Author丨Fengse Fish and Sheep

Collected all the new crown machine learning “watch movies” papers, Nothing works for one article ? !

Just as the news that “two medical staff in Guangzhou tested positive for nucleic acid” once again affected the nerves of the big guys, a new study from the University of Cambridge also ignited a barrel of explosives on the Internet.

Since the beginning of the epidemic, many researchers in the field of machine learning hope to use the power of AI to help medical staff better protect themselves:

From January to October last year, more than 2,000 related papers were published.However, just when AI is more mature Watch a movie Field, Cambridge University’s conclusions are surprising:

The collected papers on the use of AI for medical image detection and diagnosis of the new coronavirus have major flaws and deviations, and there is no possibility of clinical use. The collected papers on the use of AI for medical image detection and diagnosis of the new coronavirus have major flaws and deviations, and there is no possibility of clinical use. The paper has been published on Nature Machine Intelligence. And such “cruel” results can be said to have stirred up waves with one stone. Experts and scholars have reposted the discussion, and it has also triggered heated discussions among netizens on social media. Many relevant practitioners said: “This has taught us an important lesson.” What is going on? Why can’t it be used? Specifically, the researchers at the University of Cambridge initially used the “machine learning model” and “CXR models” in the preprinted paper platforms such as bioRxiv, medRxiv, and arXiv, as well as the EMBASE and MEDLINE databases.(Chest X-ray photo)/CT image diagnosis, prediction” such keywords, collected 2212 Related studies. After excluding papers that lack external verification, ignoring data sources, or incomplete model training information, the papers that passed the preliminary screening were: 415 articles . And to further improve the requirements for papers, such as excluding RQS(Radioactive quality score)＜6, failed to pass CLAIM(Medical imaging artificial intelligence checklist)After the model, the papers selected for the final review stage are 62 Stories . And these 62 articles, Have no potential clinical application value . In response, one of the authors of the paper, Dr. James Rude from the University of Cambridge School of Medicine, said: The international machine learning community has made tremendous efforts to help respond to the Covid-19 epidemic. These early studies gave us some hope, but their method and reporting deficiencies are very common. None of the papers we reviewed achieved the robustness and repeatability necessary to support clinical applications. The international machine learning community has made tremendous efforts to help respond to the Covid-19 epidemic. These early studies gave us some hope, but their method and reporting deficiencies are very common. None of the papers we reviewed achieved the robustness and repeatability necessary to support clinical applications. Of the 62 papers, 55 were found to be at high risk of bias due to various problems, including relying on public data sets, many of which were suspected of being positive for Covid-19 and were not detected in CT images. △ Data indicators for each paper All of these models appear to be highly accurate in the study, and they are fully revealed as soon as they arrive in the clinic (for example, different types of patients or imaging scans obtained with different equipment). Behind this “annihilation” is mainly the problem of data sets. Many of these models are trained with very small sample data sets, and some data only come from one hospital. If you change a city to a hospital, this model will not work at all. There are also models that are trained based on the publicly available “Frankenstei Data Set”. The problem with such a large data set is that as time goes by, the data set continues to develop and incorporate new data. These changes are likely to make the original results impossible to reproduce. What’s more, the same data set is used for training and testing. Of course, this may not be deliberately done by researchers. For legal and commercial reasons, many medical data sets have to be Keep secret , There are really few large and diverse data for researchers to train and verify. This also makes the machine learning research produced in the health care field particularly difficult to replicate. MIT has done a study: The recurrence rate of medical AI papers is only 23%, compared with 58% in the field of natural language processing and 80% in the field of computer vision. But the data set is only one aspect. There are also problems with improper method design and lack of participation of radiologists and clinicians. For example, a training set with a model uses CT images of children as “non-Covid-19” data, and adults as “Covid-19” data. But in fact, children treated in pediatrics are very different from adults in terms of human anatomy. Such data settings are unreasonable, and the trained model will have large deviations. “Whether you are using machine learning to predict the weather or detect diseases, it is important to ensure that different experts work together and speak the same language so that you can focus on the right issues.” Unfortunately, many models do not allow radiology departments. Doctors and clinicians get involved. In addition, time limit It can also be interpreted as an “excus” for this series of questions. “These obstacles must be overcome, Otherwise, we will face a crisis of trust” Of course, behind the fact that a large number of such papers are published but all cannot be applied, it shows that the paper a There are also problems, such as the lack of in-depth understanding of machine learning by the reviewers, or blind trust in well-known institutions or companies, etc., which led to these papers being passed hastily. But the most important thing is the lack of review institutions Consistent standards To evaluate machine learning research in the medical field. Researchers at the University of Cambridge believe that a common set of standards must be established between authors and reviewers to ensure that the research really solves practical problems. Finally, although a large number of Covid-19 models have been found to be unable to be reproduced for clinical use, researchers at the University of Cambridge said that after some key modifications, these machine learning models can still become powerful tools in the fight against the new crown. They gave some conclusions and suggestions:

