AI in Hospital Emergency Rooms Needs a Little Humanistic Concern - Interpretation of Medical Artificial Intelligence
The emergency room is the most centralized, most diseased, rescue, and administrative task department in the hospital, and it is also where all the love, persistence, and hopes come together. People-oriented, artificial intelligence find the right use is the temperature.
The telephone in the wiring room rang and the nurse skillfully connected the emergency phone. Immediately afterwards, the hospital mobilized the ambulance to rush to the scene and received the patient and then sent it to the emergency room of the hospital. A paragraph of stories, an individual, began to meet, interweave, express...
A young doctor looks at the CT that was taken - this is from a patient who was admitted to the emergency room because of a severe headache. Once the brain CT shows death and grayness, it can diagnose whether it is a stroke. The key to diagnosis is It is necessary to give results before most of the nerve cells die so that the doctor can treat them in time. In other words, every minute, a part of the brain will die. Losing time means losing the brain.
Here, time is life.
Emergency rush: AI is going to be the first threshold
What is the emergency department? Some patients think that the emergency department is the place where they can immediately see the disease. Some patients use their own feelings as the criteria for criticism. They feel that their illness is urgent and they should go to the emergency department. Therefore, it is not uncommon for people with various misperceptions to visit the emergency department, and the patients who have flocked to it make the emergency department no longer rush. Mark Reiter, the head of the American Academy of Emergency Medicine, once said, "I have a patient who visits 300 times a year."
Faced with this situation, Yan Yan, an industry analyst, believes that artificial intelligence in the emergency room may play a role at this time.
In general, except for those who passed the death line in various green passageways, other patients waited for a long time for emergency treatment. Although each patient is in a hurry, there are still medical priorities. Sometimes doctors are already treating a certain patient, but they are also called to deal with more urgent matters immediately. Of course, this situation is essentially due to the lack of medical resources.
Therefore, for patients who come to the emergency department, AI can first perform a "primary screening." Through visual recognition, big data and other technologies, the patient's continuous vital signs including ECG, breathing, pulse, blood pressure, blood oxygen, and body temperature are detected. In critically ill patients with vital signs, the AI
will immediately report the initial diagnosis to the doctor on duty. If the patient is mildly ill, the waiting time can be calculated and reported to the patient. For some patients who do not need emergency treatment, persuasive diversion can be performed.
Washington State once applied big data to the emergency room. Practice has shown that if doctors can provide pre-examination information, they can reduce the number of emergency medical visits such as gastrointestinal discomfort and headaches, and reduce the rate of emergency visits by 10%.
The advantage of this approach is that it can maximize the reasonable arrangement of medical resources, reduce the friction between pre-screen nurses and appointment patients, and at the same time, because the emergency patients are mostly in the transition period or critical period of disease development, the value of emergency big data is very high. AI makes pre-checks one step ahead of time, and can also label this key data and do deep learning.
Place of non-compliance: AI is a service person
The emergency room is a very busy place. As a result, communication between doctors and patients is often not timely. In the case of emotional instability of the patient's family, it is also prone to misunderstandings and some misunderstandings.
In the eyes of the public, health care workers are “white angels” and it seems natural to serve patients. However, nurses, especially nurses who are most easily misunderstood by patients, are essentially professional technicians, not service personnel. It is unreasonable for professional and technical personnel to assume the service's functions. The patient's cognitive errors lead to the communication between doctors and patients as if they were flammable gunpowder. Therefore, emergency room AI may be able to take this function.
The artificial intelligence service realizes an on-demand and active intelligence. That is, AI captures patient information, builds background data, and doctors' treatment data, constructs demand structure models for patients and their families, and performs data mining and intelligent analysis. In addition to analyzing explicit needs such as patient preferences, it can be further improved. Excavate the hidden needs of the hospital associated with the identity and working life status of patients and their families.
At present, the most common AI medical services are actually intelligent guidance robots. For example, the “Guangxi Xiaomei” robotic guidance robot, which recently landed in Chengdu, Sichuan, can recommend patients to relevant departments according to the information provided by the patients. List of doctors and presentations, etc.
