Home > Riviste > The Quarterly Journal of Nuclear Medicine and Molecular Imaging > Fascicoli precedenti > Articles online first > The Quarterly Journal of Nuclear Medicine and Molecular Imaging 2021 Oct 19

ULTIMO FASCICOLO
 

JOURNAL TOOLS

eTOC
Per abbonarsi PROMO
Sottometti un articolo
Segnala alla tua biblioteca
 

ARTICLE TOOLS

Publication history
Estratti
Permessi
Per citare questo articolo
Share

 

 

The Quarterly Journal of Nuclear Medicine and Molecular Imaging 2021 Oct 19

DOI: 10.23736/S1824-4785.21.03346-X

Copyright © 2021 EDIZIONI MINERVA MEDICA

lingua: Inglese

Volumetric and radiomics assessment of lesions with incidential 18F-FDG uptake in the colon: a retrospective radiomics study

Nurşin AGÜLOĞLU 1 , Emine ACAR AKKAYA 2, Ömer B. BİNİCİER 3

1 Izmir SBU Dr. Suat Seren Chest Diseases and Surgery Hospital, Bornova, Turkey; 2 İzmir Private Kent Hospital, İzmir, Turkey; 3 İzmir Tepecik Training and Research Hospital, İzmir, Turkey


PDF


BACKGROUND: To evaluate the power of volumetric and radiomic tissue data obtained from the images of the lesions in predicting the histopathological diagnosis in patients with incidental colorectal focal FDG uptake detected by 18-F FDG PET/CT imaging.
METHODS: Electronic records of patients who underwent 18F-FDG PET/CT for various malignancies between January 2016 and January 2020 were retrospectively reviewed. 98 lesions of 80 patients with colonoscopic and histopathological results were included in the study. The lesions were divided into 3 groups according to their histopathological diagnosis as benign, premalign and malign. 18F-FDG PET/CT images obtained from the patients were evaluated using LIFEx software. Volumetric and radiomic textural features were obtained by establishing the region of interest (ROI) of the primary tumor. The 18 F FDG PET/CT parameters of the lesions were compared between the groups. In order to evaluate the predictive power of the parameters obtained with 18 F FDG PET/CT, the area under the curve (AUC), sensitivity and selectivity values, negative and positive predictive values were calculated by ROC analysis.
RESULTS: The volumetric and radiomic tissue analysis parameters of the lesions in the malignant group were significantly different when compared to the other groups. Our study showed that SUVmax value can be used as a marker in the diagnosis of pathological malignancy (AUC=0.737 and p=0.001 95% CI (0.619-0.855)). Contrast and Entropy parameters of GLCM matrix, which are radiomic texture features, were found to be used in the prediction of malignancy (AUC=0.685 and AUC=0.692, p=0.041 and p=0.035, respectively. Similarly, parameters of NLGDM, GLRLM and GLZLM matrices were shown to have statistically significant diagnostic values in predicting malignancy. However, no superiority of these parameters over each other in predicting malignancy was observed.
CONCLUSIONS: In our study, the parameters of radiomic textural analysis together with volumetric parameters have been shown to have diagnostic value in noninvasively determining lesion characterization and defining tissue in lesions with incidental colorectal focal FDG uptake detected by 18F-FDG PET / CT imaging.


KEY WORDS: PET/CT; Radiomics; Tissue identification

inizio pagina