Prediction of dose deposition matrix using voxel features driven machine learning approach.

OBJECTIVES: A dose deposition matrix (DDM) prediction method using several voxel features and a machine learning (ML) approach is proposed for plan optimization in radiation therapy. METHODS: Head and lung cases with the inhomogeneous medium are used as training and testing data. The prediction model is a cascade forward backprop neural network where the input is the features of the voxel, including 1) voxel to body surface distance along the beamlet axis, 2) voxel to beamlet axis distance, 3) voxel density, 4) heterogeneity corrected voxel to body surface distance, 5) heterogeneity corrected voxel to beamlet axis, and (6) the dose of voxel obtained from the pencil beam (PB) algorithm. The output is the predicted voxel dose corresponding to a beamlet. The predicted DDM was used for plan optimization (ML method) and compared with the dose of MC-based plan optimization (MC method) and the dose of pencil beam-based plan optimization (PB method). The mean absolute error (MAE) value was calculated for full volume relative to the dose of the MC method to evaluate the overall dose performance of the final plan. RESULTS: For patient with head tumor, the ML method achieves MAE value 0.49 × 10-4 and PB has MAE 1.86 × 10-4. For patient with lung tumor, the ML method has MAE 1.42 × 10-4 and PB has MAE 3.72 × 10-4. The maximum percentage difference in PTV dose coverage (D98) between ML and MC methods is no more than 1.2% for patient with head tumor, while the difference is larger than 10% using the PB method. For patient with lung tumor, the maximum percentage difference in PTV dose coverage (D98) between ML and MC methods is no more than 2.1%, while the difference is larger than 16% using the PB method. CONCLUSIONS: In this work, a reliable DDM prediction method is established for plan optimization by applying several voxel features and the ML approach. The results show that the ML method based on voxel features can obtain plans comparable to the MC method and is better than the PB method in achieving accurate dose to the patient, which is helpful for rapid plan optimization and accurate dose calculation. ADVANCES IN KNOWLEDGE: Establishment of a new machine learning method based on the relationship between the voxel and beamlet features for dose deposition matrix prediction in radiation therapy.

Diagnostic Significance of 18F-FDG PET/CT Imaging Coupled with Magnetic Resonance Imaging of the Entire Body for Bone Metastases.

Objective: To see if 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) imaging paired with MR diffusion imaging can help doctors diagnose bone metastases. Methods: From September 2020 to December 2021, a total of 30 individuals with probable bone metastases were recruited for the trial. With an average interval of four days, MAGNETIC resonance whole-body diffusion imaging (MR whole-body diffusion imaging) was performed on each of the 30 patients who had 18F-FDG PET/CT. The SUVmax values of the group with bone metastases were compared to those of the group without bone metastases. In this study, 18F-FDG PET/CT imaging, MR whole-body diffusion imaging, and their combination were examined. The researchers compared the results when 18F-FDG PET/CT imaging, whole-body MRI diffusion scans, and their combination indicated abnormal bone lesions. By comparing the diagnostic efficacy of 18F-FDG PET/CT imaging, MR whole-body diffusion imaging, and their combination, as well as accuracy, sensitivity, and specificity, the three techniques for diagnosing bone metastases will be evaluated for diagnostic usefulness. Results: the SUV max values of patients with bone metastases were significantly different from those of patients without bone metastases, as determined by 18F-FDG PET/CT imaging (P < 0.05). Using 18F-FDG PET/CT imaging, MR whole-body diffusion imaging, and their combined detection of aberrant bone lesions in various areas, we found statistically significant differences. Conclusion: The use of 18F-FDG PET/CT imaging in conjunction with MR whole-body diffusion imaging in the diagnosis of bone metastases can be very helpful.

Specific genomic alterations and prognostic analysis of perihilar cholangiocarcinoma and distal cholangiocarcinoma.

