The System 'AI Doctor' - The Solution for the patients facing with sepsis

On October 23, scientists created an artificial intelligence system known as AI doctor, capable of treating patients with sepsis by predicting and proposing the best treatment strategy.

The Journal Nature Medicine published the AI system gave more reliable treatment strategies than human. AI Clinician is a system used by medical professionals in order to assist doctors with providing the best treatment for patients.

Sepsis is a disease that causes very high blood pressure to cause the body to have hypoxia and blood, which in turn leads to multiple organ failure and death.

The doctors provide extra fluids so as to raise blood pressure and also maintain the heart beating. The extra fluids are often in the kind of a salt mixture along with vasopressors which helps raise blood pressure and squeeze blood vessels.

After reviewing the records of 130 specialty care settings in the United States for 15 years, the researchers attempted to delve into whether AI could improve outcomes for patients. but take care of the doctor's standards.

These studies will be tested in the UK in special care units. "Sepsis claims six million lives worldwide -- so we desperately need new tools at our disposal to help patients," according to Aldo Faisal at Imperial College London.

Faisal also said that their new AI system can analyze patient data including blood pressure and heart rate and make the best decision. They found that when deciding on a doctor's treatment in line with the proposed AI system, they were given better opportunities to ensure the survival of their patients.

The AI system is designed to help clinicians determine which approach maximizes the patient's chance of survival by evaluating important signs and proposing the best treatment for each individual patients. The system 'AI doctor' analyzed 96,000 medical records of sepsis in specialized care facilities in the United States.

AI Clinicians learn decision-making and problem-solving through increased learning. Robots go through each case of each patient and find the best strategy to apply to each condition of the patient.

Based on 48 variables including important signs and age, conditions precedent and then the system AI doctor will predict the best strategy for each sepsis patient. There is 98% of the decision of AI match or better than the decision of the doctor.

Studies have shown that patients with low mortality have a dose of dilatation that matches the dose that both the physician and the AI doctor give. In contrast, two different decisions reduce the chances of survival.

As a result, the team showed a difference in the individual patients whose dosage would be different. For example, the average patient is treated with too much fluid and less vasodilator with the human doctor’s treatment.

Following to professor at Imperial College London – Anthony Gordon: “The AI Clinician was able to 'learn' from far more patients than any doctor could see in a lifetime. It has learned from 100,000 patients and 'remembered' them all equally whereas doctors are always susceptible to recall bias, where they particularly remember recent cases or unusual cases,"