이경수 교수

BACKGROUND: Integrated (18)fluorine fluorodeoxyglucose ((18)F-FDG) positron emission tomography/computed tomography (PET/CT) has shown somewhat variable sensitivity and specificity for mediastinal nodal staging in granulomatous disease endemic areas. The purpose of the study was to prospectively evaluate the efficacy of PET/CT for mediastinal nodal staging in nonsmall cell lung cancer (NSCLC) patients in a tuberculosis-endemic country. METHODS: Prospective assessment of the diagnostic efficacy o

Deep learning is a genre of machine learning that allows computational models to learn representations of data with multiple levels of abstraction using numerous processing layers. A distinctive feature of deep learning, compared with conventional machine learning methods, is that it can generate appropriate models for tasks directly from the raw data, removing the need for human-led feature extraction. Medical images are particularly suited for deep learning applications. Deep learning techniqu

Sjogren syndrome accompanying pulmonary amyloidosis and lymphoproliferative disease manifests as multiple, large, thin-walled cysts; multiple nodules; parenchymal opacity; and bronchiectasis. These findings are caused by the interstitial or peribronchial infiltration of mixed amyloid and lymphoproliferative cells.

Quantitative imaging using radiomics can capture distinct phenotypic differences between tumors and may have predictive power for certain phenotypes according to specific genetic mutations. We aimed to identify the clinicoradiologic predictors of tumors with ALK (anaplastic lymphoma kinase), ROS1 (c-ros oncogene 1), or RET (rearranged during transfection) fusions in patients with lung adenocarcinoma.A total of 539 pathologically confirmed lung adenocarcinomas were included in this retrospective

RATIONALE: Existing literature is inconclusive regarding how the nodular bronchiectatic form of Mycobacterium avium complex (MAC) disease will progress without treatment and when treatment initiation should be considered. OBJECTIVES: To assess the natural course of MAC pulmonary disease by serial thin-section computed tomography (CT). METHODS: Of 339 patients with nodular bronchiectatic form of MAC disease, we selected 265 untreated patients who had serial CTs (mean observation period, 32 ± 21 m

The halo sign in a pulmonary nodule refers to the condition in which soft tissue attenuation of a pulmonary nodule is surrounded by peripheral ground glass attenuation on high resolution CT. The halo sign can be caused by several pathologic processes: hemorrhagic pulmonary nodules, tumor cell infiltration, and nonhemorrhagic inflammatory lesions. Hemorrhagic pulmonary nodules may occur in infectious diseases including invasive pulmonary aspergillosis, mucormycosis, and candidiasis and noninfecti

By showing the distribution of micronodules in and around the secondary pulmonary lobule, HRCT enables the narrowing of the differential diagnosis of DMLD. CT findings reflect gross morphologic features of pathologic examination.

Thoracic involvement occurs more frequently in systemic lupus erythematosus than in any other connective tissue diseases, and more than half of patients with the disease suffer from the involvement. Primary intrathoracic manifestations include pleural disease (effusions and/or thickening), acute lupus pneumonitis, subacute interstitial lung disease including bronchiolitis obliterans organizing pneumonia and non-specific interstitial pneumonia with fibrosis, chronic interstitial lung disease of u

Micropapillary subtype has recently been established to be a distinct marker for poor prognosis in lung adenocarcinomas. According to the current classification of lung adenocarcinomas, all subtypes are listed semiquantitatively in 5% increments. In other words, a minimal amount of the micropapillary pattern, precisely <5% of the entire tumor is disregarded. Therefore, we sought to assess the prognostic significance and survival outcomes in patients with a micropapillary pattern proportion of <5