Better Statistical Analysis May Be Key to Earlier Mesothelioma Diagnosis

Published on April 03, 2019

Page Last Modified: September 22, 2019

One of the top goals of mesothelioma researchers is to find a way to diagnose malignant mesothelioma earlier. The rare and fatal form of cancer typically does not get diagnosed until it is in its latest stages, and that makes effective treatment virtually impossible. While scientists have spent countless hours trying to devise more accurate an innovate tests, molecular geneticists at the University of Toronto pursued a different tactic, trying to see whether by inputting the correct data into sophisticated computer programs, machines could do a better job of spotting the fatal disease in its earliest stages.

Researchers Say Machine Learning is A Valuable Tool

In conducting their mesothelioma researcher, the group of geneticists collaborated with both oncologists and cardiologists to collect data from patients who were suspected of having the rare, asbestos-related disease. They input information from hundreds of cases into several different programs to determine whether they were of help, and which were most accurate at both diagnosing and ruling out malignant pleural mesothelioma, the most common form of the disease. What they found was that the data itself was almost as important as the program that analyzed it, and that by paying attention to the right factors physicians might not even need artificial intelligence.

What is Artificial Intelligence and How Does it Help with Diagnosis?

Artificial intelligence is the use of computers that take all of the accumulated information that has been programmed into them in order to allow faster deduction and calculation than can be done manually. It is also known as machine learning, and there are a variety of forms and programs available to quickly analyze immense amounts of data. The researchers input a variety of health record information into several of these programs from a group of over 300 Turkish patients with known exposure to asbestos and who were showing a wide array of symptoms suggestive of malignant mesothelioma. They were able to identify one of the programs as most effective at early and accurate diagnosis.

According to lead researcher Davide Chicco, “Our results show that machine learning can predict diagnoses of patients having mesothelioma symptoms with high accuracy, sensitivity, and specificity in a few minutes.” Though the turnaround of information from the program known as Random Forest was remarkably fast and accurate, it also pointed to specific diagnostic factors that seemed most important to spotting the disease early. Those factors were lung side and platelet count, and based upon the results delivered by the program, the researchers believe that diagnosticians who do not have access to artificial intelligence can still use those elements in making an accurate diagnosis.

The more scientists learn about the way that mesothelioma manifests and how to diagnose it quickly, the closer we get to finding an effective treatment for this rare and fatal asbestos-related disease. If you or someone you love needs information about treatment options or the other resources available to you, contact the Patient Advocates at Mesothelioma.net today to learn more. We can be reached at 1-800-692-8608.

Written by Terri Oppenheimer

Terri Heimann Oppenheimer is the head writer of our Mesothelioma.net news blog. She graduated from the College of William and Mary with a degree in English. Terri believes that knowledge is power and she is committed to sharing news about the impact of mesothelioma, the latest research and medical breakthroughs, and victims’ stories.

Learn more about and contact Terri

Where can I Get Additional Help?

For over 20 years we’ve provided the best FREE resources to mesothelioma patients and loved ones. Our resources include information on the leading treatment options and best doctors in your area; lessons learned from survivors; claims and benefits specifically for Veterans; and how to access your share of billions of dollars in trust fund money.