eHealth News, South Africa

Google Trains Computers to Diagnose Cancer

Google researchers have developed an automated detection algorithm that is effective at identifying breast cancer.

Google - EHN

Google researchers have developed an automated detection algorithm that is effective at identifying breast cancer and can naturally complement a pathologists’ workflow.

The tech giant used deep learning, a form of machine learning that can be applied to large datasets to help recognise patterns and images, to create an algorithm which goes through the pathology data faster than a human.

“A pathologist’s report after reviewing a patient’s biological tissue samples is often the gold standard in the diagnosis of many diseases. For cancer in particular, a pathologist’s diagnosis has a profound impact on a patient’s therapy,” wrote Technical Lead, Martin Stumpe, and Product Manager at Google, Lily Peng, in a blog post.

According to the team, the reviewing of pathology data is a complex task, requiring years of training to gain the expertise and experience to do well.

However, the researchers note that even with this extensive training, there can be substantial variability in the diagnoses given by different pathologists for the same patient, which can lead to misdiagnoses. For example, agreement in diagnosis for some forms of breast cancer can be as low as 48%, and similarly low for prostate cancer.

“The lack of agreement is not surprising given the massive amount of information that must be reviewed in order to make an accurate diagnosis,” said the Google team.

To address these issues of limited time and diagnostic variability, the research team is investigating how deep learning can be applied to digital pathology by creating an automated detection algorithm that can naturally complement pathologists’ workflow.

The Google team used a large dataset of images supplied by the Radboud University Medical Centre and University Medical Centre Utrecht in the Netherlands to train the algorithms to automatically recognise signs that an individual’s breast cancer may have spread or metastasised to nearby lymph nodes.

“After additional customisation we showed that it was possible to train a model that either matched or exceeded the performance of a pathologist who had unlimited time to examine the slides,” said the researchers.

The researcher showed that the algorithm was good at identifying potential cancer but also identified a lot of false positives. According to the team, these algorithms perform well for the tasks for which they are trained, but lack the breadth of knowledge and experience of human pathologists.

“To ensure the best clinical outcome for patients, these algorithms need to be incorporated in a way that complements the pathologist’s workflow. We envision that an algorithm such as ours could improve the efficiency and consistency of pathologists,” concluded the team.

Google says the research will not be incorporated into a real product yet as it will still have to go through clinical validation and regulatory approval. However, it hopes by sharing the work it will accelerate progress in the space.

For more information contact news@eHealthNews.co.za, like us on Facebook or tweet us @eHealthNewsZA.

Did you find this eHealth article interesting or valuable? TWEET THE ARTICLE  



Subscribe now to ehealthnews.co.za and get the news as it happens