[Chest radiological lesions in COVID-19 : from classical imaging to artificial intelligence]

Area of Science:

• Medical Imaging
• Infectious Diseases
• Pulmonology

Background:

• The COVID-19 pandemic necessitated a reevaluation of diagnostic strategies for respiratory illnesses.
• Delays in Reverse Transcription Polymerase Chain Reaction (RT-PCR) testing and platform overload highlighted the need for alternative diagnostic methods.
• Thoracic imaging emerged as a vital tool for patient management during the pandemic.

Purpose of the Study:

• To assess the role of thoracic imaging in the diagnostic approach for patients with COVID-19.
• To describe the common imaging findings associated with SARS-CoV-2 infection.
• To explore the utility of advanced imaging techniques and artificial intelligence in clinical decision-making.

Main Methods:

• Systematic review of common thoracic imaging findings in COVID-19 patients.
• Description of typical Computed Tomography (CT) scan characteristics.
• Exploration of artificial intelligence applications in interpreting imaging data.

Main Results:

• Thoracic imaging, particularly CT, demonstrated significant utility in aiding the diagnosis and management of COVID-19.
• Characteristic CT findings include bilateral, peripheral, posterior, and multi-lobar ground-glass opacities.
• Imaging can detect significant lesions in asymptomatic patients, sometimes preceding symptom onset.

Conclusions:

• Thoracic imaging is an indispensable tool for diagnosing and managing COVID-19, especially when RT-PCR results are delayed.
• The characteristic imaging patterns aid clinicians in timely patient care.
• Artificial intelligence holds promise for enhancing diagnostic accuracy and clinical decision support in respiratory illness.