Artificial intelligence applications for oncological positron emission tomography imaging

Positron emission tomography (PET), a functional and dynamic molecular imaging technique, is generally used to reveal tumors’ biological behavior. Radiomics allows high-throughput extraction of multiple features from images with artificial intelligence (AI) approaches develops rapidly worldwide. Quantitative objective medical have been explored recognize reliable biomarkers, the development PET radiomics. This paper will review current clinical exploration PET-based classical machine learning deep methods, including disease diagnosis, prediction histological subtype, gene mutation status, tumor metastasis, relapse, therapeutic side effects, intervention evaluation prognosis. The applications AI in oncology be mainly discussed. image-guided biopsy or surgery assisted by introduced as well. aims present methods for imaging, which may offer important details further studies. Relevant precautions are put forward future research directions suggested.