Published: Thu, May 23, 2019
Electronics | By Kelly Massey

AI spotted lung cancer BETTER than expert radiologists, study finds

AI spotted lung cancer BETTER than expert radiologists, study finds

Lung cancer screening has garnered much more attention recently as various studies have been published in the a year ago on the outcomes of screening guidelines and programs. In this study, the deep learning model was used to foresee whether a patient has lung cancer and also find the place of the malignant tissue in the patient's lungs.

The work demonstrates the potential for Artificial Intelligence (AI) to increase both accuracy and consistency, which could help accelerate adoption of lung cancer screening worldwide.

As of now, lung cancer claims to be having more than $1.7 Million death every year and is also considered to be the deadliest of all cancers.

Unfortunately, the US Centers for Disease Control reports that less five percent of patients who met the screening criteria actually got screened.

Like several other deep learning algorithms tested for use in medical fields, the computer brain in this study was trained using scans from past lung cancer screenings. The estimates show that low-dose CT (LDCT), a cost-effective method of screening lung cancer has reduced lung cancer deaths by 20%. CEO Sundai Pichai tweeted "today we're publishing our work in @NatureMedicine showing how these methods could boost chances of survival for many people at risk around the world".

However, when the researchers provided an earlier scan, the results were a lot closer, both for the deep learning program and the human doctors.

Town hall on enemy turf? Fox News debate divides 2020 Dems
Elizabeth Warren turned down a similar chance while branding the network a "hate-for-profit racket". It's important for Democrats to "find people where they are", he said.


Next, the researchers performed a two-part retrospective reader study with the participation of six US board-certified radiologists with an average of eight years of clinical experience.

They continue and explain in the cases where the additional tomography data was not available, "the AI handily bested its human counterparts, outperforming all six of them with 11 percent fewer false positives and 5 percent fewer false negatives". "This is technically "4D" because it is not only looking at one CT scan, but two over time", Etemadi said. Warren has written for many publications including the New York Daily News, Vanity Fair and Yahoo.

The research team has created an artificial intelligence model that makes the prediction of an overall lung cancer malignancy in 3D volume and also assists in identifying malignancy tissue in the lungs. These errors typically delay the diagnosis of lung cancer until the disease has reached an advanced stage when it becomes too hard to treat. The AI-powered model uses a current CT scan and if possible, a previous CT scan, as input for each patient.

"The system can categorize a lesion with more specificity", said co-author Dr. Mozziyar Etemadi of Northwestern University in a statement.

"Most of the software we use as clinicians is designed for patient care, not for research", Etemadi said.

Like this: