Skip to main content
EJNMMI Research
Fig. 1 | EJNMMI Research

Fig. 1

From: Radiomics predictive modeling from dual-time-point FDG PET Ki parametric maps: application to chemotherapy response in lymphoma

Fig. 1

Five-step flowchart for the present study. (Step I) The Ki map was generated based on DTP imaging of pre-treatment PET data. (Step II) The SUV and Ki map were segmented to define VOI. (Step III) The LIFEx software was used to extract static and dynamic features. (Step IV) The ComBat harmonization was applied to each feature set to correct for the batch effect. A post-treatment PET scan was then used to assess the response to treatment. (Step V) prognostic models were developed to predict treatment outcomes for lymphoma patients and different classification metrics were reported for evaluation of models

Back to article page