• Public data set May cause serious risk of deviation, use with caution;
• In order to make the model applicable to different groups and independent external data sets, the training data should be kept Variety and appropriate size ；
• In addition to higher-quality data sets, you also need Reproducible and externally verified In this way, the possibility of the model being promoted and integrated into future clinical trials can be increased.

Public data set May cause serious risk of deviation, use with caution; In order to make the model applicable to different groups and independent external data sets, the training data should be kept Variety and appropriate size ； In addition to higher-quality data sets, you also need Reproducible and externally verified In this way, the possibility of the model being promoted and integrated into future clinical trials can be increased. And said that these obstacles must be overcome, otherwise, where will people’s trust in artificial intelligence start?In addition, due to privacy restrictions, it is difficult to obtain medical data. In addition to federal learning, you can also refer to the latest published on the cover of Nature. Joint learning (Swarm Learning), A medical data sharing technology superior to federal learning. Is it okay for AI to watch movies? Facing the conclusion of the University of Cambridge, some netizens are worried: Today, the value and credibility of AI/ML are being diluted. When I hear words like “AI solution” and “AI driven”, I even feel nervous. Today, the value and credibility of AI/ML are being diluted. When I hear words like “AI solution” and “AI driven”, I even feel nervous. Some netizens believe that this is not unrelated to the current trend of “watering” papers in the field of machine learning. However, some netizens objectively analyzed that AI has indeed played a role in medical imaging, but they can’t replace doctors, they are more acting as doctors. assistant And at present, AI has no way to deal with some really difficult situations. And for what the article talks about data Many netizens agree with the question: Data is not the only problem, but it seems to be the most direct cause. It’s time to build a database like “Protein Bank” for AI models and data sets. Data is not the only problem, but it seems to be the most direct cause. It’s time to build a database like “Protein Bank” for AI models and data sets. In fact, no matter whether you are skeptical about AI medical care or not, now, at least like AI watching movies Such medical services have truly come to the public. Prior to this, colleagues of qubits had asked AI to take a CT scan at Zhongguancun Hospital. In the communication with first-line doctors, the chief doctor in charge of the physical examination center in the tertiary hospital also revealed to us that single-point applications such as the detection of lung nodules have been able to help doctors reduce the burden. There are even radiologists who said, “I don’t use it now (AI assisted viewing), and I’m not used to it.” Stat News commented on this: Machine learning is booming in the medical field, but it is also facing a credibility crisis.

Machine learning is booming in the medical field, but it is also facing a credibility crisis.

So, do you think AI medical care will work?

— Finish —

This article is the original content of the NetEase News•NetEase Featured Content Incentive Program signed account[qubit]. Unauthorized reprinting of the account is prohibited.

This article is reproduced with permission from Qubit (ID: QbitAI ), please contact the original author if you need to reprint

Welcome to forward to the circle of friends.

