Human-Plus-AI Imaging Analysis Boosts Cancer Diagnosis to 99.5%

share Share

A combination of the decades-long practice of diagnosing illnesses by manually reviewing images under a microscope and artificial intelligence (AI), can help doctors improve accuracy of breast and other cancer diagnosis to a rate of 99.5 percent.

A research team from Beth Israel Deaconess Medical Center (BIDMC) in Boston and Harvard Medical School (HMS) recently developed AI methods aimed at training computers to interpret pathology images, with the long-term goal of building AI-powered systems to make pathologic diagnoses more accurate.

“Our AI method is based on deep learning, a machine-learning algorithm used for a range of applications including speech recognition and image recognition,” explained pathologist Andrew Beck, MD, PhD, Director of Bioinformatics at the Cancer Research Institute at BIDMC and an Associate Professor at Harvard Medical School.

The Beck lab’s approach was recently put to the test in a competition held at the annual meeting of the International Symposium of Biomedical Imaging (ISBI), which involved examining images of lymph nodes to decide whether or not they contained breast cancer. The research team of Beck and his lab’s post-doctoral fellows placed first in two separate categories, competing against private companies and academic research institutions from around the world. The research team posted a technical report describing their approach to the repository, an open access archive of e-prints in physics, mathematics, computer science, quantitative biology, quantitative finance and statistics.

“Peering into the microscope to sift through millions of normal cells to identify just a few malignant cells can prove extremely laborious using conventional methods. We thought this was a task that the computer could be quite good at – and that proved to be the case,” said Beck.

In an objective evaluation in which researchers were given slides of lymph node cells and asked to determine whether or not they contained cancer, the team’s automated diagnostic method proved accurate approximately 92 percent of the time, that nearly matched the success rate of a human pathologist, whose results were 96 percent accurate.

“But when we combined the pathologist’s analysis with our automated computational diagnostic method, the result improved to 99.5 percent accuracy,” said Beck. “Combining these two methods yielded a major reduction in errors.”

The team trained the computer to distinguish between cancerous tumour regions and normal regions based on a deep multilayer convolutional network.

“There have been many reasons to think that digitising images and using machine learning could help pathologists be faster, more accurate and make more accurate diagnoses for patients,” Beck added.

Jeroen van der Laak, PhD, who leads a digital pathology research group at Radboud University Medical Center in the Netherlands and was an organiser for the competition, said, “when we started this challenge, we expected some interesting results. The fact that computers had almost comparable performance to humans is way beyond what I had anticipated. It is a clear indication that artificial intelligence is going to shape the way we deal with histopathological images in the years to come.”

«« PET Tomo Helps Hodgkin Lymphoma Patients Avoid Chemo

3D Breast Imaging Tool Boosts Contrast, Cuts Dose Levels »»

Published on : Tue, 21 Jun 2016

Related Articles
Peter Johnson, Professor of Medical Oncology at the University of Southampton

Research on 1,200 patients in the UK, European and Australia has found that the use of positron emission tomography (PET)... Read more

This image shows DCIS sample highlighting aggressive biomarkers. Credit: University of Michigan Health System

A combination of imaging and mathematics can help identify if aggressive treatment is needed when a woman has been diagnosed... Read more

The left image is a tissue sample with low levels of HER2 and HER3 bonded together. The right is a sample with high levels

Scientists in the UK have used an imaging technique, that can accurately measure the distance between two protein molecules,... Read more

Human-Plus-AI Imaging Analysis Boosts Cancer Diagnosis to 99.5%, artificial intelligence, machine learning Human-Plus-AI Imaging Analysis Boosts Cancer Diagnosis to 99.5%

No comment

Please login to leave a comment...

Highlighted Products