AUG 13, 2020 8:43 AM PDT

A Call for Bigger, Better Studies on AI Breast Cancer Diagnostics

WRITTEN BY: Tara Fernandez

A staggering 1 in 8 women in the U.S. will develop an aggressive form of breast cancer in their lifetime, making it the most commonly diagnosed cancer among American women. Experts agree that early intervention through health monitoring has a massive impact on clinical outcomes.

Constance Lehman is the director of Breast Imaging at Massachusetts General Hospital. According to Lehman, the key to improving patient care lies in advanced medical diagnostic technologies such as artificial intelligence, or AI.

In a recent publication in JAMA Oncology, Lehman pushed for more AI-driven tools to support mammography interpretation, with evaluating these new technologies in the clinic under simulated trials as being the first step. Mammograms are X-ray images of the breast, taken using low doses of radiation. These help clinicians spot potentially malignant tissues within the breast.

Recent studies have reported that diagnostic advances are completely transforming the way traditional mammograms are processed in the clinic: they’re now completely digital, much faster, and can incorporate AI to search for signature tissue changes associated with potential tumors, increasing accuracy and reducing the workload burden on staff.

"In the continued evolution of AI applied to improving human health, it is time to move beyond simulation and reader studies and enter the critical phase of rigorous, prospective clinical evaluation," wrote Lehman as part of her commentary on these studies.

So far, the challenge with AI has been in “teaching” the system on how to differentiate between healthy and suspicious tissues. This learning process hinges on the quality and quantity of the images fed into the AI platform for it to learn. Because not all developers have access to large, standardized databases, performance results so far have been mixed. Future development is focused on using large databases of mammogram images featuring both normal controls and those from individuals diagnosed with breast cancer. 

What’s next? Experts like Lehman are calling for testing and validation of these diagnostic algorithms in clinical trials before they can be implemented as part of standard breast health monitoring regimes. 

"If AI models can be developed that can reliably detect on mammograms women with cancer from those without it, then quality, affordable screening mammography may finally become available to a large population of women globally who currently have no access to its life-saving potential."

 

 

Sources: Massachusets General Hospital, Medical XPress.


 

About the Author
  • Tara Fernandez has a PhD in Cell Biology and has spent over a decade uncovering the molecular basis of diseases ranging from skin cancer to obesity and diabetes. She currently works on developing and marketing disruptive new technologies in the biotechnology industry. Her areas of interest include innovation in molecular diagnostics, cell therapies, and immunology. She actively participates in various science communication and public engagement initiatives to promote STEM in the community.
You May Also Like
JUN 09, 2020
Clinical & Molecular DX
The Key to Unlocking Next Generation Wearable Biosensors Is Under the Sea
JUN 09, 2020
The Key to Unlocking Next Generation Wearable Biosensors Is Under the Sea
Wearable biosensors are the latest trend in health and diagnostic technologies — keeping track of everything from ...
JUL 11, 2020
Cardiology
Predicting a Heart Attack's Prognosis with Biomarkers
JUL 11, 2020
Predicting a Heart Attack's Prognosis with Biomarkers
Biomarker studies have taken the diagnostic field by storm these past few decades. Many diseases will have unique g ...
JUL 13, 2020
Neuroscience
Can COVID-19 Be Diagnosed by Vocal Patterns?
JUL 13, 2020
Can COVID-19 Be Diagnosed by Vocal Patterns?
Researchers from the Lincoln Laboratory have found early evidence that subtle changes in peoples' voices may indicat ...
JUL 27, 2020
Cancer
Identifying Tumor Cells in the Blood with Artificial Intelligence
JUL 27, 2020
Identifying Tumor Cells in the Blood with Artificial Intelligence
Cancer diagnostics is an area of keen interest in the fight against cancer. Cancer often slips by routine checkups and v ...
AUG 12, 2020
Cancer
Comparing Nivolumab and Pembrolizumab in the Treatment of Lung Cancer
AUG 12, 2020
Comparing Nivolumab and Pembrolizumab in the Treatment of Lung Cancer
Since the 1940s, chemotherapy has been a primary treatment option for cancer. The late 20th century brought a new type o ...
SEP 12, 2020
Cardiology
Is Hyperuricemia an Indicator of Aortic Disease?
SEP 12, 2020
Is Hyperuricemia an Indicator of Aortic Disease?
Living in a culture of excess can have its benefits, but also its drawbacks. Diabetes is a common disease that is gettin ...
Loading Comments...