According to the Wall Street Journal, a team of scientists recently said that its artificial intelligence program diagnosed pneumonia through chest X-rays more accurately than human radiologists. Should radiologists be worried about this? Maybe not. AI is more likely to make their work easier, rather than making them redundant.
Earlier this month, a team of computer scientists at Stanford University announced that their artificial intelligence programs were more accurate than human radiologists in the diagnosis of pneumonia through chest X-rays.
“A radiologist should be worried about his rice bowl?†one of the scientists, Andrew Ng, said in a tweet.
Maybe they don't have to worry. In fact, they should breathe a sigh of relief: AI is more likely to make their work easier, rather than making them redundant.
Today's radiologists face the same 21st century problems as most professionals: too much data. The latest AI iterative technology machine learning can carefully screen those data and discover patterns that humans may not see. But the machine doesn't know how to deal with that information -- at least not at the moment. Only humans can do the work of this step.
In recent decades, the number of medical images has increased dramatically, indicating that the workload of doctors is increasing, the types of scans are more, and the images produced by each physical examination are also increasing. In the past, CT scans that only captured one body dimension can now capture three dimensions, resulting in hundreds or even thousands of highly detailed images.
Steve Tolle, vice president of global strategy for Watson's medical imaging division at IBM, estimates that the number of radiological examinations in the United States is about 800 million per year, generating a total of about 60 billion images, or each. The radiologist produces one every two seconds. "And our expectation is that they can diagnose correctly every time," he said.
He pointed out that the huge burden of all those data explains why some medical images in the UK have not been viewed for several weeks, why many radiologists feel exhausted and why the hospital lacks radiologists, even though The salary of the position is very high.
IBM believes that AI can alleviate those burdens. It is one of many companies that develop AI applications for medical imaging.
IBM's team of scientists injects thousands of medical images that have been manually diagnosed for the presence of tumors or other symptoms. The computer generates a model to analyze the new image and tries to determine if a tumor is present.
In one experiment, IBM Watson's accuracy in diagnosing melanoma based on skin lesion images reached 76%, and the average accuracy of more than 8 dermatologists was 71%. The company hopes that one day, a doctor or nurse will be able to take a picture of a skin lesion and upload it to an AI app, which will then tell you the probability of developing cancer.
Since AI is learned through human-identified patterns, it cannot know more than humans, but it can operate more stably and is less biased when applying that knowledge.
John Park from Anne Arundel Medical Center is working with IBM Watson. He pointed out that radiologists tend to seek to discover what they have seen recently. By contrast, once AI has seen something, it will never forget that it will apply relevant knowledge to every image diagnosis. This may be particularly valuable in poor countries where there is a high lack of trained radiologists.
Limitations of AIHowever, AI can only derive probabilities based on what the image displays, rather than deterministic information. Jaime Murillo of Sentara Healthcare Healthcare said it "may point me out what I might not have seen, but then I have to determine if it is accurate."
This is where the application of AI in various fields is restricted. Like radiology, Facebook is also plagued by excessive information. Facebook hopes that one day AI will be able to screen controversial content for its platform.
However, troubleshooting controversial content requires Facebook users to report to the "Community Operations" team, who then decides if they should be removed. After that, the AI ​​will pay attention to similar content and prevent it from being shared. Facebook currently employs thousands of people to review content.
Avi Goldfarb, an economist at the University of Toronto who specializes in AI, points out that “in many everyday AI applications, machines predict and organize the most relevant options, but in the end, humans make actual choices. He talks about BenchSci and Atomwise, which will recommend potential research pathways to scientists based on their analysis of scientific literature and large databases of antibodies and molecules.
Keith Dreyer, chief data science officer at the American Society of Radiology, said the algorithm can be trained to look for a particular situation, but "most of the time you don't even know what you are looking for."
"The patient has a cough, there may be thousands of reasons," he said. The "white clouds" on the chest X-ray may be pneumonia or liver cancer.
Maximizing accuracy requires an algorithm to train each symptom and disease separately. This is a process that requires enormous manpower and material resources. Dreier pointed out that the US Food and Drug Administration must then approve systems and doctors to integrate algorithms into their practice.
There have been some commercial AI applications for medical imaging, but IBM Watson has not yet launched such applications. A spokesperson for the company said that the launch of the proposed application for breast and skin cancer depends on when the regulatory body passes the review.
People’s panic about AI’s robbing of rice bowls is reminiscent of offshore outsourcing concerns about all types of work nearly 20 years ago. Some economists have predicted that international broadband connections mean that many medical image interpretation work will be outsourced to Indian radiologists with much lower salaries. In fact, this has never happened.
US regulators do not want doctors who are not under their control to access medical images of American patients, and few in emerging markets have the skills and experience they need. Since 1995, the number of radiologists in the United States has increased by more than 40%.
A high-speed slip ring is an electromechanical device that allows electrical current to pass between a stationary and a rotating assembly. It is typically used in applications where the speed requirement exceeds that which is achievable with a standard slip ring. High-speed slip rings are available in a variety of sizes and configurations to meet the needs of most applications. They are commonly used in military, aerospace, and medical applications where speed and reliability are critical.
The speed is required to be at least tens of thousands of revolutions per minute, the structure of the Conductive Slip Ring directly determines the contact of high speed, as well as the size of impedance. High-speed slip ring materials have high wear resistance, high life characteristics, and multi-contact contact can ensure signal transmission and power reliability, with the development of science and technology, high-speed slip rings will be more and more widely used.
If you need to transfer large amounts of data or power between two stationary objects, a high-speed conductive slip ring is what you need. Traditional slip rings only offer a few thousand revolutions per minute (RPM), which is more than enough for most applications, but can be limiting in some circumstances. Some applications, like those found in the medical or military field, require speeds of tens of thousands of RPM in order to function properly. That's where our high-speed conductive slip ring comes in. With speeds up to a certain range, it can handle even the most demanding applications.
High Speed Slip Ring,Rotary Slip Ring Electrical,Slip Electrical Connector,Hydraulic Rotary Joint
Dongguan Oubaibo Technology Co., Ltd. , https://www.sliprobs.com