Deep learning-radiomics integrated noninvasive detection of epidermal growth factor receptor mutations in non-small cell lung cancer patients.

This study focused on a novel strategy that combines deep learning and radiomics to predict epidermal growth factor receptor (EGFR) mutations in patients with non-small cell lung cancer (NSCLC) using computed tomography (CT). A total of 1280 patients with NSCLC who underwent contrast-enhanced CT scans and EGFR mutation testing before treatment were selected for the final study. Regions of interest were segmented from the CT images to extract radiomics features and obtain tumor images. These tumor images were input into a convolutional neural network model to extract 512 image features, which were combined with radiographic features and clinical data to predict the EGFR mutation. The generalization performance of the model was evaluated using external institutional data. The internal and external datasets contained 324 and 130 EGFR mutants, respectively. Sex, height, weight, smoking history, and clinical stage were significantly different between the EGFR-mutant patient groups. The EGFR mutations were predicted by combining the radiomics and clinical features, and an external validation dataset yielded an area under the curve (AUC) value of 0.7038. The model utilized 1280 tumor images, radiomics features, and clinical characteristics as input data and exhibited an AUC of approximately 0.81 and 0.78 during the primary cohort and external validation, respectively. These results indicate the feasibility of integrating radiomics analysis with deep learning for predicting EGFR mutations. CT-image-based genetic testing is a simple EGFR mutation prediction method, which can improve the prognosis of NSCLC patients and help establish personalized treatment strategies.

Effect of papillary muscle and trabeculae on left ventricular function analysis via computed tomography: A cross-sectional study.

Deciding whether to include or exclude the papillary muscles and trabeculae to blood pool is essential, because quantifications of left ventricular (LV) functional parameters and myocardial mass are significantly affected. As a result, such inclusion or exclusion might produce different indices for diagnosis and therapy. Using cardiac computed tomography (CT), we obtained standard values of the portion of papillary muscle and trabeculae in normal adults, and to find out how the inclusion or exclusion of papillary muscle and trabeculae affect LV functional parameters depending on the patient group. Excluding the papillary muscles from the LV mass results in easier automated contour detection using CT. The percentage portions of papillary muscle and trabeculae to LV end-diastolic volume (EDV) and LV mass (LVM) were 11.9 ±â€…5.6% and 20.2 ±â€…4.3%, respectively, significantly affecting disease diagnosis. Imaging should be consistent at follow-up and include or exclude the papillary muscles and trabeculae to avoid introducing significant differences between measurements.

The diagnostic performance and clinical value of deep learning-based nodule detection system concerning influence of location of pulmonary nodule.

BACKGROUND: The deep learning-based nodule detection (DLD) system improves nodule detection performance of observers on chest radiographs (CXRs). However, its performance in different pulmonary nodule (PN) locations remains unknown. METHODS: We divided the CXR intrathoracic region into non-danger zone (NDZ) and danger zone (DZ). The DZ included the lung apices, paramediastinal areas, and retrodiaphragmatic areas, where nodules could be missed. We used a dataset of 300 CXRs (100 normal and 200 abnormal images with 216 PNs [107 NDZ and 109 DZ nodules]). Eight observers (two thoracic radiologists [TRs], two non-thoracic radiologists [NTRs], and four radiology residents [RRs]) interpreted each radiograph with and without the DLD system. The metric of lesion localization fraction (LLF; the number of correctly localized lesions divided by the total number of true lesions) was used to evaluate the diagnostic performance according to the nodule location. RESULTS: The DLD system demonstrated a lower LLF for the detection of DZ nodules (64.2) than that of NDZ nodules (83.2, p = 0.008). For DZ nodule detection, the LLF of the DLD system (64.2) was lower than that of TRs (81.7, p < 0.001), which was comparable to that of NTRs (56.4, p = 0.531) and RRs (56.7, p = 0.459). Nonetheless, the LLF of RRs significantly improved from 56.7 to 65.6 using the DLD system (p = 0.021) for DZ nodule detection. CONCLUSION: The performance of the DLD system was lower in the detection of DZ nodules compared to that of NDZ nodules. Nonetheless, RR performance in detecting DZ nodules improved upon using the DLD system. CRITICAL RELEVANCE STATEMENT: Despite the deep learning-based nodule detection system's limitations in detecting danger zone nodules, it proves beneficial for less-experienced observers by providing valuable assistance in identifying these nodules, thereby advancing nodule detection in clinical practice. KEY POINTS: • The deep learning-based nodule detection (DLD) system can improve the diagnostic performance of observers in nodule detection. • The DLD system shows poor diagnostic performance in detecting danger zone nodules. • For less-experienced observers, the DLD system is helpful in detecting danger zone nodules.

