Breaking News
April 25, 2019 - Cardiometabolic Risk Better ID’d in Children Reclassified to Higher BP
April 25, 2019 - How the obesity epidemic is taking a toll on our bones and joints
April 25, 2019 - E-cigarettes contaminated with dangerous microbial toxins
April 25, 2019 - Researchers document specific characteristics of storefront tobacco advertisements
April 25, 2019 - Oncotype DX-guided treatment could reduce cost for breast cancer care, study suggests
April 25, 2019 - New review highlights how lifestyle affects our genes
April 25, 2019 - Study provides evidence that blood tests can detect Alzheimer’s risk
April 25, 2019 - Physicians turning to antibiotic alternatives for long-term acne treatment
April 25, 2019 - Preschool Is Prime Time to Teach Healthy Lifestyle Habits
April 25, 2019 - Study finds insidious and persistent discrimination among physician mothers
April 25, 2019 - Newly identified skin-gut communication helps illuminate link between food allergy and eczema
April 25, 2019 - Thiazide use linked with reduced risk of low energy fractures in people with Alzheimer’s
April 25, 2019 - Some women are biologically more resilient than others to PTSD
April 25, 2019 - The Current issue of “The view from here” is concerned with Partnerships and Alliances
April 25, 2019 - Imaging method reveals long-lived patterns in cells of the eye
April 25, 2019 - Podcast: KHN’s ‘What The Health?’ The Abortion Wars Rage On
April 25, 2019 - Prolonged exposure therapy is more effective in treating veterans with PTSD, alcohol use disorder
April 24, 2019 - Our artificial cornea breakthrough could lead to self-assembling organs
April 24, 2019 - A Stanford black, female, gay surgery resident speaks out
April 24, 2019 - Donna Lynne on Extreme Sports, Lessons From the '60s, and Taking CUIMC to the Next Level
April 24, 2019 - Pain Clinics’ Doctors Needlessly Tested Hundreds Of Urine Samples, Court Records Show
April 24, 2019 - Researchers uncover potential clue to halt destruction of nerve cells in people with ALS
April 24, 2019 - Study uncovers reasons for poor mental health in bisexual people
April 24, 2019 - Screenings, interventions, and referrals can help adolescents overcome substance abuse
April 24, 2019 - Febrile seizures following vaccination are self-resolving and not dangerous
April 24, 2019 - Flow-UV inline UV-Visible spectrometer monitors dispersion in real time
April 24, 2019 - Rates of Marijuana Use in Cancer Patients on the Rise in U.S.
April 24, 2019 - Versatile drug may protect baby from hazards of intraamniotic infections
April 24, 2019 - Financial transparency may diminish trust in doctors, new study finds
April 24, 2019 - Calling all Riders: Velocity Extends Free Registration 
April 24, 2019 - The Homeless Are Dying In Record Numbers On The Streets Of L.A.
April 24, 2019 - Researchers use brain scans to provide better understanding of unconscious bias
April 24, 2019 - Blocking BRAF ubiquitination may be an effective treatment approach in melanoma
April 24, 2019 - Simple mobility test helps predict hospital readmission in elderly heart attack patients
April 24, 2019 - Novel fluorescence imaging system helps surgeons remove small ovarian tumors
April 24, 2019 - Uncovering the Structure of HIV Integrase to Inform Drug Discovery
April 24, 2019 - Medical Marijuana Use Rising Among Cancer Patients
April 24, 2019 - Artificial intelligence approach optimizes embryo selection for IVF
April 24, 2019 - Doctor or detective? Sleuthing mysteries in medical school
April 24, 2019 - CUIMC Community Gives Blood During Spring 2019 Columbia University Blood Drive
April 24, 2019 - Americans Overwhelmingly Want Federal Protections Against Surprise Medical Bills
April 24, 2019 - Making Laboratories More Efficient with the Most Modern LIMS on the Market
April 24, 2019 - Treating cancer patients with personalized, combination therapies improves outcomes
April 24, 2019 - Researchers engineer new molecules to help stop lung cancer
April 24, 2019 - Acupuncture can be a wonderful tool for preventing number of diseases
April 24, 2019 - Daily life disability before hip replacement may predict poor post-operative outcomes
April 24, 2019 - Study finds involuntary staying in housing estates to be a potential health risk
April 24, 2019 - Older kidney disease patients starting dialysis die at higher rates than previously thought
April 24, 2019 - Time-restricted eating shows promise for controlling blood glucose levels
April 24, 2019 - Ambiguous genitalia in newborns may be more common than previously thought
April 24, 2019 - Research provides important insight on the brain-body connection
April 24, 2019 - In 10 Years, Half Of Middle-Income Elders Won’t Be Able To Afford Housing, Medical Care
April 24, 2019 - Researchers study how E. coli clones have become major cause of drug-resistant infections
April 24, 2019 - Bacterial and fungal toxins found in popular electronic cigarettes
April 24, 2019 - Factors affecting absorption of ‘sunshine vitamin’ during spring/summer months
April 24, 2019 - Texting helps improve medication adherence, health outcomes for patients with schizophrenia
April 24, 2019 - Cochrane Review looks at different ways to use nicotine replacement therapies
April 24, 2019 - New review on relationship between COPD and Type 2 diabetes
April 24, 2019 - Brain areas linked to memory and emotion aid odor navigation in humans
April 24, 2019 - Brain stimulation reverses age-related memory loss
April 24, 2019 - Amid Opioid Prescriber Crackdown, Health Officials Reach Out To Pain Patients
April 24, 2019 - $4 million NIH award will help establish UCI Skin Biology Resource-based Center
April 24, 2019 - Cancer drugs reprogram genes in breast tumors to prevent endocrine resistance, finds study
April 24, 2019 - Combination-imaging technique provides new window into macaque brain connections
April 24, 2019 - Researchers identify new allergen responsible for allergy to durum wheat
April 24, 2019 - Researchers define role of rare, influential cells in the bone marrow
April 24, 2019 - DNA rearrangement may predict poor outcomes in multiple myeloma
April 24, 2019 - FDA Approves Skyrizi (risankizumab-rzaa) for Moderate to Severe Plaque Psoriasis
April 24, 2019 - Combination therapy might be beneficial in schizophrenia
April 24, 2019 - Blood test can help match cancer patients to early phase clinical trials
April 24, 2019 - Women tend to underreport snoring and underestimate its loudness
April 24, 2019 - Comprehensive molecular test introduced for diagnosis of malaria caused by P. vivax parasites
April 24, 2019 - New range prediction approach increases accuracy, safety and tolerability of proton therapy
April 24, 2019 - Need for Sedation Up for Regular Cannabis Users
April 24, 2019 - Lack of access to antibiotics is a major global health challenge
April 24, 2019 - New study provides better understanding on safety of deworming programs
April 24, 2019 - EEG used to detect impact of maternal stress on neurodevelopment in 2-month-old infants
April 24, 2019 - FDA Approves First Generic Naloxone Nasal Spray Against Opioid Overdose
April 24, 2019 - A new way of finding compounds that prevent aging
April 24, 2019 - Mechanical training makes synthetic hydrogels perform more like muscle
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