Of course, in the emergency room, it is not only the transmission and feedback data that is needed, but also the multi-dimensional and multi-level perception and active in-depth identification of AI.
It is worth noting that high security is the foundation of intelligent services. The security services here are not only for patients and families, but also for doctors and nurses who need security. In the event of an accident in the emergency room, the family members of the emotionally uncontrolled patient will immediately vent their grief to the doctors and nurses in front of them.
AI should provide personalized security services for doctors and nurses. For example, families of emotionally depressed patients rushed into the operating room and judged the possibility of family members angering and wounding in accordance with techniques such as visual recognition, and connected to the hospital's security warning system. Inform security and even alarm to protect the personal safety of hospitals and nurses.
AI and people: Give you the best companionship and understanding
1.AI keeps you company all the time
The most frequent sentence in the emergency room is -- "Does your family be there?" For patients who are sent to the emergency room and are struggling on the edge of life and death, the companionship of their loved ones means all. However, in many cases, it is difficult for the families of emergency patients to arrive at the first time.
Another obvious situation is that even if a family members escorted to the emergency room, they would often think that they were busy with various procedures, or they would communicate with the doctor without worrying about the patients in their pain. At this moment, the patient is lying in an icy bed, and even in a paralysed emergency room, I am afraid that there will be a great sense of loneliness and anxiety.
It should be noted that this is not because medical staff are irresponsible and they have lost patients there. However, some patients have a special situation. Further observation is required to determine the treatment plan. The medical staff may travel to the operating room and operation room for the patient only to be able to give treatment as soon as possible.
However, the patient's sense of loneliness will not be relieved because of the busyness of the medical staff. On the contrary, busy doctors cannot pay attention to the patient's emotions at any time. At this time, it seems very important to have a patient with a "big white".
Amazon is developing the companionship and emotional intelligence of its own robots. At the end of 2017, Alexa's chief scientist, Rohit Prasad, talked about how Amazon analyzed the user's emotional state based on the voice records collected each time the user interacts with Alexa-supported devices. In the future, AIexa may be able to recognize the subtle emotions in your words. This means that Amazon's robots can emotionally comfort patients and reduce human loneliness in the emergency room.
Because many emergency patients are often unable to speak, in addition to "understand", "reading" is also particularly important, which requires the maturity of the machine vision, face recognition technology. AI must accurately recognize facial expressions and movements, determine a person's emotional and emotional changes, and changes in concentration. For example, if a patient is lying in a bed, if the machine can recognize that he is in a state of tension or sadness, it can bring positive results to the doctor's treatment.
2. Human beings give you the greatest sense of security
AI can accompany people, but in the face of illness, giving patients the greatest sense of security may be the most critical. Suddenly, is it because your retina has a problem or is your brain nerves being squeezed? Is it because of overuse of an overused or overused brain? Asking a doctor why you are constantly asking for information is a way for patients to get information, and the more you know about your condition, the more patients The more you have in your heart.
The AI's deep learning system does not have any explanatory power. In fact, the more powerful the system for deep learning, the more opaque its structure becomes. As more data features are extracted, the diagnosis of AI becomes more and more accurate. But why these features will be selected from millions of data remains an ambiguity in artificial intelligence.
Therefore, the most irreplaceable place for human doctors is not to "know what they are" or "know how they are". It is neither to grasp the fact of the disease (WHAT) nor to understand how the condition is formed (HOW), but to the third area of
knowledge: Know why (WHY).
In the emergency room, the patient's ultimate sense of security still depends on the person. After the AI
has taken more responsibility to the upper body, the emergency doctors do not have to deal with wasted medical resources or patients and their family members who are not rational when treating patients. Their responsibilities will increasingly be biased towards giving families and patients a professional perspective. Can ease its anxious explanation.
The emergency room is the most centralized, most diseased, rescue, and administrative task department in the hospital, and it is also where all the love, persistence, and hopes come together. People-oriented, artificial intelligence find the right use is the temperature.