Artificial Intelligence Program to aid cancer Pathology
By Warren Miller. 24th September 2018
Doctors usually rely on the well trained eyes of pathologists to give their patients a cancer diagnosis. Now researchers are teaching machines to do that time intensive work in as little as a few seconds.
A new computer program can analyze images of patients' lung tumours, specify cancer types, and even identify altered genes driving abnormal cell growth, a new study shows. This has the potential to speed up the diagnosis of other cancers, including malignant mesothelioma and asbestos lung cancer.
An earlier diagnosis of mesothelioma can improve the odds of survival by making earlier intervention possible. When mesothelioma is at its earliest stage it is most likely to respond to treatment.
The image above demonstrates shows how the AI program analyses cancerous tissue to create a map that tells apart two lung cancer types. The squamous cell carcinoma is shown in red, lung squamous cell carcinoma in blue, and normal lung tissue in grey.
Researchers at New York University (NYU) School of Medicine found that an artificial intelligence program could distinguish with 97% accuracy between Adenocarcinoma and Squamous Cell Carcinoma, two lung cancer types that experienced pathologists at times struggle to parse without confirmatory tests.
The computer program was also able to detect whether abnormal versions of six genes linked to lung cancer were present in cells. These genetic mutations often cause the abnormal growth seen in cancer, so determining which genes have changed can help with targeted therapies that work only against cancer cells with specific genetic mutations.
An associate professor in the Department of Pathology at NYU School of Medicine stated that:
" Delaying the start of cancer treatment is never good. Our study provides strong evidence that an AI approach will be able to instantly determine cancer subtype and mutational profile to get patients started on targeted therapies sooner. "
The Artificial intelligence comes from the ability of the program to build rules and mathematical models to enable it to make decisions based on the data fed into it. The team at NYU used images from The Cancer Genome Atlas, where lung cancer diagnosis had already been determined to measure how well the AI program could be trained to accurately and automatically identify cancerous tissues from normal tissues.
A few of the tumour images were incorrectly classified by the computer program, but about half of these were also
incorrectly classified by the pathologists. Conversely, 45 of 54 of images wrongly classified by the pathologists were correctly
classified by the computer program. So at the very least the program offers a useful second opinion.
In the near future the team hope to keep training the AI program with more data until it can determine which genes are
mutated in a given cancer with more than 90 percent accuracy. This has the potential to speed up the diagnosis of mesothelioma and asbestos lung cancer.
If you have any questions about the article or would like more information about making a claim for Mesothelioma or Asbestos Lung Cancer, please call us on the number below or use the online chat facility.