BACKGROUND: Cholangiocarcinoma (CCA), which consists of intrahepatic CCA (iCCA), perihilar CCA (pCCA), and distal CCA (dCCA), is an aggressive malignancy worldwide. PCCA and dCCA are often classified as extrahepatic CCA (exCCA). However, the differences in mutational characteristics between pCCA and dCCA remain unclear. METHODS: Deep sequencing targeting of 450 cancer genes was performed for genomic alteration detection. The tumor mutational burden (TMB) was measured by an algorithm developed in-house. Correlation analysis was conducted using Fisher's exact test. RESULTS: FGFR2 and ERBB2 mutations mainly occurred in iCCA and exCCA, respectively. In exCCA, the frequencies of PIK3CA, FAT4, KDM6A, MDM2, and TCF7L2 mutations were significantly higher in pCCA compared to dCCA, while the frequencies of TP53 and KRAS mutations were markedly lower in pCCA than those in dCCA. The prognosis-related mutations were different among the CCA subtypes. NF1 mutation was associated with short disease-free survival (DFS) and overall survival (OS), and ERBB2 mutation was associated with short DFS in dCCA patients. Meanwhile, MAP2K4 mutation was associated with long DFS and OS, and TERT mutation was associated with short DFS in pCCA. A series of mutations in genes, including ARID1A, ARID2, SMAD4, TERT, TP53, and KRAS, were found to be associated with the TMB. CONCLUSIONS: In this study, we investigated the comprehensive genomic characterizations of CCA patients, identified the significant alterations in each subtype, and identified potential biomarkers for prognosis prediction. These results provide molecular evidence for the heterogeneity of CCA subtypes and evidence for further precision targeted therapy of CCA patients.

Tumor necrosis factor-related apoptosis-inducing ligand additive with Iodine-131 of inhibits non-small cell lung cancer cells through promoting apoptosis.

Non-small-cell lung cancer (NSCLC) accounts for ~80% of human lung cancer cases and is the most common cause of cancer-associated mortality worldwide. Reports have indicated that tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) and Iodine-131 (I-131) can induce tumor cell apoptosis. The purpose of the present study was to investigate the additive efficacy of TRAIL and I-131 on NSCLC cells. The present study demonstrated that additive treatment of TRAIL and I-131 (TRAIL-I-131) significantly inhibited the growth and aggressiveness of NSCLC cells compared with single TRAIL or I-131 treatment. Results demonstrated that TRAIL-I-131 treatment induced apoptosis of NSCLC cells, with western blot analysis confirming that TRAIL-I-131 treatment increased proapoptotic Bad and Bax expression levels, while antiapoptotic Bcl-2 and Bcl-w protein levels were decreased in NSCLC cells. The present study demonstrated that TRAIL-I-131 treatment inhibited vascular endothelial growth factor (VEGF) and activator protein-1 (AP-1) in NSCLC cells. Potential mechanism analyses identified that TRAIL-I-131 treatment induced apoptosis of NSCLC cells through caspase-9 activation. In vivo assays revealed that TRAIL-I-131 treatment significantly inhibited NSCLC tumor growth and increased apoptotic bodies in tumor tissues. Immunohistology demonstrated that caspase-9 was upregulated and VEGF was downregulated in tumor tissues in TRAIL-I-131-treated tumors. In conclusion, these results indicate that TRAIL combined with I-131 promoted apoptosis of NSCLC through caspase-9 activation, which may be a promising anticancer therapeutic schedule for the treatment of NSCLC.

A rare case of urachal mucinous adenocarcinoma detected by 18F-FDG PET/CT.

A 48-year-old woman presented with abdominal pain, and her initial investigations, which included an abdominal ultrasound and a pelvic MRI examination, revealed a dominant soft tissue mass abutting her left iliac bone, extensive pelvic adenopathy, and multiple osseous metastases. The findings were concerning for chondrosarcoma; however, biopsy results were consistent with mucinous carcinoma of unknown origin. A staging 18F-FDG PET/CT revealed a mildly FDG-avid soft tissue mass with scattered calcifications extending to the dome of the urinary bladder, highly suggestive of a primary tumor. Cystoscopy with tissue sampling of this mass demonstrated a primary mucinous adenocarcinoma of the urachus.

