• Proceedings of the Korean Information Science Society Conference
• /
• 한국정보과학회 2002년도 봄 학술발표논문집 Vol.29 No.1 (B)
• /
• pp.637-639
• /
• 2002

일반적으로 방사선 의사들(radialogists)이 폐 노쥴(pulmonary nodule)을 탐지하는 데는 실제적으로 30%의 실패율을 가진다고 알려져 있다. 만약 자동화된 시스템이 체스트 영상에서 의심스런 노쥴들의 위치들을 방사선 의사에게 알려줄 수 있다면 잘못 판단되는 노쥴들의 수를 잠재적으로 줄일 수 있다. 우리는 형태학적 필터들(morphological filters)과 두가지 특징-추출(feature-extraction) 기술들을 포함하는 컴퓨터 자동 처리 시스템을 구현하였다. 본 시스템에서는 첫째로 형태학적 필터 (morphological filtering)처리를 행한다. 이 과정은 원래의 영상에 침식(erosion)과 확장(dilation)을 연이어서 행하는 것으로 처리가 어려운 X 선 영상을 좀 더 다루기 쉬운 상태로 바꿔주는 역할을 하게 된다. 둘째는 일차적으로 노쥴로서 컴퓨터에 선택된 의심 부분에 가해주는 특징-추출 테스트로서 이 작용은 노쥴로 감지되었으나 실제로는 노쥴이 아닌 경우인false-positive 감지들을 줄이기 위해서 사용된다. 그리하여 본 시스템은 노쥴의 정확한 판독이 어려운 폐의 X선 영상에 적용되어 false-positive 들을 효과적으로 줄임으로써 보다 효율적인 폐 노쥴의 탐지를 가능하게 하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 한국정보처리학회 2002년도 춘계학술발표논문집 (상)
• /
• pp.767-770
• /
• 2002

일반적으로 방사선 의사들(radialogists)이 폐 노쥴(pulmonary nodule)을 탐지하는 데는 실제적으로 30%의 실패율을 가진다고 알려져 있다. 만약 자동화된 시스템이 체스트 영상에서 의심스런 노쥴들의 위치들을 방사선 의사에게 알려줄 수 있다면 잘못 판단되는 노쥴들의 수를 잠재적으로 줄일 수 있다. 우리는 형태학적 필터들(morphological filters)과 두가지 특징-추출(feature-extraction) 기술들을 포함하는 컴퓨터 자동 처리 시스템을 구현하였다. 본 시스템에서는 첫째로 형태학적 필터(morphological filtering) 처리를 행한다. 이 과정은 원래의 영상에 침식(erosion)과 확장 (dilation)을 연이어서 행하는 것으로 처리가 어려운 X 선 영상을 좀 더 다루기 쉬운 상태로 바꿔주는 역할을 하게 된다. 둘째는 일차적으로 노쥴로서 컴퓨터에 선택된 의심 부분에 가해주는 특징-추출 테스트로서 이 작용은 노쥴로 감지되었으나 실제로는 노쥴이 아닌 경우인 false-positive 갑지들을 줄이기 위해서 사용된다. 그리하여 본 시스템은 노쥴의 정확한 판독이 어려운 폐의 X 선 영상에 적용되어 false-positive 들을 효과적으로 줄임으로써 보다 효율적인 폐 노쥴의 탐지를 가능하게 하였다.

• Journal of Chest Surgery
• /
• 제49권6호
• /
• pp.451-455
• /
• 2016

Background: This study assessed the efficacy of pulmonary metastasectomy for synovial sarcoma in adult patients. Methods: Fifty patients, diagnosed with pulmonary metastasis from June 1990 to August 2010, were reviewed retrospectively. Twenty-eight patients underwent complete pulmonary metastasectomy, and their survival was evaluated. Age, sex, time to metastatic progression, laterality, number of tumors, size of largest nodule, and number of metastasectomies were analyzed as potential prognostic factors. Results: In all, 29 patients underwent at least one pulmonary metastasectomy, and 51 resections were performed. One intraoperative mortality occurred, and the 5-year survival rate was 58.4%. Bilateral metastases and early metastatic progression were associated with poor survival in multivariate analyses. Conclusion: Surgical resection can be a good option for treating pulmonary metastasis in patients with synovial sarcoma. Repeated resection was feasible with low mortality and morbidity.

