Artificial Intelligence may help to differentiate COVID-19 pneumonia from other forms of pneumonia when used in high-prevalence and symptomatic populations, according to a new JRC review study.
Differentiating accurately COVID-19 pneumonia from pneumonia of other origins remains challenging. Radiologic differences are subtle, especially in asymptomatic patients and those with early onset of symptoms.
Artificial Intelligence (AI) has great potential to improve health care, with the research community putting high expectations on AI tools for COVID-19 detection in medical imaging.
The JRC published a systematic review comparing added value of Artificial Intelligence versus human readers to detect COVID-19 by medical imaging.
The study shows that AI systems or AI-supported human readings show less performance variability compared to readings of radiologists without AI support. AI may therefore support the differentiation of COVID-19 pneumonia from other forms of pneumonia when used in high-prevalence and symptomatic populations.
However, inconsistencies related to study design, reporting of data, areas of risk of bias, as well as limitations of statistical analyses complicate clear conclusions.
The JRC performed a systematic review of 1270 peer-reviewed publications that used AI for the evaluation of lung imaging to support the detection of COVID-19. To our knowledge, this is the first published systematic review to date on this focused topic.
Artificial intelligence for imaging-based COVID-19 detection: Systematic review comparing added value of AI versus human readers
- Publication date
- 10 January 2022