Every day, hundreds of vloggers and fans come to the village to film and take pictures with the emerging internet phenomenon. For streamers, Cheng is an abundant “view mine”. Many people even stationed themselves in the village to broadcast day and night. The village with a population of less than 1,500 people sometimes welcomes up to 1,000 streamers. Cheng is famous on Chinese social media for a bowl of noodles priced at 3 yuan. Live stream army Due to the large number of visitors, Cheng could not even go to the market as usual or leisurely go up the mountain to walk like before. There were too many people to find, he had to move the noodle shop to sell it right at the gate of his house. Cheng’s house is built on the mountain, with only one main room and two small rooms on the side. Near noon, the streamer and the audience crowded outside the gate with cameras and phones held. Cheng is tucked in the middle with a kneading table, the scene is like a Roman arena. “Noodle brother, say a few words”, “Look here and say hello to the audience”, the vlogger’s invitations kept ringing. In the crowd, live streamers do all sorts of things like dancing, singing, arguing to get the attention of their audience. Since becoming famous, Cheng has mostly just stood at the gate making noodles or rested in the house. His mother set up a beverage stand outside the gate to sell to guests who came to see. Facing the curious crowd at her door, she appeared calm. However, she still couldn’t understand why her ordinary second son was so attractive. The mother also said that she did not expect this age to be able to earn money thanks to the popularity of the child. The live stream army surrounded Cheng’s house day and night, following him while he was walking in the mountains. Thanks to the popularity of the “noodle brother”, people in the village also have more jobs. Dozens of food stalls were set up near his house. Chen’s village is a backward agricultural area, with only a small asphalt road leading to the village. Previously, this road did not have much traffic, sometimes there were cars passing by without stopping. To create a good image, Cheng’s village repaired the road through the night, in cooperation with the telecommunications company to install a high-speed network. Many villagers gave up farming, turned to selling food, opened inns, some imitated vloggers and went live in hopes of making money from it. From 10am to midnight, thousands of people come and go from the village. There are many people who drive cars for 3 hours from Xuzhou, Jiangsu come here just to eat a bowl of 3 yuan noodles. Some people from Inner Mongolia came here just to take pictures with Cheng. Some people bring their children here so that the children can be inspired by the “noodle brother”. Liaocheng, who lives 390 km from Cheng’s village, uses the word “pilgrimage” to describe his journey to meet the “noodle brother”. Liaocheng said that like many others, he sees Cheng as a star. Talking about her popularity, Cheng expressed it was a coincidence, like a dream. After talking for a few sentences, he scratched his head again and smiled shyly. Cheng is tired of having to receive many curious people to film and take pictures. Cheng’s life is turned upside down. Many people scrutinize, probe and disturb his family. Some people deliberately broke bricks in the wall to look into the yard, at night, there were people who took beer bottles and rotten vegetables and threw them into the house. Cheng looked much thinner, his skin dark and the deep wrinkles at the corners of his eyes made him look older than his years. Since becoming famous, he has not taken a day off. During the past 2 days, he had a cold, his body was more emaciated. At 15:00, it was pouring rain, Cheng had finished the last dough and was about to stop selling. He looked tired, his eyes lost. He signaled to the spectators and quickly went inside, leaving his relatives to clean up the tools.

Cutting leg nerves to have slimmer calves is being seen as a horror beauty trend in China. It’s crazy that some people will in the name of beauty, or follow so many strange beauty trends to achieve “body perfection”. Take, for example, the new and disturbing trend known as “calf block,” which is essentially young, healthy girls who assume that there are “less important” nerves in their legs to be operated on. amputation to permanently atrophy the calf muscles, thereby making the legs appear slimmer and straighter. In an interview with The Paper on May 20, Doctor Zeng Chun – Deputy Director of the Department of Joint Surgery and Sports Medicine, Affiliated Hospital 3 of Nam Medical University said: He was shocked when he heard the news. One girl had some nerves removed in her legs to make her calves look slimmer. Patients with nerve injury and calf muscle atrophy, constant pain need to find ways to recover. But I didn’t expect that there are healthy people who actively destroy their nerves,” said Dr Zeng, adding that there is no such thing as a less important nerve, and once the nerve has been If it is damaged, there is a high chance that it will never fully recover. Doctor Zheng Chun explained that to treat a broken bone is quite easy, but once the nerve is damaged, there is a high chance that it cannot be restored. On online forums, the keyword “calf block” is searched by many people in the hope of having a solution to help slimmer calves. According to information shared from these service providers, customers will have some less important nerves removed in the calf, gradually the muscles will be weak and atrophy, helping to make the legs smaller and straighter. On Good Doctor Online, a girl who has performed this surgery said that two months after the operation, she still could not walk normally, her legs were numb, and she could not stand on her feet. She was very worried, scared and asked the doctor if there was any way to fix it. Following the revelations in The Paper’s article, this disturbing beauty trend has gone viral in China, where many beauty bloggers and vloggers blame the promotion of this type of trick, but also girls. undergoing unproven surgeries in a desperate attempt to achieve unrealistic beauty standards. Unfortunately, this isn’t the only shocking beauty trend that has popped up over the years. Beauty trends such as tattooing, tattooing pupils, dilating lips, splitting tongues, piercing, surgery to resemble animals … are considered the most shocking bizarre beauty trends on the planet.

