, 2019 - Review examines what influences HIV-positive patients to stay on antiretroviral drugs in Africa
January 14, 2019 - Identifying genetic factors that lead to squamous cell carcinoma
January 14, 2019 - Virtual video visits can replace office visits without compromising quality of care
January 14, 2019 - Health Highlights: Jan. 10, 2019
January 14, 2019 - Molecular hallmarks of tumor hypoxia across 19 cancer types discovered
January 14, 2019 - Scientists uncover how protein clumps damage cells in Parkinson’s
January 14, 2019 - Physician-scientist’s “indomitable spirit” prevails over personal adversity
January 14, 2019 - King’s researchers receive £1.25 million to investigate fatal eating disorder
January 14, 2019 - UCR researchers uncover how plants sense temperature
January 14, 2019 - Scientists find link between colitis and colon cancer
January 14, 2019 - New skin patch provides long-acting contraceptive protection
January 14, 2019 - Asparagine synthetase deficiency – Genetics Home Reference
January 14, 2019 - Improved stem cell approach could aid fight against Parkinson’s
January 14, 2019 - New class of sleeping pill preserves ability to wake in response to danger signals
January 14, 2019 - Cancer patients are four times more likely to commit suicide
January 14, 2019 - The human brain works in reverse order to retrieve memories
January 14, 2019 - Simple tips can lead to better food choices
January 14, 2019 - Meth’s Resurgence Spotlights Lack Of Meds To Combat The Addiction
January 14, 2019 - TARA Biosystems and Insilico Medicine collaborate to discover novel therapies for cardiac disease
January 14, 2019 - Early life stress in mice affects their offspring behavior
January 14, 2019 - Depression Tied to Worse Asthma Outcomes in Urban Teens
January 14, 2019 - Santa calorie counting
January 14, 2019 - Opiod prescriptions for pet dogs misused by their masters
January 14, 2019 - People with ASD could be better at recognizing regret and relief in others finds study
January 14, 2019 - Conducting ChIP-Seq with Low Cell Numbers
January 14, 2019 - Study explores support and social networks of family carers of people with dementia
January 14, 2019 - At Risk for an Opioid OD? There’s an App for That
January 14, 2019 - Single national electronic health record will help improve care in Canadian hospitals
January 14, 2019 - Study unearths Britain’s first speech therapists
January 14, 2019 - Study reveals nuances of racial inequalities in breast cancer prevention
January 14, 2019 - Air pollution can raise the risk of miscarriage among women finds study
January 14, 2019 - An extra meal a day cuts deaths by half in elderly with hip fractures
New automated model identifies dense breast tissue in mammograms

New automated model identifies dense breast tissue in mammograms

image_pdfDownload PDFimage_print

Researchers from MIT and Massachusetts General Hospital have developed an automated model that assesses dense breast tissue in mammograms — which is an independent risk factor for breast cancer — as reliably as expert radiologists.

This marks the first time a deep-learning model of its kind has successfully been used in a clinic on real patients, according to the researchers. With broad implementation, the researchers hope the model can help bring greater reliability to breast density assessments across the nation.

It’s estimated that more than 40 percent of U.S. women have dense breast tissue, which alone increases the risk of breast cancer. Moreover, dense tissue can mask cancers on the mammogram, making screening more difficult. As a result, 30 U.S. states mandate that women must be notified if their mammograms indicate they have dense breasts.

But breast density assessments rely on subjective human assessment. Due to many factors, results vary — sometimes dramatically — across radiologists. The MIT and MGH researchers trained a deep-learning model on tens of thousands of high-quality digital mammograms to learn to distinguish different types of breast tissue, from fatty to extremely dense, based on expert assessments. Given a new mammogram, the model can then identify a density measurement that closely aligns with expert opinion.

“Breast density is an independent risk factor that drives how we communicate with women about their cancer risk. Our motivation was to create an accurate and consistent tool, that can be shared and used across health care systems,” says second author Adam Yala, a PhD student in MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL).

The other co-authors are first author Constance Lehman, professor of radiology at Harvard Medical School and the director of breast imaging at the MGH; and senior author Regina Barzilay, the Delta Electronics Professor at CSAIL and the Department of Electrical Engineering and Computer Science at MIT.

Mapping density

The model is built on a convolutional neural network (CNN), which is also used for computer vision tasks. The researchers trained and tested their model on a dataset of more than 58,000 randomly selected mammograms from more than 39,000 women screened between 2009 and 2011. For training, they used around 41,000 mammograms and, for testing, about 8,600 mammograms.

Each mammogram in the dataset has a standard Breast Imaging Reporting and Data System (BI-RADS) breast density rating in four categories: fatty, scattered (scattered density), heterogeneous (mostly dense), and dense. In both training and testing mammograms, about 40 percent were assessed as heterogeneous and dense.

During the training process, the model is given random mammograms to analyze. It learns to map the mammogram with expert radiologist density ratings. Dense breasts, for instance, contain glandular and fibrous connective tissue, which appear as compact networks of thick white lines and solid white patches. Fatty tissue networks appear much thinner, with gray area throughout. In testing, the model observes new mammograms and predicts the most likely density category.

Matching assessments

The model was implemented at the breast imaging division at MGH. In a traditional workflow, when a mammogram is taken, it’s sent to a workstation for a radiologist to assess. The researchers’ model is installed in a separate machine that intercepts the scans before it reaches the radiologist, and assigns each mammogram a density rating. When radiologists pull up a scan at their workstations, they’ll see the model’s assigned rating, which they then accept or reject.

“It takes less than a second per image … [and it can be] easily and cheaply scaled throughout hospitals.” Yala says.

On over 10,000 mammograms at MGH from January to May of this year, the model achieved 94 percent agreement among the hospital’s radiologists in a binary test — determining whether breasts were either heterogeneous and dense, or fatty and scattered. Across all four BI-RADS categories, it matched radiologists’ assessments at 90 percent. “MGH is a top breast imaging center with high inter-radiologist agreement, and this high quality dataset enabled us to develop a strong model,” Yala says.

In general testing using the original dataset, the model matched the original human expert interpretations at 77 percent across four BI-RADS categories and, in binary tests, matched the interpretations at 87 percent.

In comparison with traditional prediction models, the researchers used a metric called a kappa score, where 1 indicates that predictions agree every time, and anything lower indicates fewer instances of agreements. Kappa scores for commercially available automatic density-assessment models score a maximum of about 0.6. In the clinical application, the researchers’ model scored 0.85 kappa score and, in testing, scored a 0.67. This means the model makes better predictions than traditional models.

In an additional experiment, the researchers tested the model’s agreement with consensus from five MGH radiologists from 500 random test mammograms. The radiologists assigned breast density to the mammograms without knowledge of the original assessment, or their peers’ or the model’s assessments. In this experiment, the model achieved a kappa score of 0.78 with the radiologist consensus.

Next, the researchers aim to scale the model into other hospitals. “Building on this translational experience, we will explore how to transition machine-learning algorithms developed at MIT into clinic benefiting millions of patients,” Barzilay says. “This is a charter of the new center at MIT — the Abdul Latif Jameel Clinic for Machine Learning in Health at MIT — that was recently launched. And we are excited about new opportunities opened up by this center.”

Source:

http://news.mit.edu/2018/AI-identifies-dense-tissue-breast-cancer-mammograms-1016

Tagged with:

About author

Related Articles