AI Analysis of Social Media Data to Help with Treatment of Long COVID
The Vector Institute for Artificial Intelligence, in collaboration with Roche Canada, Deloitte, and TELUS, released a study utilizing artificial intelligence to analyze social media data for insights into long COVID. By applying natural language processing techniques to over 460,000 Twitter posts from individuals self-reporting long COVID, the framework identified more than 100 symptoms. The most frequent symptoms included fatigue, pain, brain fog, anxiety, and headache, along with patterns of co-occurring symptoms. This research aims to assist researchers and clinicians in understanding the frequency and nature of long COVID symptoms, thereby improving prognosis and treatment strategies. Given the lack of a single diagnostic method or treatment for long COVID, these AI-driven insights provide crucial first-hand data from patients' daily lives outside clinical settings. The study highlights the potential of social media as a valuable resource for public health monitoring and demonstrates how cross-sector collaboration between AI specialists, corporate entities, and healthcare providers can address complex medical challenges. Furthermore, the developed framework has broader applications for tracking disease symptoms and drug side effects in other medical areas.
Wire timeline
AI Analysis of Social Media Data to Help with Treatment of Long COVID
The Vector Institute for Artificial Intelligence, in collaboration with Roche Canada, Deloitte, and TELUS, released a study utilizing artificial intelligence to analyze social media data for insights into long COVID. By applying natural language processing techniques to over 460,000 Twitter posts from individuals self-reporting long COVID, the framework identified more than 100 symptoms. The most frequent symptoms included fatigue, pain, brain fog, anxiety, and headache, along with patterns of co-occurring symptoms. This research aims to assist researchers and clinicians in understanding the frequency and nature of long COVID symptoms, thereby improving prognosis and treatment strategies. Given the lack of a single diagnostic method or treatment for long COVID, these AI-driven insights provide crucial first-hand data from patients' daily lives outside clinical settings. The study highlights the potential of social media as a valuable resource for public health monitoring and demonstrates how cross-sector collaboration between AI specialists, corporate entities, and healthcare providers can address complex medical challenges. Furthermore, the developed framework has broader applications for tracking disease symptoms and drug side effects in other medical areas.
Vector Institute for Artificial Intelligence