Combination of FDG PET/CT and contrast-enhanced MSCT in detecting lymph node metastasis of esophageal cancer.

BACKGROUND: Lymph node metastasis is believed to be a dependent negative prognostic factor of esophageal cancer. To explore detection methods with high sensitivity and accuracy for metastases to regional and distant lymph nodes in the clinic is of great significance. This study focused on clinical application of FDG PET/CT and contrast-enhanced multiple-slice helical computed tomography (MSCT) in lymph node staging of esophageal cancer. MATERIALS AND METHODS: One hundred and fifteen cases were examined with enhanced 64-slice-MSCT scan, and FDG PET/CT imaging was conducted for neck, chest and upper abdomen within one week. The primary lesion, location and numbers of metastatic lymph nodes were observed. Surgery was performed within one week after FDG PET/CT detection. All resected lesions were confirmed histopathologically as the gold standard. Comparative analysis of the sensitivity, specificity, and accuracy based on FDG PET/CT and MSCT was conducted. RESULTS: There were 946 lymph node groups resected during surgery from 115 patients, and 221 were confirmed to have metastasis pathologically. The sensitivity, specificity, accuracy of FDG PET/CT in detecting lymph node metastasis were 74.7%, 97.2% and 92.0%, while with MSCT they were 64.7%, 96.4%, and 89.0%, respectively. A significance difference was observed in sensitivity (p=0.030), but not the others (p>0.05). The accuracy of FDG PET/CT in detecting regional lymph node with or without metastasis were 91.9%, as compared to 89.4% for MSCT, while FDG PET/CT and MSCT values for detecting distant lymph node with or without metastasis were 94.4% and 94.7%. No significant difference was observed for either regional or distant lymph node metastasis. Additionally, for detecting para-esophageal lymph nodes metastasis, the sensitivity of FDG PET/CT was 72%, compared with 54.7% for MSCT (p=0.029). CONCLUSIONS: FDG PET/CT is more sensitive than MSCT in detecting lymph node metastasis, especially for para-esophageal lymph nodes in esophageal cancer cases, although no significant difference was observed between FDG PET/CT and MSCT in detecting both regional and distant lymph node metastasis. However, enhanced MSCT was found to be of great value in distinguishing false negative metastatic lymph nodes from FDG PET/CT. The combination of FDG PET/CT with MSCT should improve the accuracy in lymph node metastasis staging of esophageal cancer.

Value of 11C-choline PET/CT for lung cancer diagnosis and the relation between choline metabolism and proliferation of cancer cells.

The aim of this study was to investigate the efficacy of 11C-choline PET/CT imaging for lung cancer and the correlation between choline uptake of lung cancer tissue and the expression of choline kinase (ChoK), phosphorylcholine-cytidyl transferase and Ki-67 index. Between March 2008 and June 2010, 53 patients diagnosed or suspected of having lung cancer underwent integrated 11C-choline PET/CT and contrast-enhanced CT scans before surgery. After surgery, specimens from 42 patients diagnosed with lung cancer were used to detect the expression of ChoK, phosphorylcholine-cytidyl transferase and the Ki-67 index. The PET/CT results were analyzed using visual methods and the standardized uptake value (SUV) of lesions was measured using semi-quantitative methods. Finally, the analyzed results were compared to the histopathological results. The accuracy of the 11C-choline PET/CT for diagnosing lung cancer was 81.13% (43/53), compared with 71.70% (38/53) for CT scanning. The difference was not statistically significant (P=0.61). The accuracy of 11C-choline PET/CT for diagnosing lymph nodes was 83.76% (227/271), compared with 66.79% (181/271) for CT scanning. This difference was statistically significant (P=0.04); the SUVmean value of lesions correlated positively with the Ki-67 index (r=0.51, p=0.002). Of the 35 patients with positive 11C-choline PET results, 29 (82.86%) overexpressed ChoK, 26 (74.29%) overexpressed phosphorylcholine-cytidyl transferase. The seven patients with negative 11C-choline PET results did not exhibit overexpression of ChoK or phosphorylcholine-cytidyl transferase; the SUVmean value correlated positively with the expression of both ChoK and phosphorylcholine-cytidyl transferase (r=0.52, p=0.001; r=0.37, p=0.029). In conclusion, compared with contrast-enhanced CT, 11C-choline PET offers nodal staging with higher accuracy. The SUV value of PET is correlated with the proliferation of tumor cells and the mechanism of PET imaging is associated with the overexpression of ChoK and phosphorylcholine-cytidyl transferase.

