The Herman Trend Alert
April 26, 2017
AI Beats the Docs
The technology leader Google has done it again. Their Google AI (Artificial Intelligence) division has bested Board-certified pathologists in detecting breast cancer.
It began with developing an algorithm
Researchers took part in Google's Brain Residency Program---a 12-month educational course in machine and deep learning; the result was their creating an algorithm trained to detect breast cancer tumors in a dataset of digitized pathology slides, provided by the Dutch medical institute, Radboud University Medical Center.
An impressive result
The algorithm the researchers developed achieved a 92 percent sensitivity in identifying tumor cells from the slides. When you compare that score to the 73 percent achieved by the trained pathologists with no time limitation, you must be impressed by the technology. Moreover, the team matched that accuracy with different datasets from other hospitals and scanning machinery.
Not 100 percent positive
The team did report an average of eight false positives per slide, compared to none from the trained pathologists. However, researchers lowered this rate, with further work on the algorithm.
Replacing a boring and time-consuming job
In daily practice, pathology slide analysis is a time-consuming and laborious process, especially when you realize that each patient typically has 10 or more pathology slides for their suspected tumor. Even after completion, the pathology process does not always result in a definitive diagnosis.
Diagnosis in a shorter period of time
Due to this streamlined process, using machines to accurately analyze imagery for disease diagnosis provides a definitive decision in much less time. This reduction in time is particularly exciting to providers in the ongoing application of AI in healthcare.
Breast cancer is just the start
Google's own AI branch DeepMind Health is currently investigating using the technology as part of several collaborations with the UK National Health Service, including with Moorfields Eye Hospital, to improve the diagnosis of serious eye conditions. The Israeli company Zebra Medical, funded by Our Crowd, is solely dedicated to this AI application; the company is working on a wide range of different deep learning algorithms to help diagnose a range of different disorders, from spinal fractures to breast cancer. Researchers at Stanford University recently used to the technology to distinguish between benign and malignant skin growths.
A rosy future
We see a very positive future for the collaboration of medical diagnosis and AI. We imagine a future, when Dermatologist open a mobile app and scan a skin growth which will instantly be identified as cancerous, pre-cancerous, or benign.
Special thanks to Marco Ricci and Pharmaphorum for their coverage of this breakthrough.
© Copyright 1998-
by The Herman Group, Inc. -- reproduction for publication is encouraged, with the following attribution: From "The Herman Trend Alert," by Joyce Gioia, Strategic Business Futurist. 1.336.210.3548 or https://hermangroup.com. To sign up, visit https://HermanTrendAlert.com. The Herman Trend Alert is a trademark of The Herman Group, Inc."
HEAR JOYCE SPEAK TOMORROW AT 11:30 AM EDT
Tomorrow morning, Joyce will be a Fireside Chat Speakers at the Brightside Business Women Empowerment Virtual Conference where she will talk about how business women can use the insights from her new, groundbreaking book, EXPERIENCE RULES, to be successful. Register for FREE at https://brightsideglobaltrade.org/women.
GET COACHING FOR YOUR PEOPLE FOR LESS THAN $1/DAY!
Gain access to dozens of coaches for the same low price. Find out more by visiting https://ingomu.com/joyceg
JOYCE IS ON THE RADIO EVERYDAY NOW
To listen to Joyce's TV shows on the radio, simply visit RadioFutures.com. Joyce's show is streaming every day at 3:00PM Pacific Time.
To read this Herman Trend Alert on the web: https://hermangroup.com/alert/archive_3-29-2023.html.
Herman Trend Alerts are produced by the Herman Group, strategic business futurists, Certified Management Consultants, authors, and professional speakers.
New subscribers are always welcome. There is no charge for this public service. The Herman Trend Alert is read by over 30,000 people in 90 countries, including other websites and printed periodicals. Click here to sign up for the Herman Trend Alert.
Do you enjoy receiving this weekly e-mail update? Contact us about our co-branded Herman Trend Alert service.
Subscribe or Unsubscribe to weekly Herman Trend Alert
View this week's Herman Trend Alert
Archived Weekly Herman Trend Alerts