Generative Adversarial Network-Based Image Conversion Among Different Computed Tomography Protocols and Vendors: Effects on Accuracy and Variability in Quantifying Regional Disease Patterns of Interstitial Lung Disease.

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.

Enhancing the interferon-γ release assay through omission of nil and mitogen values.

PURPOSE: To address the limited utility of the interferon (IFN)-γ release assay (IGRA) caused by its variability and inconsistency. METHODS: This retrospective cohort study was based on data obtained between 2011 and 2019. QuantiFERON-TB Gold-In-Tube was used to measure IFN-γ levels in nil, tuberculosis (TB) antigen, and mitogen tubes. RESULTS: Of 9,378 cases, 431 had active TB. The non-TB group comprised 1,513 IGRA-positive, 7,202 IGRA-negative, and 232 IGRA-indeterminate cases. Nil-tube IFN-γ levels were significantly higher in the active TB group (median = 0.18 IU/mL; interquartile range: 0.09-0.45 IU/mL) than in the IGRA-positive non-TB (0.11 IU/mL; 0.06-0.23 IU/mL) and IGRA-negative non-TB (0.09 IU/mL; 0.05-0.15 IU/mL) groups (P < 0.0001). From receiver operating characteristic analysis, TB antigen tube IFN-γ levels had higher diagnostic utility for active TB than TB antigen minus nil values. In a logistic regression analysis, active TB was the main driver of higher nil values. In the active TB group, after reclassifying the results based on a TB antigen tube IFN-γ level of 0.48 IU/mL, 14/36 cases with negative results and 15/19 cases with indeterminate results became positive, while 1/376 cases with positive results became negative. Overall, the sensitivity for detecting active TB improved from 87.2 to 93.7%. CONCLUSION: The results of our comprehensive assessment can aid in IGRA interpretation. Since nil values are governed by TB infection rather than reflecting background noise, TB antigen tube IFN-γ levels should be used without subtracting nil values. Despite indeterminate results, TB antigen tube IFN-γ levels can be informative.

Diagnostic accuracy and safety of CT-guided percutaneous lung biopsy with a coaxial cutting needle for the diagnosis of lung cancer in patients with UIP pattern.

This study aimed to assess the diagnostic accuracy and safety of CT-guided percutaneous core needle biopsy (PCNB) with a coaxial needle for the diagnosis of lung cancer in patients with an usual interstitial pneumonia (UIP) pattern of interstitial lung disease. This study included 70 patients with UIP and suspected to have lung cancer. CT-guided PCNB was performed using a 20-gauge coaxial cutting needle. The diagnostic accuracy, sensitivity, specificity, and percentage of nondiagnostic results for PCNB were determined in comparison with the final diagnosis. PCNB-related complications were evaluated. Additionally, the risk factors for nondiagnostic results and pneumothorax were analyzed. The overall diagnostic accuracy, sensitivity, and specificity were 85.7%, 85.5%, and 87.5%, respectively. The percentage of nondiagnostic results was 18.6% (13/70). Two or less biopsy sampling was a risk factor for nondiagnostic results (p = 0.003). The overall complication rate was 35.7% (25/70), and pneumothorax developed in 22 patients (31.4%). A long transpulmonary needle path was a risk factor for the development of pneumothorax (p = 0.007). CT-guided PCNB using a coaxial needle is an effective method with reasonable accuracy and an acceptable complication rate for the diagnosis of lung cancer, even in patients with UIP.

Changes in Thoracic Cavity Volume After Bilateral Lung Transplantation.

