AI is Being Trained to Spot Anomalies During Tests in the Medical Sector
Audra Moran, the head of a global charity (“OCRA”) focused on researching ovarian cancer, says her area of expertise is underfunded and very deadly. Most cases of this disease originate within the fallopian tubes and spread to the ovaries. The problem is, by the time the malignancy spreads to the ovaries, there is a high likelihood that it has also spread to other parts of the body. This is why early detection is critical as life-saving interventions can be carried out.
The life-changing window of detection is approximately 5 years before the onset of ovary cancer symptoms, but it is currently difficult to detect the condition within this period.
That is where artificial intelligence (“AI”) comes in. New testing techniques are exploring how AI can be leveraged to detect ovarian cancer early enough to make a difference in the mortality of patients. Other deadly infections, such as pneumonia, can also benefit from this AI-testing method.
Doctor Daniel Heller, a biomedical engineer based at the cancer center at Memorial Sloan Kettering in New York, heads a team developing technology to conduct blood tests using nanotubes. Nanotubes were discovered two decades ago, but how to leverage them during diagnostic tests had eluded scientists. Dr. Heller and his team are working to solve this challenge.
They took blood samples from two groups of patients; one group contained samples from people who have been diagnosed with ovarian cancer. The other group didn’t have the disease. This data was uploaded into an algorithm that uses machine learning to detect patterns within those blood samples. One challenge the team encountered was the limited data on ovarian cancer patients. This is because the condition is rare, and the patients who are diagnosed with the cancer are treated by different hospitals whose data isn’t readily available to research teams.
Dr. Heller says they took a chance and used data from just 100 patients to train their AI model. It was a desperate attempt, and they were pleasantly surprised to see that the AI model was able to perform more reliably than other biomarkers currently in use to detect the cancer.
He says the system is undergoing further development and it could be 3-5 years away from commercialization. Nevertheless, the team is hopeful that this AI model can be deployed to detect other gynecological disorders so that doctors can use the tool to positively identify which particular condition a patient is suffering from.
As more use cases of AI emerge, not just in the biomedical field but also in other walks of life, the demand for minerals like gold and copper that are critical in the burgeoning AI industry will increase. Companies like McEwen Mining Inc. (NYSE: MUX) (TSX: MUX) that focus their efforts on extracting these AI metals are well positioned for the future ahead.
NOTE TO INVESTORS: The latest news and updates relating to MUX are available in the company’s newsroom at https://nnw.fm/MUX
Please see full terms of use and disclaimers on the NetworkNewsWire website applicable to all content provided by NNW, wherever published or republished: http://NNW.fm/Disclaimer
Audra Moran, the head of a global charity (“OCRA”) focused on researching ovarian cancer, says her area of expertise is underfunded and very deadly. Most cases of this disease originate within the fallopian tubes and spread to the ovaries. The problem is, by the time the malignancy spreads to the ovaries, there is a high likelihood that it has also spread to other parts of the body. This is why early detection is critical as life-saving interventions can be carried out.
The life-changing window of detection is approximately 5 years before the onset of ovary cancer symptoms, but it is currently difficult to detect the condition within this period.
That is where artificial intelligence (“AI”) comes in. New testing techniques are exploring how AI can be leveraged to detect ovarian cancer early enough to make a difference in the mortality of patients. Other deadly infections, such as pneumonia, can also benefit from this AI-testing method.
Doctor Daniel Heller, a biomedical engineer based at the cancer center at Memorial Sloan Kettering in New York, heads a team developing technology to conduct blood tests using nanotubes. Nanotubes were discovered two decades ago, but how to leverage them during diagnostic tests had eluded scientists. Dr. Heller and his team are working to solve this challenge.
They took blood samples from two groups of patients; one group contained samples from people who have been diagnosed with ovarian cancer. The other group didn’t have the disease. This data was uploaded into an algorithm that uses machine learning to detect patterns within those blood samples. One challenge the team encountered was the limited data on ovarian cancer patients. This is because the condition is rare, and the patients who are diagnosed with the cancer are treated by different hospitals whose data isn’t readily available to research teams.
Dr. Heller says they took a chance and used data from just 100 patients to train their AI model. It was a desperate attempt, and they were pleasantly surprised to see that the AI model was able to perform more reliably than other biomarkers currently in use to detect the cancer.
He says the system is undergoing further development and it could be 3-5 years away from commercialization. Nevertheless, the team is hopeful that this AI model can be deployed to detect other gynecological disorders so that doctors can use the tool to positively identify which particular condition a patient is suffering from.
As more use cases of AI emerge, not just in the biomedical field but also in other walks of life, the demand for minerals like gold and copper that are critical in the burgeoning AI industry will increase. Companies like McEwen Mining Inc. (NYSE: MUX) (TSX: MUX) that focus their efforts on extracting these AI metals are well positioned for the future ahead.
NOTE TO INVESTORS: The latest news and updates relating to MUX are available in the company’s newsroom at https://nnw.fm/MUX
Please see full terms of use and disclaimers on the NetworkNewsWire website applicable to all content provided by NNW, wherever published or republished: http://NNW.fm/Disclaimer
