Artificial intelligence can help a lot of doctors

Artificial intelligence can help a lot of doctors

A machine can better analyze environmental data, genetic data and patient history.

Several years ago, Silicon Valley investor Vinod Khosla wrote a provocative article entitled "Do We Need Doctors or Algorithms". Khosla argued that doctors have no competition with artificial intelligence. Doctors connect with patients, collect a few symptoms, check the body and send prescriptions to the patient. Sometimes (perhaps by accident) this prescription leads to the correct treatment, but doctors are only working on a piece of information available. They wrote that an algorithm could cure it better.

I am a pediatrician and pediatrician in the Bay Area of ​​San Francisco where businessmen like Khosla have been knocking on doctors' doors for years with their pilot technology and software and hardware. I can say to some authority that Khosla is the voice of an outsider who knows what he knows and what is not healthcare.

Yes, artificial intelligence can help us diagnose and treat illnesses. It can present data in a clear and concise way by mobilizing large amounts of data, which can reduce the physician's inaccurate observations.

 Doctors usually do less good tests because of pressure and complications. There is no doubt that for some doctors, whose work is highly diagnosed (for example, a radiologist or a pathologist), artificial intelligence is a constant. The risk can be overwhelming. For example, a decade ago researchers showed that artificial intelligence was better than radiologists in detecting breast cancer.

But for doctors treating children like me, who are seeing 1500 to 2000 patients, artificial intelligence is a boon. I went to medical school to connect with people and make a difference. Today, I often feel like a bookkeeper rather than a doctor who takes information and returns it to patients prescribes medicines and dosages, and orders tests. But the artificial intelligence room is pulling the medical field to the test. It allows me to get to know my patients better, explains how a disease affects them and gives me the opportunity to coach them for better results.

Imagine what artificial intelligence can do in asthma, which is the most common chronic illness in children. Six million American children face it. A total of 14 million children could not attend school in 2013 due to this disease. The treatment cost for this disease costs $ 60 billion a year.

I diagnose asthma with a thumb roll that I have received over time: If you are having a whistling sound and are exposed to asthma, you have this disease. Once the disease is diagnosed, I urge the parent of the patient to take the child's medication. I ask them what makes the disease worse. Is the child in a home where there is smoke?

 I also review his record of how many visits he made to the Emergency Room

But despite the parents 'and patients' recollection and electronic records, this knowledge is lacking. There is no forecast for this. It's not that we don't have statistics but they are dirty. Doctors inboxes are full of data. This data comes in different formats and in different directions: objective information such as laboratory results and important diagnostics to patients on the phone or in emails. 

They are all separate and we spend a great deal of time on doctors just to understand this data. Technology companies and start-up companies want to open data degrees and want to connect directly to consumers via phone, watch, blood pressure cuffs, blood sugar meters. We keep fighting this data and the doctors are tired of it.

How can artificial intelligence fix this? Let's start with the diagnosis. The clinical signs of asthma are simple, but the disease is much more complicated at the molecular and cellular level. The genes, proteins, enzymes, and asthma factors are very diverse.

Now many experts still think of asthma as they do about cancer. Symptoms of the disease vary depending on the location of the tumor and the characteristics of the cell. Imperial College, London Institute for Heart The liver ion adapter studies the connection between asthma and the environment.

He and his team collect biological samples from asthma patients' blood, urine, and lung tissue. This theory means that with such knowledge, patients can be given the best medicine.

Artificial intelligence helps to cure asthma. For many patients, asthma becomes more dangerous when air pollution levels rise, such as last summer when a fire broke out in Northern California. Conscious intelligence tells us to respond better by providing environmental information. ۔

 In 2015, researchers published a study that could help to estimate the number of emergency rooms at Fort Worth Hospital in Dallas related to asthma. They extracted patient record data, as well as pollution data through EPE sensors, Google search, and tweets. In the tweets, he reviewed the terms "whistle sound" or "asthma". Google and Twitter data are associated with the location of the user.

If I had this kind of data, I could say, "Alexa, tell me which of the asthma patients I need to be worried about today." I could reach out to help families affected. And if I had some genetic data like ADAC, I would first diagnose asthma and ask patients to do blood tests and compare them with molecular markers.

The wisdom of saving time like this frees me up to spend more time with patients. One study suggests that asthmatic children took about half of their medication. Artificial intelligence allows me more time to personally interact with children so that I can get better results.

Many questions come up. Are patients willing to share more personal data with us? If artificial intelligence shows that you are well cared for but you or your doctor feel different, will the insurance company accept it? What would happen in this situation if the algorithm skips something or applies incorrectly? Will the doctor or the machine maker be responsible?

Some time ago I saw a colorful picture made by a child in the Journal of American Medical Association. In this picture, the girl's doctor was examining it while the girl kept her eyes on the computer. I hope I will treat this little girl like this soon.