Key information
- Nearly 23 percent of U.S. adults may have symptoms of long COVID, a significantly higher prevalence than previously estimated.
- The AI-powered tool can identify subtle symptoms of long COVID and differentiate between symptoms caused by existing medical conditions or long COVID.
- This technology aims to improve diagnostic capabilities by accurately identifying potential indicators of long COVID and helping doctors navigate complex symptom presentations.
A recent study using artificial intelligence (AI) found that nearly 23 percent of American adults may have symptoms of long COVID. This finding, published in the journal Med, far exceeds previous estimates of 7 percent. The researchers attribute this higher prevalence to their artificial intelligence tool, designed to analyze electronic medical records and identify subtle symptoms of long COVID.
Long COVID manifests itself with a range of symptoms impacting different body systems, including fatigue, persistent coughs, heart problems and cognitive difficulties often referred to as “brain fog”. These symptoms usually appear weeks or months after the initial COVID-19 infection has cleared. The AI tool specifically looks for symptoms unexplained by medical history, lasting at least two months, and occurring following a confirmed COVID infection.
Improved diagnostic capabilities
This technology aims to clarify the diagnostic process by highlighting potential long-term indicators of COVID that might otherwise remain in the shadows. For example, AI can differentiate shortness of breath attributed to pre-existing conditions such as asthma or heart failure from that potentially due to a long COVID infection.
The study’s lead researcher, Dr. Alaleh Azhir, notes that this AI-powered tool can significantly help doctors navigate complex symptom presentations and patient histories, ultimately helping account to establish more precise diagnoses. The researchers intend to make this AI technology publicly available to enable wider adoption by healthcare professionals.
In Belgium and Europe
In Belgium and across Europe, the results of this study are of great relevance, as healthcare systems face the challenges of diagnosing and treating long-term COVID. The introduction of AI-driven diagnostic tools, such as the one developed in this research, could streamline the detection of subtle and complex symptoms often associated with the disease. By integrating these tools into existing electronic medical records systems such as Vitalink, Belgian healthcare providers could improve diagnostic accuracy and reduce the workload of overburdened clinicians.
Furthermore, thanks to the initiatives already in place in Belgium for the rehabilitation of people suffering from long-term COVID and its strong research infrastructure, it is possible to adapt and validate such AI solutions for the local population, thereby promoting a more effective and targeted response to long-term COVID.
If you want access to all articles, subscribe here!
