AI predicts accurate detection based on tumor markers and the nature of the disease, along with the patient’s genetic conditions.
Cancer is one of the most physically and mentally debilitating diseases, and its treatment depends largely on the timing of diagnosis. According to experts, detecting cancer at an early stage not only saves lives but also reduces the exorbitant costs of treatment. Every year, 8.8 million people die from cancer, most of them in low- and middle-income countries, according to WHO figures. The number is so high that it represents two and a half times more people who die than those who die from HIV/AIDS, tuberculosis and malaria combined.
However, new technologies are making life easier for everyone. The use of artificial intelligence (AI) is extremely important in expanding many areas of healthcare services, especially cancer detection and treatment. Experts say that researchers are using machine learning to create tools in the diagnostic realm, which could detect tumors or lesions that might go undetected in other tests.
“By using computer-aided vision and detection systems across imaging modalities, automation of repetitive and time-consuming tasks like finding nodules and image-based biomarkers is now possible through AI,” Dr Shrinidhi Nathany, consultant, molecular haematology and oncology, Fortis Memorial Research Institute, told Times Now.
Advances in cancer detection using artificial intelligence
Last year, a US-based company launched Sybil: an AI tool trained on low-dose chest CT scans, recommended for people aged 50-80 who have a significant smoking history or currently smoke.
For patients being screened for lung cancer, Sybil has been able to look at an image and accurately predict a patient’s risk of developing lung cancer within six years.
Lung cancer is the most commonly diagnosed cancer worldwide, accounting for over 12.4 percent of all cases, but doctors believe it is not often talked about because of the stigma of being considered a self-inflicted cancer, largely due to smoking.
“Lung cancer, due to its insidious nature, remains one of the most important public health dilemmas. However, the adoption of biomarker testing and the discovery of the processes that drive it have led to a paradigm shift in the prognosis of this terrible disease, leading to an improvement,” said Dr. Nathany.
Similarly, another AI tool developed and studied by researchers at Harvard Medical School, the University of Copenhagen, and the Dana-Farber Cancer Institute was able to identify those at highest risk of developing fatal pancreatic cancer based solely on the patient’s medical records, three years before the actual diagnosis.
How does AI work in diagnosis?
According to Dr. Nathany, AI predicts accurate detection based on tumor markers and the nature of the disease, along with the patient’s genetic conditions.
“If you train the machine like you do with the human brain, for example, you know that when the sun rises it is morning: the brain has been trained. We also have thousands of open source databases to train the machine with lung CT images to detect even the smallest nodule with an accuracy close to 95 percent and then use meta-attributes to predict the nature of the disease and the possible genetics of the tumor,” he added.
In addition, AI analyzes images from mammograms, X-rays, MRIs, ultrasounds, and tissue preparations to identify cell properties and highlight lesions. It can help doctors detect disease by providing detailed views of cells, detecting anti-tumor immune formations, and automatically identifying tumor and cancer cells.
“AI can also help predict outcomes in already diagnosed patients, including the risk of spread to the brain and other organs, as well as overall survival with considerable accuracy,” Dr. Nathany said.
Disclaimer
The information contained in this post is for general information purposes only. We make no representations or warranties of any kind, express or implied, about the completeness, accuracy, reliability, suitability or availability with respect to the website or the information, products, services, or related graphics contained on the post for any purpose.
We respect the intellectual property rights of content creators. If you are the owner of any material featured on our website and have concerns about its use, please contact us. We are committed to addressing any copyright issues promptly and will remove any material within 2 days of receiving a request from the rightful owner.