How can artificial intelligence help in breast cancer diagnosis?
According to the World Health Organization (WHO), breast cancer has already surpassed lung cancer to become the most common cancer in the world. On the other hand, the arrival of technology and, above all, artificial intelligence presents a new strategy to help discover the early diagnosis of the disease. Some… According to the World Health Organization (WHO), breast cancer has already surpassed lung cancer to become the most common cancer in the world. With the help of artificial intelligence, researchers have developed a method that helps doctors interpret blood tests and identify breast cancer early. The technology will analyze mammography exams to detect suspected breast cancer foci, such as microcalcifications and breast lumps. It is also possible to detect a tumor by marking in the exact spot, facilitating the doctor’s analysis.
نشرت : منذ عامين بواسطة David في Tech
According to the World Health Organization (WHO), breast cancer has already surpassed lung cancer to become the most common cancer in the world. On the other hand, the arrival of technology and, above all, artificial intelligence presents a new strategy to help discover the early diagnosis of the disease.
Some operators are already in trials to apply the technology, but how does this new way of speeding up the breast cancer detection process work? Understand it now in this article!
How does Artificial Intelligence contribute to breast cancer diagnosis?
According to the National Cancer Institute (INCA), by 2022, 66,280 new cases of breast cancer will be diagnosed in Brazil, with an estimated risk of 61.61 cases per 100,000 women. In 2017 alone, there were approximately 16,724 breast cancer deaths among women, equating to a risk of 16.16 in 100,000.
Recent studies can change this reality with the help of early diagnosis, which increases the chances of cure and survival of patients suffering from the disease.
One of these was created by researcher Daniella Castro Araújo, a doctoral student in the Postgraduate Program in Computer Science at the Federal University of Minas Gerais (UFMG).
With an artificial intelligence (AI) algorithm, the researcher has created a method that helps doctors interpret blood tests. In this way, the technology can help identify breast cancer early.
According to Daniella, artificial intelligence is used to recognize patterns in groups of patients with and without cancer and its application can help optimize the mammography queue, as there are many women who do not have access to this exam.
How was artificial intelligence developed to discover breast cancer diagnosis?
During its development, an application was used to analyze the databases of institutions such as Hospital de Amor, in Barretos (SP), and Grupo Fleury, a network of laboratories operating in São Paulo. To do this, they collected results of tests such as mammograms, blood tests and biopsies used to accurately diagnose breast cancer.
They then separated the women into two different groups, one made up of patients with the disease and the other made up of patients without.
Therefore, the researchers evaluated blood tests taken up to six months before the final diagnosis. The accuracy rate reached 70%, and if more data, such as reports and other tests were included in the history, the accuracy could be as high as 90%.
Another similar technology was created by the Korean company Lunit®️ and will be applied in the testing phase at the Beneficência Portuguesa Hospital. In this case, the diagnostic accuracy rate is 97%.
Once put into practice, AI will analyze mammography exams to detect suspected breast cancer foci, such as microcalcifications and breast lumps.
The most interesting thing is that, in this case, once the position of the possible tumor has been identified, there is a program to carry out a sort of marking in the exact spot, facilitating the doctor’s analysis. Amazing, right?
The post How can artificial intelligence help in breast cancer diagnosis? appeared first on Olhar Digital.
المواضيع: AI