Forum for Science, Industry and Business

Sponsored by:     3M 
Search our Site:

 

Computer accurately identifies and delineates breast cancers on digital tissue slides

11.05.2017

A deep-learning computer network developed through research led by Case Western Reserve University was 100 percent accurate in determining whether invasive forms of breast cancer were present in whole biopsy slides.

Looking closer, the network correctly made the same determination in each individual pixel of the slide 97 percent of the time, rendering near-exact delineations of the tumors.


This is a tumor boundary delineated by a pathologist.

Credit: Anant Madabhushhi

Compared to the analyses of four pathologists, the machine was more consistent and accurate, in many cases improving on their delineations.

In a field where time and accuracy can be critical to a patient's long-term prognosis, the study is a step toward automating part of biopsy analysis and improving the efficiency of the process, the researchers say.

Currently, cancer is present in one in 10 biopsies ordered by physicians, but all must be analyzed by pathologists to identify the extent and volume of the disease, determine if it has spread and whether the patient has an aggressive or indolent cancer and needs chemotherapy or a less drastic treatment.

Last month, the U.S. Food and Drug Administration approved software that allows pathologists to review biopsy slides digitally to make diagnosis, rather than viewing the tissue under a microscope.

"If the network can tell which patients have cancer and which do not, this technology can serve as triage for the pathologist, freeing their time to concentrate on the cancer patients," said Anant Madabushi, F. Alex Nason professor II of biomedical engineering at Case Western Reserve and co-author of the study detailing the network approach, published in Scientific Reports.

The study

To train the deep-learning network, the researchers downloaded 400 biopsy images from multiple hospitals. Each slide was approximately 50,000 x 50,000 pixels. The computer navigated through or rectified the inconsistencies of different scanners, staining processes and protocols used by each site, to identify features in cancer versus the rest of the tissue.

The researchers then presented the network with 200 images from The Cancer Genome Atlas and University Hospitals Cleveland Medical Center. The network scored 100 percent on determining the presence or absence of cancer on whole slides and nearly as high per pixel.

"The network was really good at identifying the cancers, but it will take time to get up to 20 years of practice and training of a pathologist to identify complex cases and mimics, such as adenosis," said Madabhushi, who also directs the Center of Computational Imaging and Personalized Diagnostics at Case Western Reserve.

Network training took about two weeks, and identifying the presence and exact location of cancer in the 200 slides took about 20 to 25 minutes each.

That was done two years ago. Madabhushi suspects training now -- with new computer architecture -- would take less than a day, and cancer identification and delineation could be done in less than a minute per slide.

"To put this in perspective," Madabhushi said, "the machine could do the analysis during 'off hours,' possibly running the analysis during the night and providing the results ready for review by the pathologist when she/he were to come into the office in the morning."

###

Madabhushi worked with Angel Cruz-Roa, a PhD student, and Fabio Gonzalez, professor, Department of Systems and Industrial Engineering at the Universidad Nacional de Colombia, in Bogota; Hannah Gilmore, associate professor of pathology at Case Western Reserve School of Medicine; Ajay Basavanhally of Inspirata Inc., Tampa Fla.; Michael Feldman, professor of pathology and laboratory medicine, and Natalie Shi, of the Department of Pathology, at the Hospital of the University of Pennsylvania; Shridar Ganesan, associate professor of medicine and pharmacology at the Rutgers Cancer Institute of New Jersey; and John Tomaszewski, chair of pathology and anatomical services at the University of Buffalo, State University of New York.

Much of the study was built on research by Madabhushi and Andrew Janowczyk, a biomedical engineering Postdoctoral Fellow at Case Western Reserve. They led development of what they termed "a resolution adaptive deep hierarchical learning scheme," which can cut the time for image analysis using deep learning approaches by 85 percent.

Deep-learning networks learned to identify indicators of cancer at lower resolutions to determine where further analysis at high levels of magnification, and thus greater computation time, were necessary to provide precise results. In short, the scheme eliminated time-consuming, high-resolution analysis of healthy tissue.