Purpose: End-stage lung diseases result in anatomical changes of the thoracic cavity. However, very few studies have assessed changes in the thoracic cavity after lung transplantation (LTx). This study aimed to evaluate the relationships between thoracic cavity volume (TCV) changes after LTx and underlying lung disease. Methods: We reviewed 89 patients who underwent a pre-LTx pulmonary function test (PFT), chest computed tomography (CT) scan, and 1-year follow-up CT after LTx. These patients were classified into two groups according to pre-LTx PFT as follows: obstructive group [forced expiratory volume in 1 s (FEV1)/forced vital capacity (FVC) ratio < 70%] and restrictive group (FEV1/FVC ratio > 70%). We measured TCV using CT scan before and at 1 year after LTx and compared the TCV change in the two groups. Results: In the restrictive group, TCV increased after LTx (preop: 2,347.8 ± 709.5 mL, 1-year postop: 3,224.4 ± 919.0 mL, p < 0.001). In contrast, in the obstructive group, it decreased after LTx (preop: 4,662.9 ± 1,296.3 mL, 1-year postop: 3,711.1 ± 891.7 mL, p < 0.001). We observed that restrictive lung disease, taller stature, lower body mass index, and larger donor lung were independently associated with increased TCV after LTx. Conclusion: The disease-specific chest remodeling caused by restriction and hyperinflation is at least, in part, reversible. After LTx, the chest remodeling appears to occur in the opposite direction to the disease-specific remodeling caused by the underlying lung disease in recipients.

Ciliated muconodular papillary tumor of the lung with cavitary change: A case report with 11-year preoperative follow-up.

Ciliated muconodular papillary tumor (CMPT) is a rare benign lung tumor characterized by ciliated columnar cells, mucous cells, and basal cells. Herein, we report a case of CMPT with 11 years of preoperative follow-up, depicting the natural course of the tumor and changes in computed tomography (CT) findings. A 39-year-old man had a 5-mm solid pulmonary nodule in the right lower lobe that had slowly grown and transformed into a thin-walled cavitary lesion. Right lower lobe lobectomy was performed and the tumor was confirmed to be a CMPT. Although it is difficult to diagnose CMPT with CT findings alone, CMPT should be considered as a possible diagnosis when a slowly growing nodule undergoes cavitary changes.

Minimally invasive adenomyomectomy via a laparoscopic-assisted approach compared to a laparoscopic or laparotomic approach.

OBJECTIVE: The present study assessed the safety and benefits of laparoscopic-assisted adenomyomectomy compared to laparoscopic or laparotomic adenomyomectomy. MATERIALS AND METHODS: This study was a retrospective comparative study. A total of 277 patients underwent adenomyomectomy between January 2016 and January 2019 at the Department of Obstetrics and Gynaecology, Ulsan University Hospital, including 25 with laparoscopic-assisted adenomyomectomy, 82 with laparoscopic adenomyomectomy, and 170 with laparotomic adenomyomectomy. Laparoscopic-assisted adenomyomectomy consisted of a laparoscopic uterine artery procedure to reduce blood loss and a minimal incisional for laparotomic adenomyomectomy. An additional laparoscopic surgery was performed for possible pelvic pathology. RESULTS: Data on patient demographics, surgical indications, operative times, estimated blood loss (EBL), short-term complications, and postoperative hospital stays were compared. The laparoscopic-assisted surgery (LAS) and laparotomic groups were comparable in average EBL (208.0 ± 128.8 vs. 193.6 ± 193.0 ml, p = 0.11), weight of removed mass (85.5 ± 71.7 vs. 108.2 ± 91.9 g, p = 0.39), and postoperative hospital days (HDs) (4.5 ± 1.0 vs. 4.7 ± 0.8 days, p = 0.27). These values were lower in the laparoscopic group (EBL 119.5 ± 79.6 ml, mass weight 39.3 ± 25.9 g, HD 3.6 ± 0.8 days). Additional procedures, including myomectomy and combined severe endometriosis surgery, were more frequently performed in the LAS group than the laparotomic group. The mean operating time was longer in the LAS group (179.8 ± 36.6 min) than the other groups (laparoscopy 99.9 ± 40.6 min, p < 0.00; laparotomy 133.0 ± 41.1 min, p < 0.00). The three groups did not differ significantly in transfusion rates, hemoglobin changes, or perioperative complications. However, febrile morbidity was lower in the laparoscopic group than the LAS and laparotomic groups. CONCLUSION: LAS adenomyomectomy allows for maximal debulking of adenomyosis via extracorporeal and intracorporeal procedures while retaining the advantages of the laparoscopic approach. Additional pelvic surgery for benign uterine and adnexal pathology may easily be performed with this approach.

Usefulness of Staging Chest CT in Breast Cancer: Evaluating Diagnostic Yield of Chest CT According to the Molecular Subtype and Clinical Stage.

