Chest Radiography of Tuberculosis: Determination of Activity Using Deep Learning Algorithm / 결핵및호흡기질환

Background@#Inactive or old, healed tuberculosis (TB) on chest radiograph (CR) is often found in high TB incidence countries, and to avoid unnecessary evaluation and medication, differentiation from active TB is important. This study develops a deep learning (DL) model to estimate activity in a single chest radiographic analysis. @*Methods@#A total of 3,824 active TB CRs from 511 individuals and 2,277 inactive TB CRs from 558 individuals were retrospectively collected. A pretrained convolutional neural network was fine-tuned to classify active and inactive TB. The model was pretrained with 8,964 pneumonia and 8,525 normal cases from the National Institute of Health (NIH) dataset. During the pretraining phase, the DL model learns the following tasks: pneumonia vs. normal, pneumonia vs. active TB, and active TB vs. normal. The performance of the DL model was validated using three external datasets. Receiver operating characteristic analyses were performed to evaluate the diagnostic performance to determine active TB by DL model and radiologists. Sensitivities and specificities for determining active TB were evaluated for both the DL model and radiologists. @*Results@#The performance of the DL model showed area under the curve (AUC) values of 0.980 in internal validation, and 0.815 and 0.887 in external validation. The AUC values for the DL model, thoracic radiologist, and general radiologist, evaluated using one of the external validation datasets, were 0.815, 0.871, and 0.811, respectively. @*Conclusion@#This DL-based algorithm showed potential as an effective diagnostic tool to identify TB activity, and could be useful for the follow-up of patients with inactive TB in high TB burden countries.

Pulmonary Emphysema: Visual Interpretation and Quantitative Analysis

Pulmonary emphysema is a cause of chronic obstructive pulmonary disease. Emphysema can be accurately diagnosed via CT. The severity of emphysema can be assessed using visual interpretation or quantitative analysis. Various studies on emphysema using deep learning have also been conducted. Although the classification of emphysema has proven clinically useful, there is a need to improve the reliability of the measurement.

Pulmonary Emphysema: Visual Interpretation and Quantitative Analysis

Pulmonary emphysema is a cause of chronic obstructive pulmonary disease. Emphysema can be accurately diagnosed via CT. The severity of emphysema can be assessed using visual interpretation or quantitative analysis. Various studies on emphysema using deep learning have also been conducted. Although the classification of emphysema has proven clinically useful, there is a need to improve the reliability of the measurement.

Usefulness of CT-Guided Percutaneous TransthoracicNeedle Lung Biopsies in Patients with SuspectedPulmonary Infection

Objective@#This study aimed to evaluate the clinical benefits and risks of CT-guided percutaneous transthoracic needle lung biopsies (PTNBs) in patients with a suspected pulmonary infection. @*Materials and Methods@#This study included 351 CT-guided PTNBs performed in 342 patients (mean age, 58.9 years [range,17–91 years]) with suspected pulmonary infection from January 2010 to December 2016. The proportion of biopsies that revealed the causative organism for pulmonary infection and that influenced patient’s treatment were measured. Multivariateanalyses were performed to identify factors associated with PTNB that revealed the causative organism or affected the treatment. Finally, the complication rate was measured. @*Results@#CT-guided PTNB revealed the causative organism in 32.5% of biopsies (114/351). The presence of necrotic components in the lesion (odds ratio [OR], 1.7; 95% confidence interval [CI], 1.1–2.7; p = 0.028), suspected pulmonary tuberculosis (OR, 2.0; 95% CI, 1.2–3.5; p = 0.010), and fine needle aspiration (OR, 2.5; 95% CI, 1.1–5.8; p = 0.037) were factors associated with biopsies that revealed the causative organism. PTNB influenced patient’s treatment in 40.7% (143/351) of biopsies. The absence of leukocytosis (OR, 1.9; 95% CI, 1.0–3.7; p = 0.049), presence of a necrotic component in the lesion (OR, 2.4; 95% CI, 1.5–3.8; p < 0.001), and suspected tuberculosis (OR, 1.7; 95% CI, 1.0–2.8; p = 0.040) were factors associated with biopsies that influenced the treatment. The overall complication rate of PTNB was 19% (65/351). @*Conclusion@#In patients with suspected pulmonary infection, approximately 30–40% of CT-guided PTNBs revealed the causative organism or affected the treatment. The complication rate of PTNB for suspected pulmonary infection was relatively low.

Quantitative Thoracic Magnetic Resonance Criteria for the Differentiation of Cysts from Solid Masses in the Anterior Mediastinum

OBJECTIVE: To evaluate quantitative magnetic resonance imaging (MRI) parameters for differentiation of cysts from and solid masses in the anterior mediastinum. MATERIALS AND METHODS: The development dataset included 18 patients from two institutions with pathologically-proven cysts (n = 6) and solid masses (n = 12) in the anterior mediastinum. We measured the maximum diameter, normalized T1 and T2 signal intensity (nT1 and nT2), normalized apparent diffusion coefficient (nADC), and relative enhancement ratio (RER) of each lesion. RERs were obtained by non-rigid registration and subtraction of precontrast and postcontrast T1-weighted images. Differentiation criteria between cysts and solid masses were identified based on receiver operating characteristics analysis. For validation, two separate datasets were utilized: 15 patients with 8 cysts and 7 solid masses from another institution (validation dataset 1); and 11 patients with clinically diagnosed cysts stable for more than two years (validation dataset 2). Sensitivity and specificity were calculated from the validation datasets. RESULTS: nT2, nADC, and RER significantly differed between cysts and solid masses (p = 0.032, 0.013, and 0.63; nT2 > 0.39. In validation dataset 1, the sensitivity of the RER, nADC, and nT2 criteria was 87.5%, 100%, and 75.0%, and the specificity was 100%, 40.0%, and 57.4%, respectively. In validation dataset 2, the sensitivity of the RER, nADC, and nT2 criteria was 90.9%, 90.9%, and 72.7%, respectively. CONCLUSION: Quantitative MRI criteria using nT2, nADC, and particularly RER can assist differentiation of cysts from solid masses in the anterior mediastinum.

