By Stephen Beech via SWNS
Cancer could soon be detected by artificial intelligence – “dramatically” improving patients’ survival chances, according to a new study.
Doctors will be able to use state-of-the-art machine learning technology to diagnose cancer in patients which would lead to earlier treatment, say scientists.
A prototype developed by British scientists identified 13 different types of cancer with an accuracy rate above 98%.
Cancer causes the death of 10 million people around the world with its evolutionary nature making it tough to treat late-stage tumors.
Scientists explained that genetic information is encoded in DNA by patterns of the four bases – denoted by A, T, G and C – that make up its structure.
Environmental changes outside the cell can cause some DNA bases to be modified by adding a methyl group, a process known as “DNA methylation.”
Each individual cell possesses millions of DNA methylation marks.
Researchers who have observed changes to those marks in the early stages of cancer believe they could assist in early diagnosis.
They say it’s possible to examine which bases in DNA are methylated in cancers and to what extent, compared to healthy tissue.
But scientists described identifying the specific DNA methylation signatures indicative of different cancer types is “akin to searching for a needle in a haystack”.
However, the team that conducted the new study, published in the journal Biology Methods & Protocols, believes that AI can help.
Cambridge University and Imperial College London researchers trained an AI mode, using a combination of machine and deep learning, to look at the DNA methylation patterns.
It identified 13 different types of cancer – including breast, liver, lung, and prostate cancer – from non-cancerous tissue with 98.2 percent accuracy.
The model relies on tissue samples and would need additional training and testing on a more diverse collection of biopsy samples to be ready for clinical use.
Researchers believe that an important aspect of the study was the use of an “explainable and interpretable” core AI model, which provided insights into the reasoning behind its predictions.
The team explored the inner workings of their model and showed that it “reinforces and enhances” understanding of the underlying processes contributing to cancer.
They said that spotting the unusual methylation patterns – potentially from biopsies – would allow doctors to detect cancer earlier.
The team believe the system could potentially improve patient outcomes “dramatically” as most cancers are treatable or curable if detected early enough.
Study lead author Dr. Shamith Samarajiwa said: “Computational methods such as this model, through better training on more varied data and rigorous testing in the clinic, will eventually provide AI models that can help doctors with early detection and screening of cancers.”
Dr. Samarajiwa, of Cambridge University, added: “This will provide better patient outcomes.”
Cancer could soon be detected by artificial intelligence – “dramatically” improving patients’ survival chances, according to a new study.
Doctors will be able to use state-of-the-art machine learning technology to diagnose cancer in patients which would lead to earlier treatment, say scientists.
A prototype developed by British boffins identified 13 different types of cancer with an accuracy rate above 98%.
Cancer causes the death of 10 million people around the world with its evolutionary nature making it tough to treat late-stage tumors.
Scientists explained that genetic information is encoded in DNA by patterns of the four bases – denoted by A, T, G and C – that make up its structure.
Environmental changes outside the cell can cause some DNA bases to be modified by adding a methyl group, a process known as “DNA methylation.”
Each individual cell possesses millions of DNA methylation marks.
Researchers who have observed changes to those marks in the early stages of cancer believe they could assist in early diagnosis.
They say it’s possible to examine which bases in DNA are methylated in cancers and to what extent, compared to healthy tissue.
But scientists described identifying the specific DNA methylation signatures indicative of different cancer types as “akin to searching for a needle in a haystack”.
However, the team that conducted the new study, published in the journal Biology Methods & Protocols, believes that AI can help.
Cambridge University and Imperial College London researchers trained an AI mode, using a combination of machine and deep learning, to look at the DNA methylation patterns.
It identified 13 different types of cancer – including breast, liver, lung, and prostate cancer – from non-cancerous tissue with 98.2 percent accuracy.
The model relies on tissue samples and would need additional training and testing on a more diverse collection of biopsy samples to be ready for clinical use.
Researchers believe that an important aspect of the study was the use of an “explainable and interpretable” core AI model, which provided insights into the reasoning behind its predictions.
The team explored the inner workings of their model and showed that it “reinforces and enhances” understanding of the underlying processes contributing to cancer.
They said that spotting the unusual methylation patterns – potentially from biopsies – would allow doctors to detect cancer earlier.
The team believe the system could potentially improve patient outcomes “dramatically” as most cancers are treatable or curable if detected early enough.
Study lead author Dr. Shamith Samarajiwa said: “Computational methods such as this model, through better training on more varied data and rigorous testing in the clinic, will eventually provide AI models that can help doctors with early detection and screening of cancers.”
Dr. Samarajiwa, of Cambridge University, added: “This will provide better patient outcomes.”
