Machine learning cuts mammogram volume radiologists need to read

share Share
Over 40 million mammography exams are conducted in the United States every year, with the volume of images growing exponentially. Radiologists workloads further increase with the implementation of innovative screening technologies like digital breast tomosynthesis (DBT) that require more reading and testing time, increasing the pressure to deliver timely and accurate diagnosis. 

As the majority of mammograms a radiologist reviews are negative, a new study applies machine learning to accurately triage examinations found to be negative and refer any other findings to a breast imager, significantly reducing the day-to-day strain of radiologists workloads, helping to free up valuable time for them to focus on examinations that may present higher risk of suspicious findings and diagnostic workups. 

The goal of this study published in JACR was to determine whether machine learning could reduce the number of mammograms the radiologist must read by using a machine learning classifier to correctly identify normal mammograms and to select the uncertain and abnormal examinations for radiological interpretation.

Unlike related research using convolutional neural networks (CNNs)for mammography developed to completely automate and replace the radiologist, this research explores a more conservative approach. Specifically, the study explores a hybridized approach of mammography triage where some mammograms are autonomously diagnosed as being negative by a machine learning classifier and the remaining mammograms are read by a radiologist.

Mammograms in a research data set from over 7,000 women who were recalled for assessment at six UK National Health Service Breast Screening Program centres were used. A convolutional neural network in conjunction with multitask learning was used to extract imaging features from mammograms that mimic the radiological assessment provided by a radiologist, the patient’s non-imaging features, and pathology outcomes. A deep neural network was then used to concatenate and fuse multiple mammogram views to predict both a diagnosis and a recommendation of whether or not additional radiological assessment was needed.

Ten-fold cross-validation was used on 2,000 randomly selected patients from the data set; the remainder of the data set was used for convolutional neural network training. While maintaining an acceptable negative predictive value of 0.99, the proposed model was able to identify 34% (95% confidence interval, 25%-43%) and 91% (95% confidence interval: 88%-94%) of the negative mammograms for test sets with a cancer prevalence of 15% and 1%, respectively.

Machine learning was leveraged to successfully reduce the number of normal mammograms that radiologists need to read without degrading diagnostic accuracy.

Source: JACR 
Image credit: iStock

«« Dense breasts, breast cancer risk, and screening choices in 2019

After 5pm your doc may not screen you for cancer »»



Published on : Wed, 5 Jun 2019

Related Articles

Approximately 50% of women in the United States between the ages of 40 to 74 years old have dense breasts (an estimated 27.6... Read more

The new University of Arkansas for Medical Sciences (UAMS) radiology residents recently found themselves locked in a room,... Read more

The addition of breast MRI to mammography in screening breast cancer survivors for new cancers results in higher cancer detection... Read more

Mammography, Radiology, breast cancer, machine learning, deep learning, JACR As the majority of mammograms a radiologist reviews are negative, a new study applies machine learning to accurately triage examinations found to be negative and refer any other findings to a breast imager, significantly reducing the day-to-day strain of

One comment

michell kell

michell kell

(2020-09-29 14:08)
I just got my ex husband back through the help of Dr Nkosi love spells Experts. My husband left me for another woman, This was just 3 years of our marriage. The most painful thing is that I was pregnant with our second baby. I wanted him back. I did everything within my reach to bring him back but all was in vain, I wanted him back so badly because of the love I had for him, I begged him with everything, I made promises but he refused. I explained my problem to my friend and she suggested that I should rather contact a spell caster that could help me cast a spell to bring him back , I had no choice than to try it. I messaged the spell caster called dr nkosi, and he assured me there was no problem and that everything will be okay before 11 hours. He cast the spell and surprisingly 11 hours later my husband called me. I was so surprised, I answered the call and all he said was that he was so sorry for everything that had happened He wanted me to return to him. He also said he loved me so much. I was so happy and went to him that was how we started living together happily again.thanks to dr nkosi. if you are here and your Lover is turning you down, or your husband moved to another woman, do not cry anymore, contact Dr.Nkosi for help now..Here his contact, WhatsApp him: +27820502562 {or} Email him at: [email protected] ,his website:

Please login to leave a comment...

Highlighted Products