Artificial intelligence predicts dangerous lung attacks a week before it occurs

Researchers have succeeded in using artificial intelligence to analyze urine samples for patients, and to predict the time of exacerbation of symptoms of chronic obstructive pulmonary disease. These results, published in the ERJ Open Research Patrol, provide great potential to improve treatment and prevention methods for patients with this chronic disease, and during the study, patients performed a simple daily test using urinary test strips, and they sent their results via their cell phones. Using artificial intelligence to analyze this data, researchers make the prediction of the decline of symptoms a week before it takes place, allowing proactive steps, such as adjusting the treatment to reduce or prevent the symptoms. Chronic pulmonary obstruction of severe and long -term lung disease, and according to the World Health Organization, chronic pulmonary obstruction is the third leading cause of death worldwide, and the aggravation of symptoms includes breathing and coughing. Chronic pulmonary obstruction occurs with chronic pulmonary obstruction attacks when the patient develops a serious condition that needs additional treatment at home or in hospital, and current treatments depend on the response to acute disease. The researchers say: “It would be better if we could predict an attack before it occurs, and then assign the treatment to avoid a seizure or reduce its effect.” The researchers began to analyze the urine samples a group of 55 people with chronic pulmonary blockage, with the aim of determining the changes in the urine composition that precedes the aggravation of the symptoms, and this has led to the determination of a group of “important indicators” that usually change when the disease is annoyed. The researchers measured the levels of a group of important indicators in urine samples for patients, including substances such as protein, or chemical molecules that indicate changes in the body due to the aggravation of the disease. These indicators include NGAL, a protein associated with inflammation, and Timp1 protein housers of the destroyed enzymes of tissues, interactive protein and fibrinogen, a protein participating in blood clots and infection, CC16, a protein being extracted from the protein Stimulate, in addition to protein stimulates enzymes of tissue-2. Artificial intelligence and medical diagnosis have used researchers some kind of artificial intelligence called the artificial nerve network to analyze changes in important indicators and expect the aggravation of symptoms. The results of the analysis showed that artificial intelligence can predict symptoms about seven days before any clinical signs appeared. Urine used, as a way to collect biomedies, is an effective tool in medical and research studies, due to its many benefits. One of the most prominent of these benefits is the ease of collecting the sample, as participants can collect urine daily, without the need for complicated medical intervention, which facilitates continuous succession, especially in remote places. Urine also contains a wide variety of chemical and biological compounds that can be analyzed to detect changes in the body, such as hormones, protein and amino acids, which provide comprehensive information on patient health information. It follows the biological functions of patients and the process of urine collection is painful compared to other methods such as the collection of blood samples or tissues, making them more comfortable for participants, and urine provides the ability to monitor health changes over a long period, which allows an accurate image of continuous physiological changes. The collection and analysis of the urine is cheaper than other methods, such as advanced medical examinations or medical imaging, making it an ideal choice in research that needs to collect data from a large number of participants. Therefore, urine samples analysis is an ideal tool in many scientific and medical studies that require a continuous follow -up of biological and physiological functions for participants. The study is the first of its kind to examine many materials in urine samples from people with chronic obstructive pulmonary disease before the symptoms worsen and at the time of the stability of the condition. The researchers found that a small group of these materials could indicate the beginning of the seizure, and then developed a forecasting tool using artificial intelligence to analyze these materials. The researchers pointed out that more work is needed to improve the artificial intelligence algorithm using data from a larger group of patients. They also hope that it enables the development of tests aimed at patients with chronic pulmonary obstruction, so that the algorithm learns what is “of course” for each person, and predicts aggravation attacks.