The aim of this study is to investigate the clinical utility of staging chest CT in breast cancer by evaluating diagnostic yield (DY) of chest CT in detection of metastasis, according to the molecular subtype and clinical stage. This retrospective study included 840 patients with 855 breast cancers from January 2017 to December 2018. The number of patients in clinical stage 0/I, II, III and IV were 457 (53.5%), 298 (34.9%), 92 (10.8%) and 8 (0.9%), respectively. Molecular subtype was identified in 841 cancers and there were 709 (84.3%) luminal type, 55 (6.5%) human epidermal growth factor receptor 2 (HER2)-enriched type and 77 (9.2%) triple-negative (TN) type. The DYs in clinical stage 0/I, cII, cIII and cIV were 0.2% (1/457), 1.7% (5/298), 4.3% (4/92) and 100.0% (8/8), respectively. The DYs in luminal type, HER2-enriched type and TN type were 1.7% (12/709), 3.6% (2/55) and 2.6% (2/77), respectively. Clinical stage was associated with the DY (p = 0.000). However, molecular subtype was not related to the DY (p = 0.343). Molecular subtype could not provide useful information to determine whether staging chest CT should be performed in early-stage breast cancer. However, chest CT should be considered in advanced breast cancer.

Allergic disorders and susceptibility to and severity of COVID-19: A nationwide cohort study.

BACKGROUND: There is inconclusive and controversial evidence of the association between allergic diseases and the risk of adverse clinical outcomes of coronavirus disease 2019 (COVID-19). OBJECTIVE: We sought to determine the association of allergic disorders with the likelihood of a positive severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) test result and with clinical outcomes of COVID-19 (admission to intensive care unit, administration of invasive ventilation, and death). METHODS: A propensity-score-matched nationwide cohort study was performed in South Korea. Data obtained from the Health Insurance Review & Assessment Service of Korea from all adult patients (age, >20 years) who were tested for SARS-CoV-2 in South Korea between January 1, 2020, and May 15, 2020, were analyzed. The association of SARS-CoV-2 test positivity and allergic diseases in the entire cohort (n = 219,959) and the difference in clinical outcomes of COVID-19 were evaluated in patients with allergic diseases and SARS-CoV-2 positivity (n = 7,340). RESULTS: In the entire cohort, patients who underwent SARS-CoV-2 testing were evaluated to ascertain whether asthma and allergic rhinitis were associated with an increased likelihood of SARS-CoV-2 test positivity. After propensity score matching, we found that asthma and allergic rhinitis were associated with worse clinical outcomes of COVID-19 in patients with SARS-CoV-2 test positivity. Patients with nonallergic asthma had a greater risk of SARS-CoV-2 test positivity and worse clinical outcomes of COVID-19 than patients with allergic asthma. CONCLUSIONS: In a Korean nationwide cohort, allergic rhinitis and asthma, especially nonallergic asthma, confers a greater risk of susceptibility to SARS-CoV-2 infection and severe clinical outcomes of COVID-19.

Amyloid deposition in thymic extranodal marginal zone lymphoma of mucosa-associated lymphoid tissue in a patient with myasthenia gravis: A case report.

Extranodal marginal zone lymphoma of mucosa-associated lymphoid tissue (MALT) in the thymus is a rare disease. Furthermore, amyloid deposition in thymic MALT lymphoma has not been previously described. Here, we report a case of a 35-year-old man with thymic MALT lymphoma with amyloid deposition and myasthenia gravis. Chest computed tomography revealed an anterior mediastinal mass with internal cystic component and extensive calcification. Total thymectomy was performed and histopathologic findings were compatible with a diffuse amyloid deposition in extranodal marginal zone MALT lymphoma. The results indicate that thymic MALT lymphoma should be considered as a possible diagnosis in patients with a solid and cystic thymic mass and autoimmune disease, including myasthenia gravis.

Visual assessment of calcification in solitary pulmonary nodules on chest radiography: correlation with volumetric quantification of calcification.

