Although the LDR effects of CT are still controversial, the current problems include the high frequency-use and abuse of CT scans, the increase of radiation dose and accumulative dose in high-accuracy CT, and the poor understanding of carcinogenic risks. The underlying biological basis needs further exploring and the ratio of risks and benefits should be considered.

No abstract

Recently, image-based diagnostic technology has made encouraging and astonishing development. Modern medical care and imaging technology are increasingly inseparable. However, the current diagnosis pattern of Signal-to-Image-to-Knowledge inevitably leads to information distortion and noise introduction in the procedure of image reconstruction (Signal-to-Image). Artificial intelligence (AI) technologies that can mine knowledge from vast amounts of data offer opportunities to disrupt established workflows. In this prospective study, for the first time, we developed an AI-based Signal-to-Knowledge diagnostic scheme for lung nodule classification directly from the CT rawdata (the signal). We found that the rawdata achieved almost comparable performance with CT indicating that we can diagnose diseases without reconstructing images. Meanwhile, the introduction of rawdata could greatly promote the performance of CT, demonstrating that rawdata contains some diagnostic information that CT does not have. Our results break new ground and demonstrate the potential for direct Signal-to-Knowledge domain analysis.

No abstract

No abstract