Smartphone-based detection of COVID-19 and associated pneumonia using thermal imaging and a transfer learning...

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Pulmonology

Background:

• Traditional COVID-19 pneumonia diagnosis relies on hospital-based imaging like X-rays and CT scans.
• These methods are inaccessible for remote or widespread screening.
• Thermal cameras offer a portable and cost-effective alternative for medical monitoring.

Purpose of the Study:

• To develop and evaluate a smartphone-based application for COVID-19 detection using thermal imaging.
• To assess the feasibility of using noninvasive thermal imaging for early screening of COVID-19 pneumonia.

Main Methods:

• Development of a smartphone application integrating thermal imaging capabilities.
• Utilized thermal images of the human back for data acquisition.
• Employed a deep learning algorithm for image analysis and classification.

Main Results:

• The developed deep learning model achieved a sensitivity of 88.7% for COVID-19 detection.
• The model demonstrated a specificity of 92.3%.
• The system successfully utilized smartphone-connected thermal cameras.

Conclusions:

• Smartphone-based thermal imaging presents a promising noninvasive approach for COVID-19 detection.
• This technology could be valuable for primary screening of COVID-19 and associated pneumonia in diverse settings.
• Further research can explore broader clinical applications of thermal imaging in respiratory illness detection.