Objective: The development of computed tomography (CT) and magnetic resonance imaging (MRI) has resulted in the discovery of unsuspected endocrinologically silent pituitary masses (pituitary incidentalomas). The aim of this study was to perform a national survey on pituitary incidentalomas in order to establish an appropriate approach to them. Design and methods: Five hundred and six patients with pituitary incidentalomas were obtained by questionnaire from March 1999 to May 2000 under the auspices of the Ministry of Health, Labor and Welfare in Japan. Two hundred and fifty-eight patients underwent surgery (surgical group), while 248 patients were followed up conservatively for a mean period of 26.9 months (range 6-173 months) (non-surgical group). Clinical and biochemical assessment, CT or MRI of the pituitary, and visual field testing by Goldman perimetry were assessed at baseline and 6 months and yearly thereafter. Results: Thirty-three patients with pituitary incidentalomas (13.3%) developed tumor enlargement during the mean follow-up period of 45.5 months. Of 115 estimated non-functioning adenomas, 23 tumors (20.0%) increased during a mean follow-up period of 50.7 months (range 10 -173 months), while 5 of 94 (5.3%) estimated Rathke's cysts increased in size during follow-up. Pituitary apoplexy occurred in one of 248 patients (0.4%). Conclusions: Pituitary incidentalomas usually follow a benign course. We recommend transsphenoidal adenectomy for a solid mass attached to the optic chiasma estimated to be a pituitary adenoma by MRI. Other patients should be followed up by MRI every 6 months for the first 2 years, and then yearly.

Incidental lesions should be considered a cause of false-positive findings (6.1%) when an imaging diagnosis is made of a functioning pituitary microadenoma.

Matrix metalloproteinases (MMPs) have been implicated to play a critical role in glioma invasiveness. In this study, we aimed to investigate the expression of MMP-2 and MMP-9 in human gliomas of different degrees of malignancy, and evaluated the correlation between MMP-2 and MMP-9 expression in gliomas. The samples from 65 cases of glioma were divided into four groups according to the WHO classification: there were 16 cases of grade I, 17 cases of grade II, 20 cases of grade III, and 12 cases of grade IV. Normal brain samples served as the control group, and biopsy specimens were obtained from 8 glioma patients with a needle placed into the adjacent brain 1 cm from the margin after tumor resection. All the samples were stained with hematoxylin and eosin and immunohistochemistry. A computer-aided image-analysis system was employed to measure the integral optical density (IOD) of positive slides. No positive staining was found in the control group. The positive staining was localized in the cytoplasm of glioma cells, the extracellular matrix (ECM), the basement membrane (BM), and the endothelial cells of blood vessels. Positive staining rates increased significantly when the degree of malignancy of gliomas was elevated. The IOD value of MMP-2 and MMP-9 also indicated that the intensity of MMP-2 and MMP-9 expression was elevated significantly with the degree of malignancy of the gliomas. There was a positive correlation between MMP-2 and MMP-9 expression in gliomas. Glioma invasion and angiogenesis were particularly seen in the biopsied tissues, and MMP-9 immunostaining seemed to be much more intense and extensive than MMP-2 immunostaining in these samples. These results suggest that MMP-2 and MMP-9 staining in gliomas is localized in the cytoplasm of tumor cells, BM, and endothelial cells, and that MMP-2 and MMP-9 together play an important role in the invasiveness of gliomas, mediating the degradation of the ECM and angiogenesis. MMP-2 and MMP-9 could be molecular targets in the treatment of malignant glioma.

An improved FP-reduction scheme using CNN technique has been developed for the detection of pulmonary nodules in PET/CT images. The authors' ensemble FP-reduction method eliminated 93% of the FPs; their proposed method using CNN technique eliminates approximately half the FPs existing in the previous study. These results indicate that their method may be useful in the computer-aided detection of pulmonary nodules using PET/CT images.

Lung cancer is a leading cause of death worldwide. Currently, in differential diagnosis of lung cancer, accurate classification of cancer types (adenocarcinoma, squamous cell carcinoma, and small cell carcinoma) is required. However, improving the accuracy and stability of diagnosis is challenging. In this study, we developed an automated classification scheme for lung cancers presented in microscopic images using a deep convolutional neural network (DCNN), which is a major deep learning technique. The DCNN used for classification consists of three convolutional layers, three pooling layers, and two fully connected layers. In evaluation experiments conducted, the DCNN was trained using our original database with a graphics processing unit. Microscopic images were first cropped and resampled to obtain images with resolution of 256 × 256 pixels and, to prevent overfitting, collected images were augmented via rotation, flipping, and filtering. The probabilities of three types of cancers were estimated using the developed scheme and its classification accuracy was evaluated using threefold cross validation. In the results obtained, approximately 71% of the images were classified correctly, which is on par with the accuracy of cytotechnologists and pathologists. Thus, the developed scheme is useful for classification of lung cancers from microscopic images.