COVID‑19 pathology imaging: A one-year perspective

Area of Science:

• Radiology
• Medical Imaging
• Artificial Intelligence in Medicine

Background:

• Coronavirus disease 2019 (COVID-19) emerged in Wuhan, China, in December 2019, leading to a global pandemic declared by the WHO.
• COVID-19 presents with nonspecific symptoms including respiratory distress, fever, cough, and loss of smell/taste, with diagnosis confirmed by RT-PCR.
• Medical imaging plays a crucial role in assessing disease extent and potential complications, extending beyond the respiratory system to include cardiac, vascular, and neurological damage.

Purpose of the Study:

• To present a comprehensive overview of the radiographic features of COVID-19 as reported in published literature.
• To summarize common pathologies and classifications of COVID-19 findings in chest radiography (CXR) and computed tomography (CT).
• To explore the utility of lung ultrasound (LUS) and the emerging role of artificial intelligence (AI) in COVID-19 detection and management.

Main Methods:

• Systematic review of published scientific literature focusing on chest radiography (CXR), computed tomography (CT), and lung ultrasound (LUS) in COVID-19 patients.
• Analysis and summarization of common radiographic pathologies and classification systems for COVID-19.
• Review of AI algorithms developed for COVID-19 detection and discussion of their role in the pandemic.

Main Results:

• Common pathologies identified in CXR and CT include ground-glass opacities and consolidations, with specific patterns associated with COVID-19.
• CT imaging is highly sensitive for detecting COVID-19 pneumonia, while CXR can show characteristic findings, especially in moderate to severe cases.
• AI algorithms show promise in automated detection and severity assessment of COVID-19 from medical images, aiding radiologists.

Conclusions:

• Medical imaging, particularly CT and CXR, is indispensable for diagnosing, staging, and monitoring COVID-19 and its complications.
• Lung ultrasound (LUS) offers a portable alternative for evaluating lung involvement in COVID-19.
• Artificial intelligence holds significant potential to enhance the efficiency and accuracy of radiological assessments in the current and future pandemics.