Nuclear Medicine Communications · 2012 · 69 citations · 18 references
MedicineNuclear MedicineOncologyBiomedical ImagingDiagnostic ImagingDiagnostic RadiologyEsophageal CancerPositron Emission TomographySquamous Cell CarcinomaTumor CharacterizationTumor GradingTumor ImagingCancer ImagingOncologic Pet ImagingPet QuantificationPreoperative Tumor CharacterizationRadiologic ImagingTextural PropertiesTumor Heterogeneity AssessmentTumor Location Assessment
OBJECTIVE: To explore the relationship of a new PET image parameter, (18)F-fluorodeoxyglucose ((18)F-FDG) uptake heterogeneity assessed by texture analysis, with maximum standardized uptake value (SUV(max)) and tumor TNM staging. MATERIALS AND METHODS: Forty consecutive patients with esophageal squamous cell carcinoma were enrolled. All patients underwent whole-body preoperative (18)F-FDG PET/CT. Heterogeneity of intratumoral (18)F-FDG uptake was assessed on the basis of the textural features (entropy and energy) of the three-dimensional images using MATLAB software. The correlations between the textural parameters and SUV(max), histological grade, tumor location, and TNM stage were analyzed. RESULTS: Tumors with higher SUV(max) were seen to be more heterogenous on (18)F-FDG uptake. Significant correlations were observed between T stage and SUV(max) (r(s)=0.390, P=0.013), entropy (rs=0.693, P<0.001), and energy (r(s)=-0.469, P=0.002). Correlations were also found between SUV(max), entropy, energy, and N stage (r(s)=0.326, P=0.04; r(s)=0.501, P=0.001; r(s)=-0.413, P=0.008). The American Joint Committee on Cancer stage correlated significantly with all metabolic parameters. The receiver-operating characteristic curve demonstrated an entropy of 4.699 as the optimal cutoff point for detecting tumors above stage II(b) with an areas under the ROC curve of 0.789 (P<0.001). CONCLUSION: This study provides initial evidence for the relationship between the new parameter of tumor uptake heterogeneity and the commonly used simplistic parameter of SUV and tumor stage. Our findings suggest a complementary role of these parameters in the staging and prognosis of esophageal squamous cell carcinoma.
18
Ahmedin Jemal, Freddie Bray, Jacques Ferlay et al. · CA A Cancer Journal for Clinicians · 2011 · 55K citations · Full text
Florent Tixier, Catherine Cheze Le Rest, Mathieu Hatt et al. · Journal of Nuclear Medicine · 2011 · 684 citations · Full text
Medicine, Nuclear Medicine, Oncology +17
Exploring feature-based approaches in PET images for predicting cancer treatment outcomes
Perry W. Grigsby, Aditya Apte, Elizabeth Kidd et al. · Pattern Recognition · 2008 · 471 citations · Full text