63 ], which is prospective in its validation. This distinction highlights one of the key limitations of the research landscape, as retrospective studies may introduce biases that can affect the interpretation of the findings." width="100%" height="100%">
Journal: Cancers
Article Title: Enhancing Lymphoma Diagnosis, Treatment, and Follow-Up Using 18 F-FDG PET/CT Imaging: Contribution of Artificial Intelligence and Radiomics Analysis
doi: 10.3390/cancers16203511
Figure Lengend Snippet: Summary of radiomics and AI-based radiomics studies conducted on lymphoma using 18 F-FDG-PET/CT images. All studies included in this Table are retrospective, except Ceriani et al. [ 63 ], which is prospective in its validation. This distinction highlights one of the key limitations of the research landscape, as retrospective studies may introduce biases that can affect the interpretation of the findings.
Article Snippet: Lue et al. [ ] , 2020 , 35 , Response to Therapy, OS, PFS , Retrospective , CGITA (MATLAB 2012a) , HL , Statistical analysis , Treatment Response Prediction: HIR_GLRM PET : OR = 36.4, p = 0.014 RLNU_GLRM CT : OR = 30.4, p = 0.020 PFS: INU_GLRM PET : HR = 9.29, p = 0.023 Wavelet SRE_GLRM CT : HR = 18.40, p = 0.012 OS: ZSNU_GLSZM PET : HR = 41.02, p = 0.001 A prognostic stratification model: PFS ( p < 0.001), OS ( p < 0.001). , Predictive features for treatment response HIR_GLRMPET and RLNU_GLRMCT. Independent predictive features for survival: ZSNU(GLSZM PET ), INU(GLRM PET ), and wavelet SRE (GLRM CT ).
Techniques: Biomarker Discovery, Introduce, Extraction, Software, Diagnostic Assay, Selection, Imaging, Positron Emission Tomography, Gene Expression, Amplification, Marker, Transformation Assay