Google is Diagnosing Lung Cancer With Artificial Intelligence Better than Humans can
Better and Earlier than Radiologists
Artificial intelligence is officially joining the fight against cancer. AI is set to especially change early diagnosis, it would seem, for some of the world’s biggest killers. Lung cancer is a case in point.
A new study found that Google artificial intelligence did better than six radiologists at identifying whether certain patients had lung cancer. When examining a single scan, it diagnosed 5% more cancers and also cut the false positives by 11%.
Deep Learning — 1, Lung Cancer — 0
Deep learning — a form of artificial intelligence — was able to detect malignant lung nodules on low-dose chest computed tomography (LDCT) scans with a performance meeting or exceeding that of expert radiologists, reports a new study from Google and Northwestern Medicine.
While Google says the technology is aimed at aiding — not replacing — radiologists, it’s clear that AI and deep learning and machine intelligence broadly speaking will indeed replace a lot of humans at their jobs, from industries like transportation, finance, retail and yes, to healthcare. To suppose AI only augments human-AI hybrid performance is, to me as a futurist, somewhat deceptive.
- In order to test the AI, Google showed it 45,856 chest CT scans, comparing the AI’s diagnoses with those of six board-certified radiologists.
- Google says its AI can spot early-stage lung cancer, in some cases better than doctors can.
- This is a promising step in how we predict lung cancer. Read Google’s report here.
Like all forms of the disease, an early diagnosis of lung cancer can greatly improve a patient’s chance of survival, but for all cancers, this is much easier said than done. That deep learning can improve upon this is really great for healthcare all over the world.
AI will Radically Change Healthcare and Early Diagnosis
This technology demonstrates just how important AI will become in healthcare, so read the following carefully.
Radiologists typically have to view up to hundreds of images from a single CT scan in order to make a successful diagnosis for lung cancer; Google’s AI is able to generate a three-dimensional image instead of 2-D ones, as well as detect specific areas of malignant tissue in the lungs, which radiologists are often unable to do from images alone.
The study published in Nature Medicine reported that the algorithm, trained on 42,000 patient CT scans taken during a National Institutes of Health clinical trial, outperformed six radiologists in determining whether patients had cancer.
The story first broken on May 20th, 2019 but there had been hints that Google had achieved this for a while. Google essentially unveiled an artificial intelligence system that — in early testing — demonstrates a remarkable talent for seeing through lung cancer’s disguises.
AI Doesn’t Just Augment Doctors, it Improves Upon Them
It’s important to note that this system performs better and differently than humans and white collar doctors can. As Google, Amazon, Apple and others take on healthcare, they will augment the current system even as global population aging becomes a bigger deal in our lives and on the global economy, with ballooning healthcare costs.
We cannot assume that healthcare professionals will be free from disruption either. Deep learning has a surprisingly ability to change the roles of even white collar professionals. We are seeing this with surgery, diagnosis and even in law and legal processes.
Early Diagnosis is Key for Lung Cancer Survival Rates
Think about it: lung cancer is by far the most common killer of Americans among cancers, resulting in about 160,000 deaths in 2018. Artificial Intelligence is transforming the world, mostly for the better.
Meanwhile, about 50 percent of lung cancers are diagnosed after they have already spread, when the disease is harder to effectively treat. How AI is regulated and goes mainstream in healthcare must be a real debate, since these BigTech companies will increasingly become utilities and their technologies will have access to our healthcare data.
Experts who were not involved in the study said Google’s system could make screening much more viable. The research was published online March 20 in Nature Medicine. Google’s technology will have to undergo further research and testing before becoming available as a diagnostic tool, but the company hopes to “make early detection more accessible.”