Comparison of (11)C-choline PET/CT and enhanced CT in the evaluation of patients with pulmonary abnormalities and locoregional lymph node involvement in lung cancer.

UNLABELLED: Carbon-11 ((11)C)-choline positron emission tomography/computed tomography (PET/CT) is a new diagnostic method for detecting lung cancer. To prove whether it is superior to enhanced CT, we compared the diagnostic efficacy of these 2 approaches in 108 patients with pulmonary lesions and drew the conclusion that (11)C-choline PET/CT is not better at diagnosing primary lesions but is better at lymph node staging. This helps us to fully understand (11)C-choline PET/CT. BACKGROUND: This study compares the diagnostic abilities of integrated (11)C-choline PET/CT imaging and contrast-enhanced helical CT imaging in pulmonary lesions and locoregional lymph node metastases in patients with lung cancer. PATIENTS AND METHODS: One hundred eight patients with proven or suspected lung cancer underwent integrated (11)C-choline PET/CT and contrast-enhanced CT, followed by surgical resection and nodal staging. RESULTS: The (11)C-choline PET/CT and CT diagnoses of pulmonary lesions and locoregional lymph node metastases were compared with pathologic findings, which revealed benign lesions in 26 patients (tuberculoma [8 patients], inflammatory pseudotumor [7 patients], hamartoma [6 patients], sclerosing hemangioma [4 patients], and pulmonary sequestration [1 patient]) and lung cancers in 82 patients (adenocarcinoma [39 patients], squamous cell carcinoma [23 patients], carcinoid [7 patients], small-cell lung cancer [5 patients], adenosquamous carcinoma [5 patients], and large-cell lung cancer [3 patients]). The accuracy, sensitivity, and specificity of (11)C-choline PET/CT for diagnosing lung cancer were 82.4%, 85.4%, and 73.1%, respectively, compared with 73.1%, 76.8%, and 61.5%, respectively, for CT. Differences between (11)C-choline PET/CT and CT in diagnosing lung cancer were not statistically significant (p = .503, .118, and .375, respectively). We used receiver operating characteristic (ROC) curve for analysis, finding the ROC of standard uptake value (SUV(max)) for diagnosing lung cancer. The cutoff value of SUV(max) was 3.54. Preoperative nodal staging was compared with postoperative histopathologic staging. (11)C-choline PET/CT correctly staged 80.5% of patients, 12.2% were overstaged, and 7.3% were understaged; for CT these values were 58.5%, 24.4%, and 17.1%, respectively. The accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of (11)C-choline PET/CT for lymph nodes were 83.8%, 82.4%, 84.1%, 50.3%, and 96.1%, respectively, compared with 69.3%, 63.7%, 71.2%, 30.2%, 91.0%, respectively, for CT. CONCLUSION: Differences in the accuracy, sensitivity, specificity, PPV, and NPV between (11)C-choline PET/CT and CT are thus statistically significant for nodal staging (p = .003, .007, .000, .000, and .004, respectively). Although (11)C-choline PET/CT is not significantly better at diagnosing pulmonary lesions than is enhanced CT, (11)C-choline PET/CT has improved sensitivity, specificity, accuracy, PPV, and NPV relative to enhanced CT in the evaluation of locoregional lymph nodes.

The clinical application value of PET/CT in adenocarcinoma with bronchioloalveolar carcinoma features.