• Journal of Yeungnam Medical Science
• /
• 제33권1호
• /
• pp.72-75
• /
• 2016

Pulmonary epithelioid hemangioendothelioma (PEH) is a rare, low-to-intermediate malignant tumor of endothelial origin. Computed tomography (CT) findings of PEH demonstrate multiple small bilateral nodules; however, to the best of our knowledge, there were no reports on PEH coexisting with other malignancies. Here, we reported on a case involving PEH in a patient with colon cancer and breast cancer which was misconceived as pulmonary meta- stasis. A 63-year-old woman who suffered from constipation for 2 weeks visited our hospital. Colonoscopy showed a large mass with obstruction on hepatic flexure. The histological diagnosis was adenocarcinoma of the ascending colon. Multiple nodules in both lungs and breast were observed on a chest CT scan. A core biopsy of a breast nodule was performed and a diagnosis of invasive ductal carcinoma of the left breast was made. Pulmonary nodules observed on the chest CT scan was considered as pulmonary metastasis from colon or breast cancer. Laparoscopic right hemicolectomy was performed. At the same time, wedge resection of the lung was performed and pathological diagnosis was PEH. Radiologic features of PEH were difficult to distinguish from lung metastasis. Therefore the author reported a rare case involving PEH in a patient with primary malignancy of colon and breast.

• Tuberculosis and Respiratory Diseases
• /
• 제75권2호
• /
• pp.67-70
• /
• 2013

A 52-year-old man was referred to our clinic for an 11.3 mm nodule in the left lower lobe that was discovered on a chest computed tomography (CT) scan. Eleven small nodules were subsequently found in both lungs. Initially, we performed a transthoracic needle aspiration using CT scan guidance. The pathologic report showed a few clusters of atypical cells that were suspicious for malignancy. The positron emission tomography images revealed multiple lung nodules scattered throughout both lungs. The largest nodule (11.3 mm) in the left lower lobe did not have any discernible fludeoxyglucose uptake. For pathologic confirmation, we consulted a thoracic surgeon to perform the video-assisted thoracoscopic surgery. The final diagnosis was minute pulmonary meningothelial-like nodules (MPMNs). MPMNs are benign in nature, and only a few cases require treatment. However, when clinicians are suspicious of potential malignancy, a pathological correlation is essential, even if the final diagnosis is MPMNs.

• Journal of rehabilitation welfare engineering & assistive technology
• /
• 제10권1호
• /
• pp.87-92
• /
• 2016

In this paper, we improved classification performance of benign and malignant lung nodules by including the parenchyma features. For small pulmonary nodules (4-10mm) nodules, there are a limited number of CT data voxels within the solid tumor, making them difficult to process through traditional CAD(computer aided diagnosis) tools. Increasing feature extraction to include the surrounding parenchyma will increase the CT voxel set for analysis in these very small pulmonary nodule cases and likely improve diagnostic performance while keeping the CAD tool flexible to scanner model and parameters. In AdaBoost learning using naive Bayes and SVM weak classifier, a number of significant features were selected from 304 features. The results from the COPDGene test yielded an accuracy, sensitivity and specificity of 100%. Therefore proposed method can be used for the computer aided diagnosis effectively.

• Korean Journal of Radiology
• /
• 제24권2호
• /
• pp.155-165
• /
• 2023

Objective: Little is known about the effects of using different expert-determined reference standards when evaluating the performance of deep learning-based automatic detection (DLAD) models and their added value to radiologists. We assessed the concordance of expert-determined standards with a clinical gold standard (herein, pathological confirmation) and the effects of different expert-determined reference standards on the estimates of radiologists' diagnostic performance to detect malignant pulmonary nodules on chest radiographs with and without the assistance of a DLAD model. Materials and Methods: This study included chest radiographs from 50 patients with pathologically proven lung cancer and 50 controls. Five expert-determined standards were constructed using the interpretations of 10 experts: individual judgment by the most experienced expert, majority vote, consensus judgments of two and three experts, and a latent class analysis (LCA) model. In separate reader tests, additional 10 radiologists independently interpreted the radiographs and then assisted with the DLAD model. Their diagnostic performance was estimated using the clinical gold standard and various expert-determined standards as the reference standard, and the results were compared using the t test with Bonferroni correction. Results: The LCA model (sensitivity, 72.6%; specificity, 100%) was most similar to the clinical gold standard. When expert-determined standards were used, the sensitivities of radiologists and DLAD model alone were overestimated, and their specificities were underestimated (all p-values < 0.05). DLAD assistance diminished the overestimation of sensitivity but exaggerated the underestimation of specificity (all p-values < 0.001). The DLAD model improved sensitivity and specificity to a greater extent when using the clinical gold standard than when using the expert-determined standards (all p-values < 0.001), except for sensitivity with the LCA model (p = 0.094). Conclusion: The LCA model was most similar to the clinical gold standard for malignant pulmonary nodule detection on chest radiographs. Expert-determined standards caused bias in measuring the diagnostic performance of the artificial intelligence model.

