Artificial intelligence (AI) is a technology that is rapidly advancing across all industries. While AI technology is extremely important for business purposes, it is critical in life-altering fields such as healthcare.
For example, researchers have recently developed a new AI system to aid radiologists in diagnosing prostate cancer.
The AI system, called FocalNet, can significantly help radiologists predict and identify the aggressiveness of prostate cancer in a patient.
FocalNet uses magnetic resonance imaging (MRI) scans to evaluate the disease. Normally, radiologists use MRI to detect and assess malignant prostate tumor aggression. It analyzes the pixels in MRI scans for lesions in the prostate area.
For a radiologist, it normally takes practicing on thousands of scan to learn how to accurately determine if a tumor is cancerous or benign and to estimate cancer grade.
The amazing thing is that FocalNet is nearly as accurate as experienced radiologists, as reported by researchers from the University of California, Los Angeles (UCLA).
In tests, radiologists with at least a decade of experience were 83.9 percent accurate while FocalNet was found 80.5 percent accurate, according to a study published in the IEEE Transactions on Medical Imaging Journal.
FocalNet is an artificial neural network which utilized an algorithm comprising more than a million variables.
The network was trained by analyzing MRI scans of 417 men with prostate cancer. The scans were fed into the system, so it could “learn” to assess and classify tumors alongside the results of an actual pathology patient.
Researchers are hopeful that an AI system such as FocalNet could save time and potentially provide diagnostic guidance to less-experienced radiologists.
If caught in the early states, prostate cancer can usually be cured. Advanced systems such as FocalNet could significantly help the diagnosis of such cancer.