OBJECTIVE: The goal of our study was to demonstrate the clinical usefulness of positron emission tomography/computed tomography (PET/CT) for adenocarcinoma with bronchioloalveolar carcinoma (BAC) features, through evaluating the relationship between the intrathoracic lymph node metastases and maximum standardized uptake value (SUVmax), tumor size of the primary tumor and the ratio of BAC component and analysing the correlation of SUVmax, tumor size and the ratio of BAC component. METHODS: This was a retrospective study. Forty-five patients with focal peripheral lung adenocarcinoma with BAC features were included in this study and underwent the PET/CT scan. Twenty-one patients were women and 24 were men. None of the patients had insulin-dependent diabetes and the serum glucose levels in all patients just before (18)F-FDG was injected were less than 120 mg/dl. The diagnosis of the lesion was made by surgical histopathology. RESULTS: All patients underwent successful surgery, and pathologic examination confirmed that 34 of 118 excised nodal groups in 18 patients were proved to be positive for malignancy. Univariate analysis revealed 3 potential factors related to intrathoracic lymph node metastases: SUVmax (P = 0.002); the ratio of BAC component (P = 0.002); maximum dimension of a tumor on mediastinal window setting images (mDmax, P = 0.025). The maximum dimension of a tumor on pulmonary window setting images (pDmax, P = 0.373) had no significance. A receiver operating characteristic (ROC) curve based on SUVmax, mDmax and the ratio of BAC component was constructed, the area under curve (AUC) was 85.2, 70.3 and 81.5% separately. There was no statistical significance between AUC of SUVmax and AUC of the ratio of BAC component (Z = 0.901, P = 0.368). The AUC of SUVmax and AUC of the ratio of BAC component were significantly higher than AUC of mDmax (Z = 2.112, P = 0.035; Z = 2.016, P = 0.042).The SUVmax and the ratio of BAC component had significant inverse correlation (r = -0.85, P < 0.01). The mDmax and the ratio of BAC component had significant inverse correlation (r = -0.69, P < 0.01). The SUVmax and mDmax had significant correlation (r = 0.60, P < 0.01). CONCLUSIONS: PET/CT would be clinically useful for adenocarcinoma with BAC features, because SUVmax obtained by PET/CT can predict the incidence of intrathoracic lymph node metastases at preoperative stages and even for inoperable patients.

Parkinson's disease: functional changes in frontal and parietal cortex using 18 F- fluoro-deoxy glucose positron emission tomography/computed tomography.

BACKGROUND: In Parkinson's disease (PD) there is increasing evidence to suggest motor function changes of the cerebral cortex occur in addition to the pathological changes in the extrapyramidal system. AIMS: To explore the functional changes in the frontal and parietal cortex in PD cat model. SETTINGS AND DESIGN: Twenty-four healthy male cats were divided into four animal model groups with the injection of 1-methyl-4-phenyl-1, 2, 3, 6-tetrahydropyridine (MPTP), and one control group. MATERIALS AND METHODS: Cats in both the animal model and control groups were observed for the behavioral changes. They were also examined by 18 F-fluoro-deoxy glucose positron emission tomography/computed tomography ( 18 F-FDG PET/CT). Region of Interest (ROI) was determined by 18 F-FDG intensity and semi-quantitative analysis was employed after detecting the standard absorption value (SUV). Statistical Analysis : Statistical significance was evaluated by ANOVA. RESULTS: Compared to control group, the 18 F-FDG intensity and SUV were found normal on both the sides of frontal and parietal cortex in the PD models on the second day (P > 0.05), and on the fifth day, they were reduced significantly on both the sides of frontal cortex exclusively (P < 0.05). Moreover on the eighth day, the SUV of both frontal cortexes was reduced, while it was increased in both parietal cortex (P < 0.05). Finally on the eleventh day, the SUV remained stable in both the frontal cortex, and was back to normal level in both the parietal cortex. CONCLUSIONS: Functional disorders exist in the frontal cortex of PD animals and aggravate with time. Transient functional enhancement in the parietal cortex of PD cats might be a compensation for the dysfunction of frontal cortex.

