Artificial Intelligence (AI)-derived computer-aided detection (CAD) software may be a game-changer in the field of lung cancer screening. These technologies can potentially enhance the detection and measurement of lung nodules in CT scan images, making them a cost-effective solution. However, more evidence is needed to ascertain which technologies are the most clinically and cost-effective.
There is little clinical-effectiveness evidence available on any of the individual technologies which means that more research is needed to fully understand the potential benefits and drawbacks of these technologies. For example, the software could potentially identify more people with benign lung nodules, leading to unnecessary anxiety and CT surveillance.
Despite these uncertainties, the potential benefits of AI-derived CAD software cannot be ignored. Not only can it potentially increase the detection of lung nodules, but it can also improve reporting of nodule characteristics, assess the growth of lung nodules, and reduce the time to review and report CT scans.
NICE encourages centres using AI-derived CAD software as part of targeted lung cancer screening to continue generating evidence and sharing their findings. This will help to realise the full potential of these technologies and facilitate comparisons between different software.
Recommendations for further research on the effect of using the software on clinical decision making and the effect of using the software on scan review and reporting time, are listed in the guideline.
Reference url