Biology Methods & Protocols: Artificial intelligence will soon be able to detect cancer

A new paper published in Biology Methods & Protocols reports that doctors may soon be able to use artificial intelligence (AI) to detect and diagnose cancer in patients, allowing treatment to begin earlier. Cancer remains one of the most complex human diseases, with more than 19 million cases and 10 million deaths per year. The evolutionary nature of cancer makes it difficult to treat tumors in advanced stages.

Genetic information is encoded in DNA using patterns of four bases, labeled A, T, G, and C, that make up its structure. Changes in the environment outside the cell can cause some DNA bases to be modified by the addition of a methyl group. This process is called "DNA methylation." Each individual cell contains millions of these DNA methylation tags. Researchers have observed changes in these tags in the early stages of cancer development: they may help in the early diagnosis of cancer.

It is possible to study which bases in DNA are methylated in cancer and to what extent, compared to healthy tissues. Identifying specific features of DNA methylation that indicate different types of cancer is akin to finding a needle in a haystack. Researchers involved in this study believe that artificial intelligence can help do just that.

Researchers from the University of Cambridge and Imperial College London have trained artificial intelligence, using a combination of machine and deep learning, to learn DNA methylation patterns and identify 13 different types of cancer (including breast, liver, lung and prostate) from non-cancerous tissue with 98.2% accuracy. This model is based on tissue samples (rather than DNA fragments in blood) and requires additional training and testing on a more diverse collection of biopsy specimens to be ready for clinical use.

Follow NEWS.am Medicine on Facebook and Twitter