• Korean Journal of Radiology
• /
• v.24 no.8
• /
• pp.807-820
• /
• 2023

Objective: To assess whether computed tomography (CT) conversion across different scan parameters and manufacturers using a routable generative adversarial network (RouteGAN) can improve the accuracy and variability in quantifying interstitial lung disease (ILD) using a deep learning-based automated software. Materials and Methods: This study included patients with ILD who underwent thin-section CT. Unmatched CT images obtained using scanners from four manufacturers (vendors A-D), standard- or low-radiation doses, and sharp or medium kernels were classified into groups 1-7 according to acquisition conditions. CT images in groups 2-7 were converted into the target CT style (Group 1: vendor A, standard dose, and sharp kernel) using a RouteGAN. ILD was quantified on original and converted CT images using a deep learning-based software (Aview, Coreline Soft). The accuracy of quantification was analyzed using the dice similarity coefficient (DSC) and pixel-wise overlap accuracy metrics against manual quantification by a radiologist. Five radiologists evaluated quantification accuracy using a 10-point visual scoring system. Results: Three hundred and fifty CT slices from 150 patients (mean age: 67.6 ± 10.7 years; 56 females) were included. The overlap accuracies for quantifying total abnormalities in groups 2-7 improved after CT conversion (original vs. converted: 0.63 vs. 0.68 for DSC, 0.66 vs. 0.70 for pixel-wise recall, and 0.68 vs. 0.73 for pixel-wise precision; P < 0.002 for all). The DSCs of fibrosis score, honeycombing, and reticulation significantly increased after CT conversion (0.32 vs. 0.64, 0.19 vs. 0.47, and 0.23 vs. 0.54, P < 0.002 for all), whereas those of ground-glass opacity, consolidation, and emphysema did not change significantly or decreased slightly. The radiologists' scores were significantly higher (P < 0.001) and less variable on converted CT. Conclusion: CT conversion using a RouteGAN can improve the accuracy and variability of CT images obtained using different scan parameters and manufacturers in deep learning-based quantification of ILD.

• Tuberculosis and Respiratory Diseases
• /
• v.46 no.5
• /
• pp.685-696
• /
• 1999

Background : Idiopathic pulmonary fibrosis (IPF) is a diffuse inflammatory and fibrosing process that occurs within the interstitium and alveolus of the lung with invariably poor prognosis. The major problem in management of IPF results from the variable rate of disease progression and the difficulties in predicting the response to therapy. The purpose of this retrospective study was to evaluate the short-term efficacy of steroid and immunosuppressive therapy for IPF and to identify the pre-treatment determinants of favorable response. Method : Twenty patients of IPF were included. Diagnosis of IPF was proven by thoracoscopic lung biopsy and they were presumed to have active progressive disease. The baseline evaluation in these patients included clinical history, pulmonary function test, bronchoalveolar lavage (BAL), and chest high resolution computed tomography (HRCT). Fourteen patients received oral prednisolone treatment with initial dose of 1mg/kg/day for 8 to 12 weeks and then tapering to low-dose prednisolone (0.25mg/kg/day). Six patients who previously had experienced significant side effects to steroid received 2mg/kg/day of oral cyclophosphamide with or without low-dose prednisolone. Follow-up evaluation was performed after 6 months of therapy. If patients met more than one of followings, they were considered to be responders : (1) improvement of more than one grade in dyspnea index, (2) improvement in FVC or TLC more than 10% or improvement in DLco more than 20% (3) decreased extent of disease in chest HRCT findings. Result : One patient died of extrapulmonary cause after 3 month of therapy, and another patient gave up any further medical therapy due to side effect of steroid. Eventually medical records of 18 patients were analyzed. Nine of 18 patients were classified into responders and the other nine patients into nonresponders. The histopathologic diagnosis of the responders were all nonspecific interstitial pneumonia (NSIP) and that of nonresponders were all usual interstitial pneumonia (UIP) (p<0.001). The other significant differences between the two groups were female predominance (p<0.01), smoking history (p<0.001), severe grade of dyspnea (p<0.05), lymphocytosis in BAL fluid ($23.8{\pm}16.3%$ vs $7.8{\pm}3.6%$, p<0.05), and less honeycombing in chest HRCT findings (0% vs $9.2{\pm}2.3%$, p<0.001). Conclusion : Our results suggest that patients with histopathologic diagnosis of NSIP or lymphocytosis in BAL fluid are more likely to respond to steroid or immunosuppressive therapy. Clinical results in large numbers of IPF patients will be required to identify the independent variables.

