Radiomics is a newly developing research field in radiology that enables comprehensive analysis of radiographical images, which process aims to directly connect radiological images and molecular characteristics of neoplasm. Numerous texture analysis is performed with the images and those parameters are then sent into statistical modeling for predicting underlying biological characteristics of the tumor. This technique has become possible with the aid of high performing computational power and in some cases assistance of artificial intelligence. In this short review, the author would like to discuss the current trend of radiomics in gliomas and also comment on the future direction that this research field is desired to pursue.

In this paper, we proposed a guided-MAP（Maximum a Posterior）reconstruction method which introduced image-based dynamic image guided filter（IDIGF）as a prior and investigated an effect of a reduction in statistical noise. The IDIGF uses a static positron emission tomography（PET）image as the guidance image, acquiring the entire data from the start to end of the data acquisition. In the evaluation, dynamic PET simulation data was used, based on the glucose metabolism of [¹⁸F]FDG. As a result, the proposed method improved peak signal-to-noise ratio and structural similarity index in all of the time frames, compared to the other conventional methods. These results indicated that the proposed method can improve the quantitative accuracy of the dynamic PET images.