ABSTRACT
PURPOSE: We propose three support vector machine (SVM) classifiers, using pre-and post-contrast T2 fluid-attenuated inversion recovery (FLAIR) subtraction and/or pre-and post-contrast T1WI subtraction, to differentiate treatment-related effects (TRE) from glioma recurrence. MATERIALS AND METHODS: Fifty-six postoperative high-grade glioma patients with suspicious progression after radiotherapy and chemotherapy from two centers were studied. Pre-and post-contrast T1WI and T2 FLAIR were collected. Each pre-contrast image was voxel-wise subtracted from the co-registered post-contrast image. Dataset was randomly split into training, and testing on a 7:3 ratio, accordingly subjected to a five fold cross validation. Best feature subsets were selected by Pearson correlation coefficient and recursive feature elimination, whereupon a radiomics classifier was built with SVM. The discriminating performance was assessed with the area under receiver-operating characteristic curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). RESULTS: In all, 186 features were extracted on each subtraction map. Top nine T1WI subtraction features, top thirteen T2 FLAIR subtraction features and top thirteen combination features were selected to build optimal SVM classifiers accordingly. The accuracies/AUCs/sensitivity/specificity/PPV/NPV of SVM based on sole T1WI subtraction were 80.00%/80.00% (CI: 0.5370-1.0000)/100%/70.00%/62.50%/100%. Those results of SVM based on sole T2 FLAIR subtraction were 86.67%/84.00% (CI: 0.5962-1.0000)/100%/80%/71.43%/100%. Those results of SVM based on both T1WI subtraction and T2 FLAIR subtraction were 93.33%/94.00% (CI: 0.7778-1.0000)/100%/90%/83.33%/100%, respectively. CONCLUSION: Pre- and post-contrast T2 FLAIR subtraction provided added value for diagnosis between recurrence and TRE. SVM based on a combination of T1WI and T2 FLAIR subtraction maps was superior to the sole use of T1WI or T2 FLAIR for differentiating TRE from recurrence. The SVM classifier based on combination of pre-and post-contrast subtraction T2 FLAIR and T1WI imaging allowed for the accurate differential diagnosis of TRE from recurrence, which is of paramount importance for treatment management of postoperative glioma patients after radiation therapy.
ABSTRACT
OBJECTIVE: To explore the significance of assessing asthma control by high-resolution computed tomography (HRCT) and biological markers in induced sputum. METHODS: Forty-eight patients with asthma (asthma group) and 10 healthy subjects (control group) were retrospectively analyzed. The asthma patients were divided into 4 groups based on severity: 6 with near-fatal attacks, 12 with severe, 14 with moderate and 16 with mild asthma. These patients received step therapy for 6 months based on the guidelines for the prevention and treatment of asthma. After achieving asthma control or partial control, HRCT, lung function and cytokine levels in induced sputum were measured. The ratio of wall area to total airway area (WA%), the ratio of 2 airway wall thickness to outer diameter (2T/D) and lung densities in both the inspiratory and expiratory phases were measured. Matrix metalloproteinase-9 (MMP-9), tissue inhibitor of metalloproteinases-1 (TIMP-1), and transformation growth factor-ß(1) (TGF-ß(1)) levels in the sputum were assessed by enzyme-linked immunosorbent assay. RESULTS: There were significant differences in forced vital capacity and forced expiratory volume in 1 second as the percentage of predicted value (FVC% and FEV(1)%, respectively), the ratio of FEV(1)/FVC, and diffusing capacity of the lung for carbon monoxide (D(LCO)) among groups (F = 5.526, 15.064, 16.326, 2.945, respectively, P < 0.05). Sputum levels of MMP-9, TIMP-1 and TGF-ß(1) were significantly increased in the near-fatal asthma, severe asthma, moderate asthma and mild asthma groups [MMP-9: (80 ± 16), (70 ± 9), (59 ± 6), and (52 ± 7) µg/L, respectively; TIMP-1: (212 ± 95), (258 ± 167), (28 ± 51), and 98 ± 60 µg/L, respectively; TGF-ß(1): (586 ± 81), (513 ± 54), (401 ± 45) and (351 ± 57) µg/L, respectively]compared with the control group [MMP9: (46 ± 5) µg/L; TIMP: (19 ± 13) µg/L; and TGF-ß(1): (258 ± 29) µg/L]. These parameters were progressively increased in the asthma groups with the severity of disease (F = 11.179, 49.914, 9.286, respectively, P < 0.05). The ratio of MMP-9/TIMP-1 in sputum was decreased in the near-fatal attack, severe, moderate and mild asthma groups (0.50 ± 0.28, 0.34 ± 0.13, 0.53 ± 0.22, and 0.87 ± 0.75, respectively) compared with the control group (2.93 ± 1.13). The MMP-9/TIMP-1 ratio in the severe asthma group was lowest among the asthma groups (F = 43.335, P < 0.05). 2T/D and WA% were higher in both the near-fatal asthma group (0.51 ± 0.01 and 0.75 ± 0.01, respectively) and the severe asthma group (0.53 ± 0.03 and 0.77 ± 0.03, respectively) as compared to the moderate asthma group (0.43 ± 0.04 and 0.67 ± 0.04, respectively) or the mild group (0.42 ± 0.04 and 0.66 ± 0.04, respectively). 2T/D and WA% were higher in the asthma groups than in the control group (0.35 ± 0.03 and 0.57 ± 0.04, respectively), (F = 40.224, 41.294, respectively, P < 0.05). Lung densities in both the inspiratory and expiratory phases were lower in the near-fatal attack group as compared to those in the other asthma groups or the control group; and the lung density differences between the two phases in the near-fatal attack group were smaller than those in the other asthma groups or the control group (F = 5.048, 13.247, 11.541, respectively, P < 0.05). 2T/D and WA% were correlated positively with MMP-9, TIMP-1 and TGF-ß(1) levels, but negatively with the MMP-9/TIMP-1 ratio, respectively. CONCLUSIONS: HRCT and biological markers in induced sputum could be used to accurately evaluate asthma control. These findings suggest that the severity of asthma, especially, near-fatal attack of asthma, is correlated not only with the degree of airway remodeling, but also with the degree of air trapping.
