According to a study presented at the 2023 American Society of Surgeons Clinical Conference, researchers at Northwestern University have developed an artificial intelligence based tool to estimate the long-term survival chances of newly diagnosed cancer patients. Currently, the estimated survival rate of cancer patients mainly depends on their cancer stage. Researchers suggest that in addition to staging criteria, there are many other factors that may affect patient survival. The team used machine learning methods to create a prototype tool called the "Cancer Survival Calculator" and tested it on a large dataset. Preliminary tests estimated the 5-year survival rate of patients with breast cancer, thyroid cancer and pancreatic cancer. This time, the researchers collected relevant information about breast cancer, thyroid cancer and pancreatic cancer patients diagnosed in some years. The collected 3/4 data is used to train machine learning algorithms to identify patterns between diagnostic features and 5-year survival rates of patients, and then ranked according to the factors that have the greatest impact on survival. Using the remaining data, the team used statistical methods to test the accuracy of the prototype's estimated survival rate. The team included data of 259485 breast cancer patients, 76624 thyroid cancer patients and 84514 pancreatic cancer patients. According to the location of cancer, the four major factors that affect the survival of patients within 5 years after diagnosis are as follows: breast cancer (whether the patient has received cancer surgery, the age of the patient at diagnosis, the size of the tumor, and the time from diagnosis to treatment); Thyroid cancer (age at diagnosis, tumor size, treatment time, and lymph node involvement); Pancreatic cancer (cancer surgery, histological or microscopic analysis of cancer, tumor size, and age at diagnosis). Although some predictive factors such as tumor size are part of cancer staging, research results indicate that there are more factors that affect the survival of cancer patients after the disease stage. Researchers report that validation tests have shown that the calculator is highly accurate in estimating survival rates for all three cancer sites. Meanwhile, the advantage of this' cancer survival calculator 'lies in a more comprehensive dataset and the use of new machine learning data models, which result in faster processing speed and more accurate risk prediction. When doctors evaluate the prognosis of tumor patients, they often use a term -5-year survival rate. This is a statistical indicator, which is the probability of a cancer patient surviving for more than 5 years after clinical comprehensive treatment. Of course, any probability is relative to the group, and each individual's situation may be different, leading to miracles or regrets. This time, researchers have developed an AI tool that incorporates more comprehensive indicators to evaluate the 5-year survival rate of patients. Of course, evaluating how long a patient can live should not be the ultimate goal. If key factors that affect the survival period of cancer patients can be identified, it may extend their lives for more people. Author: Zhang Mengran (Xinhua News Agency)