Artificial intelligence has learned to detect diabetes from chest X-rays
A new artificial intelligence model aimed at early detection of diabetes based on the results of chest X-rays has been created in the USA. The service is able to distinguish signs of diabetes even in patients who do not meet the criteria of a high-risk group.Emory University (USA) has created a model using artificial intelligence (AI) that can help doctors detect diabetes in its early stages and prevent its complications. In some cases, chest X-rays warned of a high risk of diabetes three years before a patient was diagnosed, Healthcare in Europe reported, citing a study published in Nature Communications.
Current guidelines suggest screening all patients for type 2 diabetes between the ages of 35 and 70 who have a body mass index (BMI) ranging from overweight to obese. However, many studies indicate that this strategy misses a significant number of cases, especially among racial/ethnic minorities, for whom BMI is a less effective predictor of diabetes risk. Patients with undiagnosed disease have a higher likelihood of complications, including irreversible organ damage and even death.
Yet every year, millions of Americans get chest X-rays for chest pain, breathing problems, injuries, or before surgery. Emory University Hospital alone takes an average of about 200,000 X-rays each year. The images become part of the patient's medical record and can be analyzed later for diabetes or other conditions.
The researchers intend to find out how well chest X-rays can help diagnose other conditions such as vascular disease, congestive heart failure and chronic obstructive pulmonary disease (COPD).