To manage the variance in staining of digitized biopsy images that can confound computer analysis, the researchers developed a technique called Stain Normalization using Sparse AutoEncoders. The technique partitions images into tissue sub-types so color standardization for each can be performed independently.

To speed research in the field, Janowczyk and Madabhushi also published a tutorial on deep learning for digital pathology image analysis. The paper was recently awarded the most cited paper award from the Journal of Pathology Informatics.

Media Contact

William Lubinger
william.lubinger@case.edu
216-368-4443

 @cwru

http://www.case.edu 

Deep-learning network possible step toward automating biopsy slide analysis | EurekAlert!

More articles from Medical Engineering:

nachricht Novel chip-based gene expression tool analyzes RNA quickly and accurately
18.01.2018 | University of Illinois College of Engineering

nachricht Potentially life-saving health monitor technology designed by Sussex University physicists
10.01.2018 | University of Sussex

All articles from Medical Engineering >>>

The most recent press releases about innovation >>>

Die letzten 5 Focus-News des innovations-reports im Überblick:

Im Focus: Artificial agent designs quantum experiments

On the way to an intelligent laboratory, physicists from Innsbruck and Vienna present an artificial agent that autonomously designs quantum experiments. In initial experiments, the system has independently (re)discovered experimental techniques that are nowadays standard in modern quantum optical laboratories. This shows how machines could play a more creative role in research in the future.

We carry smartphones in our pockets, the streets are dotted with semi-autonomous cars, but in the research laboratory experiments are still being designed by...

Im Focus: Scientists decipher key principle behind reaction of metalloenzymes

So-called pre-distorted states accelerate photochemical reactions too

What enables electrons to be transferred swiftly, for example during photosynthesis? An interdisciplinary team of researchers has worked out the details of how...

Im Focus: The first precise measurement of a single molecule's effective charge

For the first time, scientists have precisely measured the effective electrical charge of a single molecule in solution. This fundamental insight of an SNSF Professor could also pave the way for future medical diagnostics.

Electrical charge is one of the key properties that allows molecules to interact. Life itself depends on this phenomenon: many biological processes involve...

Im Focus: Paradigm shift in Paris: Encouraging an holistic view of laser machining

At the JEC World Composite Show in Paris in March 2018, the Fraunhofer Institute for Laser Technology ILT will be focusing on the latest trends and innovations in laser machining of composites. Among other things, researchers at the booth shared with the Aachen Center for Integrative Lightweight Production (AZL) will demonstrate how lasers can be used for joining, structuring, cutting and drilling composite materials.

No other industry has attracted as much public attention to composite materials as the automotive industry, which along with the aerospace industry is a driver...

Im Focus: Room-temperature multiferroic thin films and their properties

Scientists at Tokyo Institute of Technology (Tokyo Tech) and Tohoku University have developed high-quality GFO epitaxial films and systematically investigated their ferroelectric and ferromagnetic properties. They also demonstrated the room-temperature magnetocapacitance effects of these GFO thin films.

Multiferroic materials show magnetically driven ferroelectricity. They are attracting increasing attention because of their fascinating properties such as...

All Focus news of the innovation-report >>>

Anzeige

Anzeige

Event News

10th International Symposium: “Advanced Battery Power – Kraftwerk Batterie” Münster, 10-11 April 2018

08.01.2018 | Event News

See, understand and experience the work of the future

11.12.2017 | Event News

Innovative strategies to tackle parasitic worms

08.12.2017 | Event News

 
Latest News

Let the good tubes roll

19.01.2018 | Materials Sciences

How cancer metastasis happens: Researchers reveal a key mechanism

19.01.2018 | Health and Medicine

Meteoritic stardust unlocks timing of supernova dust formation

19.01.2018 | Physics and Astronomy

VideoLinks
B2B-VideoLinks
More VideoLinks >>>