However, livestream has become a “trend” pursued by many people, but not all “creators” create value. Bullshit, vulgar content is posted by many vloggers defiantly to attract attention and receive donations (money and gifts) from viewers. On major live-streaming platforms in the country of billions of people, such as Bilibili, Douyin, and Kuaishou, it is not difficult for viewers to come across images of female vloggers dressed in revealing, offensive clothing and performing erotic acts. Many streamers are banned from showing off their bodies or showing sexual acts to receive donations. Showing off, deceiving viewers Many live broadcast studios often hack the number of followers to create credibility, attract viewers so that they are willing to donate money. The account holders also deliberately fabricated sensational stories, creating conflicts with others to attract viewers of their livestream. “Poor students borrow more than 100,000 yuan to reward female streamers in Nanjing”, “16-year-old teenager rewards female streamers with 1.58 million yuan for 3 months”, “Male cashier appropriated more than 48 million yuan to donate to the live broadcast room”, a series of negative news related to direct money rewards through livestream has caused many controversial streams in the Chinese online community. On May 17, Kuaishou posted a notice saying that live streamer Yin Shihang, an account holder with 8 million followers, had been banned from the platform for “sensational marketing”, “vulgar performances” and false advertising during the livestream on May 15. Initially, Yin introduced to live stream his marriage proposal with girlfriend Tao Lulu. Millions of people watched to witness the special moment of the popular streamer. In the end, he tricked the audience by turning the broadcast into a 5-hour sales show. Yin Shihang was banned from broadcasting when tricking viewers into the livestream. Various live broadcasting platforms are working to strengthen control over violations. Kuaishou has issued a special regulatory notice on “Scripting, Acting and Purchases”, which bans accounts or imposes fines on those who take advantage of family conflicts and other factors unrelated to the issue. sell merchandise to entice fans to buy. In June 2020, Douyin issued a new rule that allows minors to privately award rewards after verification and can be fully refunded. In April 2021, Douyin revealed that a total of 36,000 minors had been refunded in 11 months, with an average refund time of 19.7 hours. At the end of November 2020, the State Administration of Radio, Film and Television of China issued a notice on strengthening the management of livestream and live broadcast of e-commerce. In which, methods such as facial recognition are applied to implement real-name management for broadcasters and users. The new rule also limits the number of rewards users can receive by time, daily and monthly, if there are too many rewards, there will be a reminder message and the reward function will be paused if the limit is exceeded.

Chinese blogger was criticized for sharing about “goblin ear” surgery. “Goblin ears help me look a lot younger. With the elongated earlobes, my face becomes smaller, the lines are more elegant,” Chen shared. Chen’s story quickly caused a great controversy when many people believed that the female blogger’s shares were promoting a beauty trend that was weird, toxic and dangerous to health. Li Zhengyong, deputy director of the Department of Aesthetics and Burns, West China Hospital of Sichuan University, said that “goblin ear” actually refers to a birth defect in the ear. A person with normal ears who wants to have surgery to become “goblin ears” will have to take cartilage from several places on the body to cover the earlobe. “This type of surgery can cause dangerous complications, and the consequences are unpredictable,” Dr. Li said. Not only doing “goblin ear” surgery, Chinese young people also follow many strange beauty trends to achieve “body perfection”. Another controversial cosmetic trend recently is cutting nerves to have slim calves. In an interview on May 20, Dr. Zeng Chun, Deputy Director of the Department of Joint Surgery and Sports Medicine, Affiliated Hospital 3 of Southern Medical University, said he was shocked when he heard that a girl had died. get rid of some nerves in your legs to make your calves look leaner. Many people blindly believe in neurectomy for thinner calves. “Patients with nerve injury and calf muscle atrophy are very painful, they find ways to recover. I did not expect that there are healthy people actively destroying their nerves,” said the doctor. doctor expressed. On online forums, the keyword “calf block” is searched by many people in the hope of a solution to help slim calves. According to information shared from these service providers, customers will have some less important nerves removed in the calf, gradually the muscles will be weak and atrophy, helping to make the legs smaller and straighter. Doctor Zheng Chun explained that to treat a broken bone is quite easy, but once the nerve is damaged, there is a high chance that it cannot be restored. Zheng also asserts that no nerve is considered less important in the body. On Good Doctor Online, a girl who has performed this surgery said that two months after the operation, she still could not walk normally, her legs were numb, and she could not stand on her feet. She was worried, scared and asked the doctor if there was any way to fix it. However, she could not find an answer. Many netizens expressed frustration when many girls still believe in unscientific surgery because they want to have a perfect appearance and harm themselves. On the other hand, bloggers sharing about the “calf block” method also received a lot of criticism when creating a malicious trend.