Diameter of the Solid Component in Subsolid Nodules on Low-Dose Unenhanced Chest Computed Tomography: Measurement Accuracy for the Prediction of Invasive Component in Lung Adenocarcinoma

OBJECTIVE: To determine if measurement of the diameter of the solid component in subsolid nodules (SSNs) on low-dose unenhanced chest computed tomography (CT) is as accurate as on standard-dose enhanced CT in prediction of pathological size of invasive component of lung adenocarcinoma. MATERIALS AND METHODS: From February 2012 to October 2015, 114 SSNs were identified in 105 patients that underwent low-dose unenhanced and standard-dose enhanced CT pre-operatively. Three radiologists independently measured the largest diameter of the solid component. Intraclass correlation coefficients (ICCs) were used to assess inter-reader agreement. We estimated measurement differences between the size of solid component and that of invasive component. We measured diagnostic accuracy of the prediction of invasive adenocarcinoma using a size criterion of a solid component ≥ 6 mm, and compared them using a generalized linear mixed model. RESULTS: Inter-reader agreement was excellent (ICC, 0.84.0.89). The mean ± standard deviation of absolute measurement differences between the solid component and invasive component was 4 ± 4 mm in low-dose unenhanced CT and 5 ± 4 mm in standard-dose enhanced CT. Diagnostic accuracy was 81.3% (95% confidence interval, 76.7.85.3%) in low-dose unenhanced CT and 76.6% (71.8.81.0%) in standard-dose enhanced CT, with no statistically significant difference (p = 0.130). CONCLUSION: Measurement of the diameter of the solid component of SSNs on low-dose unenhanced chest CT was as accurate as on standard-dose enhanced CT for predicting the invasive component. Thus, low-dose unenhanced CT may be used safely in the evaluation of patients with SSNs.

Nodule Classification on Low-Dose Unenhanced CT and Standard-Dose Enhanced CT: Inter-Protocol Agreement and Analysis of Interchangeability

OBJECTIVE: To measure inter-protocol agreement and analyze interchangeability on nodule classification between low-dose unenhanced CT and standard-dose enhanced CT. MATERIALS AND METHODS: From nodule libraries containing both low-dose unenhanced and standard-dose enhanced CT, 80 solid and 80 subsolid (40 part-solid, 40 non-solid) nodules of 135 patients were selected. Five thoracic radiologists categorized each nodule into solid, part-solid or non-solid. Inter-protocol agreement between low-dose unenhanced and standard-dose enhanced images was measured by pooling κ values for classification into two (solid, subsolid) and three (solid, part-solid, non-solid) categories. Interchangeability between low-dose unenhanced and standard-dose enhanced CT for the classification into two categories was assessed using a pre-defined equivalence limit of 8 percent. RESULTS: Inter-protocol agreement for the classification into two categories {κ, 0.96 (95% confidence interval [CI], 0.94–0.98)} and that into three categories (κ, 0.88 [95% CI, 0.85–0.92]) was considerably high. The probability of agreement between readers with standard-dose enhanced CT was 95.6% (95% CI, 94.5–96.6%), and that between low-dose unenhanced and standard–dose enhanced CT was 95.4% (95% CI, 94.7–96.0%). The difference between the two proportions was 0.25% (95% CI, −0.85–1.5%), wherein the upper bound CI was markedly below 8 percent. CONCLUSION: Inter-protocol agreement for nodule classification was considerably high. Low-dose unenhanced CT can be used interchangeably with standard-dose enhanced CT for nodule classification.

Diagnostic Performance of CT Angiography in Patients Visiting Emergency Department with Overt Gastrointestinal Bleeding

OBJECTIVE: To investigate the diagnostic performance of computed tomography angiography (CTA) in identifying the cause of bleeding and to determine the clinical features associated with a positive test result of CTA in patients visiting emergency department with overt gastrointestinal (GI) bleeding. MATERIALS AND METHODS: We included 111 consecutive patients (61 men and 50 women; mean age: 63.4 years; range: 28-89 years) who visited emergency department with overt GI bleeding. They underwent CTA as a first-line diagnostic modality from July through December 2010. Two radiologists retrospectively reviewed the CTA images and determined the presence of any definite or potential bleeding focus by consensus. An independent assessor determined the cause of bleeding based on other diagnostic studies and/or clinical follow-up. The diagnostic performance of CTA and clinical characteristics associated with positive CTA results were analyzed. RESULTS: To identify a definite or potential bleeding focus, the diagnostic yield of CTA was 61.3% (68 of 111). The overall sensitivity, specificity, positive predictive value (PPV), and negative predictive value were 84.8% (67 of 79), 96.9% (31 of 32), 98.5% (67 of 68), and 72.1% (31 of 43), respectively. Positive CTA results were associated with the presence of massive bleeding (p = 0.001, odds ratio: 11.506). CONCLUSION: Computed tomography angiography as a first-line diagnostic modality in patients presenting with overt GI bleeding showed a fairly high accuracy. It could identify definite or potential bleeding focus with a moderate diagnostic yield and a high PPV. CTA is particularly useful in patients with massive bleeding.