Earlier Mesothelioma Diagnosis with Machine Learning Model

A new machine learning model can diagnose mesothelioma better than any current model. The model provides new hope for early diagnosis and better treatment of malignant mesothelioma.

A machine learning tool is a computer program that gets “smarter” the more it runs. One challenge in predicting mesothelioma is that mesothelioma is very rare. With the exception of mesothelioma specialists, most doctors never even see a single case. The same is true for pathologists. If they do not see mesothelioma patients, they are less likely to see differences that may influence the diagnosis.

Doctors trained the new machine learning tool to recognize and diagnose malignant mesothelioma. It works more quickly and accurately. Earlier mesothelioma diagnosis would allow for earlier intervention. This could mean longer survival for victims of one of the world’s deadliest cancers.

An Easier and Less Painful Diagnosis

Mesothelioma is a rare cancer caused by asbestos exposure. It can take decades to develop, but it is usually fatal within 18 months of diagnosis.

One reason is that earlier mesothelioma diagnosis is rare. Early malignant mesothelioma is difficult to detect using standard imaging tools. By the time a mesothelioma tumor is large enough to be seen on a CT scan, it may already be very advanced. Late-stage pleural mesothelioma is resistant to most kinds of cancer treatments.

“With an increase in incidence and a continuous lack of non-invasive diagnosis methods, a machine learning model has been proposed for the effective diagnosis of malignant mesothelioma,” says Shakir Shabbir from the University of Engineering and Technology in Pakistan.

The study team emphasized that this new machine learning model is non-invasive. No surgery or tissue sample is required. Current diagnostic methods are rather painful for the patient. This may include a biopsy or minor surgery. And they are often expensive and unavailable in rural areas.

The machine learning model uses social, economic, geographic, and clinical data from the patient file. The study used the health records of 324 Turkish patients with symptoms related to mesothelioma.

Source:

Shabbir, Shakir, Muhammad S. Asif, Talha Mahboob Alam, and Zeeshan Ramzan. “Early prediction of malignant mesothelioma: an approach towards non-invasive method.” Current Bioinformatics 16, no. 10 (2021): 1257-1277. https://doi.org/10.2174/1574893616666210616121023