“In the broadcast room, he introduced a beauty device, excited screams broke out, but I just watched with an outsider mentality. Then they asked a girl in the room to try the product. , the left side of her face looks a bit smaller than the right side. Admittedly at that moment I was ‘frozen’ because I was so surprised.” Shengxinte had heard about the lifting and firming effects of that product, but its price was not cheap. “When I was hesitating, the livestream room offered 24 offers, I thought the opportunity had come to me. Quick calculation in my head, I found it was a really good price, decided to buy immediately with one click” . Young Chinese people prefer home beauty devices such as face washes and lifting massage machines. Shengxinte is happy to have successfully purchased the item. However, when she told her friends about buying a beauty machine, they all said that it was not necessary because she was only 23 years old, did not have wrinkles and her skin was also healthy, no need to lift or tighten. “I think choosing to buy is also a psychological factor, but now that I have used the device for 6 months, I am satisfied. I see smaller pores and a clear jawline.” “Buy it on livestream for 4,492 yuan, about 1,000 yuan cheaper than usual, I think it’s a bargain,” said Xiaosheng (23 years old) praising her new face lift device. Like many other young people, she also started the “anti-aging path” early. According to the The Pape r, as beauty technology is more and more developed and popular, young Chinese are also becoming more willing to spend on beauty care. According to survey data of CBNData, 91.3% of people born after 1990 believe that cosmetic surgery is one of the anti-aging methods, they believe it much more than options like “care” skin care” or “exercise”. The market for anti-aging beauty devices for women has experienced a boom since 2020, especially with products priced at around 3,000 yuan growing strongly. Home beauty tools with technology are sought after and become a new hobby among young people. Regret when rushing to spend money Lili is a banker in Hangzhou. During a trip to Japan in 2016, she was led by a tour guide to an electronics center to buy a Hitachi face wash. “I am willing to spend a large amount of money to buy. The function of the machine was introduced as cleaning the skin, shrinking pores, but after a few days of use, I realized I was ‘paying IQ tax’. to use, i need to take a cotton pad and pour makeup remover on it, after washing the cotton pad will be yellow, which means the skin looks clean.In fact, just holding the cotton pad with your hand gives the results the same, similar”. Many people regret after buying expensive but ineffective beauty products. She says it feels cheated when there’s no visible effect in shrinking pores. After a few uses, she left the device idle. “It’s a waste of money to buy such an item when a cotton ball is equally effective.” A beauty blogger said she often looks at skin care products, buys them to try and review them on her channel. In 2014, to thin her face, she bought gold masks through Taobao Japan for 700-800 USD. After a few uses, she saw many people using gold bars for skin care mainly due to propaganda effects, the actual beauty results were not high. “I wouldn’t say they’re useless, but for me personally, it’s not very effective. The more I study how this product works, the more tasteless it becomes.” Seeing crow’s feet appear around his eyes when looking in the mirror, Li Zijun (Xiamen) suddenly realized the importance of taking care of his eyes. She immediately went online to search for products, initially intending to buy eye cream, eventually she was drawn to beauty devices. After consulting, she bought an eye massage roller for 1,500 yuan. The product is advertised to improve wrinkles around the eyes, puffiness and dark circles, and can also be used on the nose. After receiving the goods, Zijun was a bit disappointed because the material of the machine is plastic. Before use, she needs to apply cream around her eyes and the steps are quite troublesome. Having used it a few times, she did not notice any improvement. “I can’t blame the product because I have to persevere to be beautiful. But I’m the type of person who rushes, wants to see results immediately. Now I’m reselling this device for 500 yuan. I recommend everyone Consider, don’t be too hasty when you want to buy home beauty tools.”