Search

To examine the prognostic significance of pretreatment 2-deoxy-2-[18F]fluoro-d-glucose ([18F]-FDG) positron emission tomography (PET)-based radiomic features using a machine learning approach in patients with endometrial cancers. Included in this retrospective study were 53 patients with endometrial cancers who underwent [18F]-FDG PET/X-ray computed tomography (CT) before treatment. Since two different PET scanners were used, post-reconstruction harmonization was performed for all PET parameters using the ComBat harmonization method. Four clinical (age, histological type, stage, and treatment method) and 40 [18F]-FDG PET-based radiomic features were ranked, and a subset of useful features was selected based on the decrease in the Gini impurity in terms of associations with disease progression. The machine learning algorithms (random forest, neural network, k-nearest neighbors (kNN), naive Bayes, logistic regression, and support vector machine) were compared using the areas under the receiver operating characteristic curve (AUC) and validated by the random sampling method. Progression-free survival (PFS) and overall survival (OS) were assessed by the Cox regression analysis. The five best predictors of disease progression were coarseness, gray-level run length nonuniformity, stage, treatment method, and gray-level zone length nonuniformity. The kNN model obtained the best performance classifier for predicting the disease progression (AUC =0.890, accuracy =0.849, F1 score =0.848, precision =0.857, and recall =0.849). Coarseness which was the first ranked radiomic feature was selected for survival analyses, and only coarseness remained as a significant and independent factor for both PFS (hazard ratios (HR), 0.65; 95 % confidence interval [CI], 0.49–0.86; p=0.003) and OS (HR, 0.52; 95 % CI, 0.36–0.76; p

This article was update to correct the spelling of Takashi Yoshiura's name; it is correct as displayed here.

To examine whether the machine learning (ML) analyses using clinical and pretreatment 2-deoxy-2-[This retrospective study included 100 patients with hypopharyngeal cancer who underwent [The five most important features for predicting disease progression were UICC stage, N stage, gray level co-occurrence matrix entropy (GLCM_Entropy), gray level run length matrix run length non-uniformity (GLRLM_RLNU), and T stage. Patients who experienced disease progression displayed significantly higher UICC stage, N stage, GLCM_Entropy, GLRLM_RLNU, and T stage than those without progression (each, p0.001). In both cohorts, the logistic regression model constructed by these 5 features was the best performing classifier (training: AUC=0.860, accuracy=0.800; validation: AUC=0.803, accuracy=0.700). In the logistic regression model, 5-year PFS was significantly higher in patients with predicted non-progression than those with predicted progression (75.8% vs. 8.3%, p0.001), and this model was only the independent factor for PFS in multivariate analysis (hazard ratio = 3.22; 95% confidence interval = 1.03-10.11; p=0.045).The logistic regression model constructed by UICC, T and N stages and pretreatment [

To examine the prognostic significance of pretreatment 2-deoxy-2-[Included in this retrospective study were 53 patients with endometrial cancers who underwent [The five best predictors of disease progression were coarseness, gray-level run length nonuniformity, stage, treatment method, and gray-level zone length nonuniformity. The kNN model obtained the best performance classifier for predicting the disease progression (AUC =0.890, accuracy =0.849, F1 score =0.848, precision =0.857, and recall =0.849). Coarseness which was the first ranked radiomic feature was selected for survival analyses, and only coarseness remained as a significant and independent factor for both PFS (hazard ratios (HR), 0.65; 95 % confidence interval [CI], 0.49-0.86; p=0.003) and OS (HR, 0.52; 95 % CI, 0.36-0.76; p0.001) at multivariate Cox regression analysis.[

To examine whether the heterogeneous texture parameters in primary tumor can predict prognosis of patients with non-small cell lung cancer (NSCLC) received surgery after 2-deoxy-2-[18F]fluoro-D-glucose ([18F]FDG) positron emission tomography (PET)/X-ray computed tomography (CT). This retrospective study included 55 patients with NSCLC who underwent [18F]FDG-PET/CT before surgery from January 2011 and December 2015. SUV-related (SUVmax and SUVmean), volumetric (metabolic tumor volume [SUV ≥ 2.5], and total lesion glycolysis) and texture parameters (local parameters; entropy, homogeneity, and dissimilarity and regional parameters; intensity variability [IV], size-zone variability [SZV], and zone percentage [ZP]) were obtained. Tumor size, TNM stage, SUV-related, volumetric, and texture parameters were compared between the patients with progression and without progression using Mann-Whitney’s U or χ2 test and progression-free survival (PFS) and prognostic significance were assessed by Kaplan-Meier method and Cox regression analysis, respectively. Nineteen patients eventually showed progression, and 36 patients were alive without progression during clinical follow-up (median follow-up PFS; 23 months [range, 1–71]). The patients with progression showed significantly larger tumor size (p

To examine whether the heterogeneous texture parameters in primary tumor can predict prognosis of patients with non-small cell lung cancer (NSCLC) received surgery after 2-deoxy-2-[This retrospective study included 55 patients with NSCLC who underwent [Nineteen patients eventually showed progression, and 36 patients were alive without progression during clinical follow-up (median follow-up PFS; 23 months [range, 1-71]). The patients with progression showed significantly larger tumor size (p 0.001), higher IV (p = 0.010), and higher SZV (p = 0.007) than those without progression. PFS was significantly shorter in patients with large tumor size (p = 0.008), high T stage (p = 0.009), high stage (p = 0.013), high IV (p = 0.012), and high SZV (p = 0.015) at univariate analysis. At multivariate analysis, stage (hazard ratio [HR] 1.62, p = 0.035) and IV (hazard ratio 6.19, p = 0.048) were only remained independent predictors for PFS.The regional heterogeneity texture parameters IV and SZV can predict tumor progression, and IV has the potential to predict prognosis of surgically treated NSCLC patients.