C-11 choline PET/CT imaging for differentiating malignant from benign prostate lesions.

PURPOSE: To investigate the potential of C-11 choline PET/CT imaging for differentiating prostate cancer (PCa) from benign prostate hyperplasia (BPH). MATERIALS AND METHODS: Forty-nine patients with prostate lesions underwent C-11 choline PET/CT imaging that was performed 5 minutes after injection of 7.4 MBq/kg (0.2 mCi/kg C-11 choline in the supine position over 2 bed positions (3 minutes per position), covering the pelvis, and the whole body (6 bed) when necessary. After attenuation correction, PET data were analyzed visually and semiquantitatively by measuring maximum standardized uptake value (SUVmax) of the prostate lesions (target) and the muscles (nontarget) and calculating their ratios (P/M). RESULTS: Twenty-one PCa and 28 BPH lesions were proven histologically. The mean values of the SUVmax of PCa and BPH were 7.87 +/- 5.74 and 4.95 +/- 5.14, respectively without a significant difference between these 2 groups (t = 2.02; P > 0.05). The mean P/M of PCa and BPH were 4.21 +/- 1.61 and 1.87 +/- 0.98. The statistical difference of P/M between them was significant (t = 2.04; P < 0.01). Using 2.3 (P/M) as the criterion, C-11 choline PET/CT imaging showed a sensitivity of 90.48%, a specificity of 85.71%, and a negative predictive value of 92.31%. PET/CT precise localization of the hot spot in different parts of the prostate could contribute to the diagnosis. CONCLUSIONS: C-11 choline PET/CT is a valuable noninvasive imaging technology in the diagnosis of PCa. The parameter P/M could differentiate PCa from benign lesions better than SUV.

[The value of dual time point 11C-choline PET-CT in differentiating malignant from benign lesion of mediastinum].

OBJECTIVE: To evaluate the value of dual time point 11C-choline PET-CT in differentiating malignant from benign lesions of mediastinum. METHODS: Thirty-five patients with mediastinal diseases, including 8 non-small cell lung cancer or highly suspected lung cancer patients with mediastinal lymphadenectasis, were subject to CT, dual time point PET-CT and videomediastinoscopy within four weeks. 11C-choline was used as PET tracers to visualize various masses. The imaging protocol included the first PET scanning 5-10 min after the-injection of 370 MBq 11C-choline and then a second PET scanning 25-30 min later. The PET data were evaluated using the standardized uptake value (SUV) and the difference between the two point (DeltaSUV). Then the results were analyzed in accordance with the pathologic data. RESULTS: Eleven of the 35 patients with mediastinal diseases were diagnosed as with sarcoidosis, 6 with tuberculosis, 5 with lymphoma, 11 with nodal metastasis (8 had their modes from the lung and the primary lesions of the other 3 failed to be identified), and 2 with lung cancer with reactive hyperplasia lymph node. The SUV of the delayed images of the 16 malignant lesions was 6.48 (3.0-11.2), higher than that of the early images [6.17 (3.2-9.8)] with a DeltaSUV of 0.31 (-0.4-1.4). The value of SUV of delayed images of the 19 benign lesions was 4.99 (2.2-9.3), lower than that of early images [5.11 (2.9-8.3)] with a DeltaSUV of -0.12 (-0.9-1.0). The DeltaSUV of the benign lesions was significantly lower than that of the malignant lesions (F = 1.939, P = 0.04). The accuracy rates of diagnosis of mediastinal masses of CT, first-time PET-CT, dual time point PET-CT, and videomediastinoscopy were 54.3% (19/35), 74.3% (26/35), 82.9% (29/35), and 100% (35/35) respectively. Conclusion With a high diagnostic yield, videomediastinoscopy remains the gold standard in differentiation of malignant and benign lesions located in the middle mediastinum. Dual time point PET-CT may improve the accuracy.