PURPOSE: To assess the ability of digital chest radiography (CXR) to reveal calcification in solitary pulmonary nodules (SPNs), and to examine the correlation between a visual assessment and volumetric quantification of the calcification. MATERIALS AND METHODS: This study was a retrospective review of 220 SPNs identified by both CXR and chest CT. Eleven observers did blind review of the CXR images and scored nodule calcification on a confidence scale of 1 to 5. The area under the receiver operating characteristics (ROC) curve (AUC) was obtained to analyze the diagnostic performance. The intraclass correlation coefficient (ICC) for interrater reliability was calculated. The AUC and ICC were calculated according to the following nodule diameter groups: group 1 (< 10 mm), group 2 (≥ 10 mm and < 20 mm), and group 3 (≥ 20 mm). RESULTS: Of the 220 SPNs, 145 SPNs (65.6%) were identified as non-calcified and 75 (34.4%) as calcified. The average percentage of calcification volume in SPN > 160 HU (Vol160HU) among the 75 calcified nodules was 47.5%. The mean Vol160HU of the 68 SPNs classified as having definite calcification was 51.1%. The overall AUC was 0.71. The AUCs for groups 1, 2, and 3 was 0.835, 0.639, and 0.620, respectively. The ICCs for groups 1, 2, 3 was 0.65, 0.48, and 0.33, respectively. CONCLUSION: The overall diagnostic performance of digital CXR to predict calcification in SPNs was moderately accurate and the diagnostic performance for predicting calcification in SPNs was significantly higher, and interobserver reproducibility was good when SPN < 10 mm compared with ≥ 10 mm in diameter. KEY POINTS: • The misdiagnosis of a non-calcified nodule as a calcified one by CXR could lead to poor management choices for the SPN. • The diagnostic performance of CXR in predicting calcification was best for nodules < 10 mm in diameter. SPNs with calcification of approximately 50% of their volume tend to be considered calcified. • The diagnostic performance of CXR in identifying calcification was low for nodules ≥ 10 mm in diameter; therefore, we should carefully evaluate calcification carefully for nodules ≥ 10 mm.

Transjugular Access for Endovascular Treatment of Immature Autogenous Arteriovenous Fistulae.

PURPOSE: To assess the feasibility and outcome of transjugular access for endovascular treatment of immature arteriovenous fistulae (AVFs). MATERIALS AND METHODS: Between August 2013 and January 2016, 90 patients (mean age, 64.5 y ± 12.8) underwent endovascular treatment of immature AVFs via transjugular access. The mean age of fistulae was 3.3 months ± 1.8. Total procedure time and technical and clinical success rates of endovascular procedures were assessed. Primary and secondary patency rates were calculated according to the Kaplan-Meier method, and complications were assessed. RESULTS: All patients had inflow lesions, among which 19 (21.1%) had occlusions. The juxtaanastomotic segment was the most common site (44.3%). Transjugular access was successful in 83 patients (92.2%), and 7 required additional standard or transarterial access. The mean procedure time was 36.5 minutes. Technical and clinical success rates were 98.9% and 90.5%, respectively. Mean primary and secondary patency durations were 14.3 months ± 1.7 and 31.0 months ± 0.7, respectively. Primary patency rates at 3, 6, and 12 months were 84.4%, 67.3%, and 48.8%, respectively. Secondary patency rates at 6 and 18 months were 98.6% and 95.5%, respectively. Venous rupture occurred as a result of balloon inflation in 9 patients (10%), and was managed by balloon tamponade. There were no complications related to transjugular access during a mean follow-up period of 12.6 months. CONCLUSIONS: Transjugular access for angioplasty of immature AVFs is feasible and safe. Potential problems associated with access in the outflow vein could be avoided by transjugular access.

Relationship between indexed epicardial fat volume and coronary plaque volume assessed by cardiac multidetector CT.

We explored whether baseline indexed epicardial fat volume (EFVi) and serial changes in EFVi were associated with increase in coronary plaque volume as assessed by multidetector computed tomography.We retrospectively reviewed 87 patients with coronary artery plaque, identified during either baseline or follow-up cardiac computed tomography (CT) examinations. Each plaque volume was measured in volumetric units using a semiautomatic software tool. EFVi was quantified by calculating the total volume of epicardial tissue of CT density -190 to -30 HU, indexed to the body surface area. Clinical cardiovascular risk factors were extracted by medical record review at the time of the cardiac CT examinations. The relationship between EFVi and coronary plaque volume was explored by regression analysis.Although the EFVi did not change significantly from baseline to the time of the follow-up CT (65.7 ±â€Š21.8 vs 66.0 ±â€Š21.8 cm/m, P = 0.620), the plaque volumes were increased significantly on the follow-up CT scans. The annual change in EFVi was not accompanied by a parallel change in coronary plaque volume (P = 0.096-0.500). On univariate analysis, smoking, hypercholesterolemia, 10-year coronary heart disease risk, obesity, and baseline EFVi predicted rapid increases in lipid-rich and fibrous plaque volumes. On multivariate analysis, baseline EFVi (odds ratio = 1.029, P = 0.016) was an independent predictor of a rapid increase in lipid-rich plaque volume.EFVi was shown to be an independent predictor of a rapid increase in lipid-rich plaque volume. However, changes in EFVi were not associated with parallel changes in coronary plaque volume.