An Artificial Intelligence System that can detect prostate cancer
Researchers have developed an artificial intelligence system that can diagnose prostate cancer. The prostate is a gland surrounding the neck of the bladder in males. This artificial intelligence system is trained to identify cancer and to classify it. It can also identify the stage of cancer and its aggressiveness. This system is called FocalNet. It predicts disease-evaluating MRI or scans. The system was trained by having it analyze 417 men with prostate cancer. Outputs were fed into the framework so it could figure out how to evaluate and order tumors in a steady way and have it compare the results with the genuine pathology example.
Is it better than radiologists?
A detailed comparison was done between this system and experienced radiologist, to check the accuracy level. It was examined that the accuracy level was almost the same. An experienced radiologist with at least 10 years’ experience is 83.9 percent accurate in reading MRI’s, while FocalNet is 80.5 percent accurate. It is very difficult to detect cancer by reading MRI’s. And the problem is that many hospitals don’t have the resources to implement highly specialized training for detecting cancer. It is also very difficult to measure the grade of cancer that if the tumor is benign or cancerous. But this system detects cancer by reading an MRI in a consistent way. The researchers concluded that this artificial intelligence system can save time and it can also provide guidance to the less experienced radiologists.
Will it be a success?
For an MRI or CT scan, you need to take appointments because there are a lot of patients that are to be entertained each day. And then the results of the MRI also take a lot of time. If any patient is in serious condition then it is very difficult for the patient to wait so long. Moreover, you need 10 years experienced radiologist for reading MRIs. So this system is far better as it saves a lot of time and the accuracy level is almost the same as the experienced radiologist. It will be a success if it performs the tasks as claimed.