The role of 11C-choline positron emission tomography-computed tomography and videomediastinoscopy in the evaluation of diseases of middle mediastinum.

BACKGROUND: Middle mediastinal masses comprise a wide variety of tumors but may also reflect lymphadenopathy, and thus remain an interesting diagnostic challenge. We performed positron emission tomography (PET) of mediastinal masses in order to evaluate the ability of PET to predict the malignancy of these tumors. We compared histologic findings, videomediastinoscopy, computed tomography (CT), and PET-CT in patients with mediastinal disease. METHODS: Thirty-two patients were evaluated with CT, PET-CT and videomediastionoscopy, and all studies were performed within four weeks in each patient. (11)C-choline as a PET tracer was used to visualize masses. PET data were evaluated using the standardized uptake value (SUV) and were compared with pathologic data. RESULTS: There were 13 men and 19 women aged from 21 to 74 (mean 45.2) years. Among the patients with mediastinal diseases, sarcoidosis was diagnosed in 12 patients, tuberculosis in 5 patients, lymphoma in 5 patients, and noncaseating granulomata without classical "sarcoid" finding in 3 patients. N2 or N3 nodal metastasis was revealed in 6 of 7 patients who had non-small cell lung cancer or suspected lung cancer, and one was negative (the pathological diagnosis was reactive hyperplasia). The accuracies for correctly diagnosing mediastinal masses for CT, PET-CT and videomediastinoscopy were 38% (12/32), 63% (20/32), and 91% (29/32) respectively. The diagnostic accuracy of videomediastinoscopy was superior to that of PET-CT (chi(2) = 11.130, P < 0.001). The SUVs were similar among these diseases. On the other hand, if the diagnostic classification was benign vs malignancy, the accuracies for CT, PET-CT and videomediastinoscopy were 53% (17/32), 75% (24/32), 100% (32/32) respectively. The diagnostic accuracy of videomediastinoscopy was superior to that of PET-CT (chi(2) = 22.042, P < 0.001). The SUV of malignant lesions (6.9, 3.2 - 9.8; n = 11) appeared to be higher than that of benign lesions (4.9, 2.9 - 8.3; n = 21), however, this difference was not statistically significant (P = 0.054). CONCLUSIONS: To diagnose lesions located in the middle mediastinum, videomediastinoscopy possesses the highest diagnostic accuracy, and therefore remains the gold standard. PET-CT is valuable for differential diagnosis of benign vs malignant lesions, CT alone or PET alone (SUV) may provide misdiagnosis in a substantial proportion of patients with mediastinal masses.

[Clinical evaluation of 18F-FDE PET-CT in detecting malignant liver tumors].

OBJECTIVE: To evaluate the clinical value of (18)F-fluorodeoxyglucose ((18)F-FDG) positron-emission tomography (PET) combined with computer tomography (PET-CT) in the diagnosis and clinical staging of liver cancer. METHODS: (18)F-FDG PET-CT was performed preoperatively in 16 cases of primary and 8 metastatic liver cancers. The imaging features of the primary foci were analyzed, followed by measurement of standardized (18)F-FDG uptake. For the metastatic foci, the abnormal metabolism of (18)F-FDG was observed and CT, PET and PET-CT fusion images were obtained for accurate localization of these foci. RESULTS: (18)F-FDG uptake occurred in the supraclavicular region in 6 (37.5%) of the 16 patients with primary liver cancer, but was detected in the 8 patients with metastatic liver cancer. Fourteen metastatic nodules were found in 5 of the 16 patients with primary liver cancer, located in the lungs (2 cases) or the abdominal cavity (3 cases). CONCLUSIONS: Negative results of (18)F-FDG PET-CT imaging should be carefully evaluated for diagnosing primary liver cancers, considering the very low sensitivity (37.5%) of this imaging modality in this study. But in the cases of metastatic liver cancers this imaging modality may exhibit high sensitivity, and can also be of great value in clinical staging of the primary liver cancers.