Google study shows that machine learning is capable of detecting lung cancer

Google has been working on a number of fronts to show how machine learning can help things that go beyond technology such as arts and health. A new survey conducted by scientists at the Mountain View giant found that it is possible to detect signs of lung cancer.

Researcher responsible for the study, Danial Tse has created a deep learning algorithm that can detect malignant lung nodules in more than 42,000 advanced tomography exams. And the results were encouraging: the algorithm revealed 11% less false positives and 5% fewer false negatives when compared to human-made analyzes.

Lung cancer is a disease with a high mortality rate. In the United States alone, more than 160,000 people died of the disease last year. In Brazil, it is the mode that kills the most, with rates that reach 82%. Therefore, studies on the disease are so important.

As was said earlier in the text, it is not the first time that artificial intelligence and machine learning are used in the treatment of cancer. Alphabet, Google's parent company, is researching for cancer detection.

A team from Google Brain, the company's artificial intelligence research division, presented a paper on cancer detection using machine learning allied to a modified light microscope, which presents an augmented reality view.

The idea is that these measures help make tracking more reliable with artificial intelligence. The discovery is still considered a small step in the field of science, as technology will still have to deal with privacy issues and the quality of the data collected.

Source: Technology Review

Stay Tuned For More :)

Post a Comment


  1. Inquiries around the potential effect of AI on the working environment are unavoidable. machine learning certification