• Tuberculosis and Respiratory Diseases
• /
• 제58권6호
• /
• pp.576-581
• /
• 2005

Background : Thoracic actinomycosis is a relatively uncommon anaerobic infection caused by Actinomyces israelii. There have been only a few case reports of endobronchial actinomycosis. The aim of this study was to evaluate the clinical manifestation and treatment of endobronchial actinomycosis. Material and Methods : Seven patients with endobronchial actinomycosis, who were diagnosed in the past 10 years, were retrospectively reviewed. Results : Cough and sputum were the most common symptoms. The chest radiograph and computed tomography showed necrotic consolidation (n=3), atelectasis (n=2), mass (n=1) and an endobronchial nodule (n=1). Proximal broncholithiasis was observed in five patients. All cases were initially suspected to have either lung cancer or tuberculosis. In these patients, the median duration of intravenous antibiotics was 3 days (range 0-12 days) and the median duration of oral antibiotics was 147 days (range 20-412 days). Two patients received oral antibiotic therapy only. There was no clinical evidence of a recurrence. Conclusion : Endobronchial actinomycosis frequently manifests as a proximal obstructive calcified endobronchial nodule that is associated with distal post-obstructive pneumonia. The possibility of endobronchial actinomycosis is suggested when findings of broncholithiasis are present at chest CT. The traditional recommendation of 2-6 weeks of intravenous antibiotics and 6-12 months of oral antibiotic therapy are not necessarily essential in all cases of endobronchial actinomycosis.

• Tuberculosis and Respiratory Diseases
• /
• 제53권1호
• /
• pp.56-65
• /
• 2002

A pulmonary epithelioid hemangioendothelioma (PEH) is a rare tumor with a vascular origin. A PEH can arise in many organ systems, such as the lung, liver, bone and soft tissues. It is a borderline malignancy but the clinical course is usually benign. In this report, we describe three cases of PEH. Case 1, a 61-year-old man, had nonspecific chest discomfort and the chest X-ray showed a solitary lung nodule. This nodule was diagnosed by an open lung biopsy. The pathologic findings including abundant necrosis, mitosis and hyperchromatic and pleomorphic nuclei, indicated a malignancy. The electron microscopic study showed Weibel-Palade bodies and the immunohistochemical stain for CD31 showed a positive reaction in the tumor cells, and linear staining along the vascular lumina. Case 2, a 34-year-old man, was admitted for an evaluation of multiple small nodules, incidentally detected a screening chest X-ray. The nodules were diagnosed by a immunohistochemical stain for the factor VIII-related antigen. Case 3, a 34-year-old woman, complained of left pleuritic chest pain. A simple chest film and the chest CT scan revealed multiple bilateral nodules and a left pleural effusion. An immunohistochemical stain for the factor VIII-related antigen was used to diagnose the nodules. Forth-one months after the diagnosis, she died of pulmonary insufficiency.

• Journal of Yeungnam Medical Science
• /
• 제25권2호
• /
• pp.102-107
• /
• 2008

Background : Malignant pulmonary nodules account for 30 to 40 percent of all solitary pulmonary nodules (SPNs). Therefore, characterization of SPNs is very important for treatment. Recently, dynamic CT has been widely used for tissue characterization and formation of differential diagnoses. The purpose of this study was to evaluate the ability of dynamic CT to formulate the differential diagnosis of SPNs. Materials and Methods : Nineteen patients with SPNs underwent dynamic CT (unenhanced scans, followed by a series of images at 20, 40, 60, 80, 100, 120, 140, 160, and 180 sec after intravenous injection of contrast medium). Diagnosis of SPN was performed based on pathologic findings in needle biopsy samples. Peak enhancement, net enhancement, slope of enhancement, and maximum relative enhancement ratio of the SPN were measured on dynamic CT, and Levene's test was performed to assess benignancy and malignancy. Results : Twelve SPNs were confirmed to have malignant pathology. There were no significant differences between benign and malignant nodules with respect to peak enhancement (p=0.787), net enhancement (p=0.135), or slope of enhancement (p=0.698). The maximal enhancement ratio was increased in malignancy compared to benignancy, but the difference was not statistically significant (p=0.094). Conclusion : In our study, the hemodynamic characteristics of dynamic CT were not significantly different between benign and malignant nodules. Therefore, long-term studies of larger patient samples are required to confirm our findings.