• Tuberculosis and Respiratory Diseases
• /
• v.48 no.4
• /
• pp.530-542
• /
• 2000

Background : Nonspecific interstitial pneumonia (NSIP) has been reported recently to have shown much better response to medical treatment and better prognosis compared with idiopathic UIP. However, clinical characteristics of idiopathic NSIP discriminating it from UIP have not been clearly defined. Method : Among 120 patients with biopsy-proven diffuse interstitial lung diseases admitted to the Samsung Medical Center between July 1996 and March 2000, 18 patients with idiopathic NSIP were included in this study. Retrospective chart review and radiographic analysis were performed. Results : 1) At diagnosis, 17 patients were female and the average age was $55.2{\pm}8.4$ years (44~73 years). The average duration from development of respiratory symptom to surgical lung biopsy was $9.9{\pm}17.1$ months. Increase in bronchoalveolar lavage fluid lymphocytes ($23.0{\pm}13.1%$) was noted. On HRCT, ground glass and irregular linear opacity were observed, but honeycombing was absent in all patients. 2) Corticosteroids were initially given to 13 patients, but the medication was stopped in 3 patients due to severe side effects. Further medical therapy was not possible in 1 patient who experienced streroid-induced psychosis. Herpes zoster (n=3), tuberculosis (n=1), avascular necrosis of the hip (n=1), cataract (n=2) and diabetes mellitus (n=1) developed during prolonged corticosteroid administration. Of the 7 patients receiving oral cyclophosphamide therapy, hemorrhagic cystitis hindered one patient from continuing with the medication. 3) After medical treatment, 14 of 17 patients improved, and 3 patients remained stable (mean follow-up ; $24.1{\pm}11.2$ months). FVC increased by $20.2{\pm}11.2%$ of predicted value and the extent of ground glass opacity on HRCT decreased significantly ($15.7{\pm}14.7%$). 4) Of the 14 patients who had stopped medication, 5 showed recurrence of NSIP and 2 became aggravated during steroid tapering. All patients with recurrence showed deterioration within one year after completion of initial treatment. Conclusion : Since idiopathic NSIP has unique clinical profiles and shows good prognosis, diagnosis different from UIP, and aggressive medical treatment are needed.

• Tuberculosis and Respiratory Diseases
• /
• v.49 no.5
• /
• pp.601-613
• /
• 2000

Background : Idiopathic pulmonary fibrosis (IPF) is a fatal progressive fibrous disease of the lung of unknown etiology. Recently it has been classified into several distinct entities on the basis of pathologic and clinical characteristics, ie : usual interstitial pneumonia (UIP), desquamative interstitial pneumonia (DIP), acute interstitial pneumonia (AIP), bronchiolitis obliterans with organizing pneumonia (BOOP), and nonspecific interstitial pneumonia (NSIP). IPF is now applied only for UIP, which has the worst prognosis. The previous reports of 3-5 year median survival appears to be overoptimistic because other types with better prognosis like NSIP or BOOP might have been included. Therefore, this study was performed to determine the clinical course and the prognostic factors of UIP as diagnosed by surgical lung biopsy. Methods : The subjects were 72 UIP patients (age $58.2{\pm}11.6$ years, M : F=45 : 27, median follow up period : 18.1 months (0.7-103.6) diagnosed by surgical lung biopsy at the Asan Medical Center (68 patients) and the Paik Hospital in Seoul (4 patients). Clinical scores (level of dyspnea : 1-20 points), radiologic score (honeycombing : HC score 0-5 points, ground glass : GG score 0-5 points), and physiologic scores (FVC : 1-12 points, $FEV_1$ : 0-3 points, TLC : 0-10 points, $D_{LCO)$ : 0-5 points, $AaDO_2$ : 0-10 points) were summed into a total CRP score. Results : 1) The one year survival rate was 78.3%, while the rate for three year survival was 58.1%, and the median survival period was 42.5months. 2) Short term (1 year) prognosis : The patients who died within one year of diagnosis (14 patients) had the higher initial total CRP score ($28.6{\pm}8.3$ vs. $16.6{\pm}9.7$) than those who lived longer than one year (46 patients). The difference in the total CRP score was attributed to the symptom score ($8.4{\pm}2.1$ vs. $5.7{\pm}3.9$) and the physiologic score ($15.7{\pm}7.1$ vs. $6.7{\pm}5.7$) including FVC, $D_{LCO)$ and $AaDO_2$. 3) Long-term (3year) prognosis : The total CRP score ($12.2{\pm}6.7$ vs. $28.7{\pm}7.9$ : including symptom score, FVC, $D_{LCO)$ and $AaDO_2$) at the time of diagnosis were also different for the long-term survivors and those who lived less than 3 years. 4) Cox regression analysis showed $D_{LCO)$ (${\geq}$60%) (Hazard ratio : 4.56, 95% CI : 2.30-16.04) was the independent prognostic factors of UIP (P<0.05). Conclusion : These results suggest that $D_{LCO)$ at the time of diagnosis seem to be a prognostic markers of biopsy-proven UIP.