At present, the diagnosis of breast cancer by artificial intelligence is as good as that by human experts! Which one do you choose?

July 18, 2019 / BIOON / - breast cancer is the most common cancer in the UK It accounts for 15% of all new cases in the country and about one eighth of women will be diagnosed with the disease in their lifetime In the NHS, breast cancer screening usually includes mammography However, as the number of experts able to read such early tests decreases, the future of such tests will be at risk While this shortage of skills cannot be made up immediately, the expected progress in the field of artificial intelligence may help Source: http://cn.bing.com explains that mammography is a complex process, usually performed by specially trained radiologists and radiographers Their skills are essential for the early detection and diagnosis of this cancer At a meeting, they examined several batches of mammograms for signs of breast cancer But these signs are often ambiguous or hard to see The false negative rate of mammography was between 20% and 30% These errors are either perceptual errors or explanatory errors, which can be attributed to readers' sensitivity or particularity It is generally agreed that the key to developing the expertise needed to interpret mammography is rigorous training, extensive practice and experience Although researchers are studying training strategies and perceptual learning modules to speed up the transition from novice readers to experts, others have been studying how to use artificial intelligence to speed up diagnosis and improve the accuracy of diagnosis As in many other fields, the potential of AI algorithm to help cancer diagnosis has not been ignored In addition to breast cancer, researchers have been studying how artificial intelligence can improve the treatment effect and efficiency of lung cancer, brain cancer and prostate cancer to meet the growing demand for diagnosis Even Google is looking at how artificial intelligence can be used to diagnose cancer The search giant has trained an algorithm to detect metastatic tumors with a 99% success rate Photo source: http://cn.bing.com for breast cancer, the current focus has been on how artificial intelligence can help diagnose diseases through mammography Each mammogram is interpreted by two experts, which may lead to a delay in diagnosis if there is a lack of expertise But researchers have been considering the introduction of artificial intelligence systems in screening The idea is that it will support the discovery of experts without waiting for a second opinion from other professionals This will reduce waiting time and related anxiety for women's tests Artificial intelligence has made great progress in cancer image recognition At the end of 2018, researchers reported that the accuracy of a commercial system was comparable to that of 101 radiologists who interpreted more than 28000 screening mammograms This means that its cancer detection accuracy is comparable to that of radiologists In another study led by the same researcher, radiologists supported by artificial intelligence systems found an increase in the detection rate of breast cancer, from 83% to 86% The study also found that the use of artificial intelligence systems reduced the time it took radiologists to analyze images on the screen While the potential for fine-tuning and improving AI has been welcomed by some radiologists, it has also raised doubts among others Although other researchers have found that AI is as good as humans in detecting breast cancer through mammograms, it also brings a warning that more fine-tuning and software improvements are needed before AI can be safely applied to breast screening programs Photo source: http://cn.bing.com what's exciting is that AI may be used to help detect this ubiquitous cancer Before introducing AI, we need to consider the confidence of experts and the public It's crucial to embrace this technology so that patients and healthcare professionals know that they're getting the right results So far, there is not much research on the public's cognition of artificial intelligence in breast cancer screening, but more extensive research on artificial intelligence and health care has found that 39% of people are willing to engage in artificial intelligence / robot medicine In the 18 - to 24-year-old population, the proportion rose to 55% AI systems are still in the research phase and no company plans to use them to diagnose patients in the UK But these promising results suggest that there is a huge opportunity for the provision of radiation health care services, which could eventually lead to the discovery of more breast cancer and other cancer patients Reference materials: [1] breast cancer diagnosis by AI now as good as human experts [2] breast cancer statistics [3] can artistic intelligence identify normal mammograms in screening? 【4】 Why AI and robotics will define New Health 【5】Alejandro Rodríguez-Ruiz et al Detection of Breast Cancer with Mammography： Effect of an Artificial Intelligence Support System Radiology DOI： https：//doi.org/10.1148/radiol.2018181371 【6】 Artificial intelligence versus 101 radiologists 【7】Alejandro Rodriguez-Ruiz et al Stand-Alone Artificial Intelligence for Breast Cancer Detection in Mammography： Comparison With 101 Radiologists JNCI： Journal of the National Cancer Institute， djy222， https：//doi.org/10.1093/jnci/djy222
【8】 Mammograms