Radiological management and challenges of the twin earthquakes of February 6th.

Two major earthquakes measuring 7.8 and 7.7 on the Richter scale struck Turkey and Northern Syria on February 6, claiming more than 50,000 lives. In such an unprecedented disaster, radiologists were confronted with very critical tasks of stepping out of the routine reporting process, performing radiological triage, managing acute adverse events, and optimizing imaging protocols. In our experience, radiologists can take three different positions in such disasters: (1) in the scene of the disaster, (2) serving in teleradiology, and (3) working in tertiary hospital for transported patients. With this article, we aimed to describe the challenges radiologists face on the three main fronts and how we manage these challenges.

Radiological Findings in SARS-CoV-2 Viral Pneumonia Compared to Other Viral Pneumonias: A Single-Centre Study.

Background: Thorax computed tomography (CT) imaging is widely used as a diagnostic method in the diagnosis of coronavirus disease 2019 (COVID-19)-related pneumonia. Radiological differential diagnosis and isolation of other viral agents causing pneumonia in patients have gained importance, particularly during the pandemic. Aims: We aimed to investigate whether there is a difference between CT images from patients with COVID-19-associated pneumonia compared to CT images of patients with pneumonia due to other viral agents and which finding may be more effective in diagnosis. Study Design. The study included 249 adult patients with pneumonia identified by thorax CT examination and with a positive COVID-19 RT-PCR test compared to 94 patients diagnosed with non-COVID-19 pneumonia (viral PCR positive but no bacterial or fungal agents detected in other cultures) between 2015 and 2019. CT images were retrospectively analyzed using the PACS system. CT findings were evaluated by two radiologists with 5 and 20 years of experience, in a blinded fashion, and the outcome was decided by consensus. Methods: Demographic data (age, gender, and known chronic disease) and CT imaging findings (percentage of involvement, number of lesions, distribution preference, dominant pattern, ground-glass opacity distribution pattern, nodule, tree in bud sign, interstitial changes, crazy paving sign, reversed halo sign, vacuolar sign, halo sign, vascular enlargement, linear opacities, traction bronchiectasis, peribronchial wall thickness, air trapping, pleural retraction, pleural effusion, pericardial effusion, cavitation, mediastinal/hilar lymphadenopathy, dominant lesion size, consolidation, subpleural curvilinear opacities, air bronchogram, and pleural thickening) of the patients were evaluated. CT findings were also evaluated with the RSNA consensus guideline and the CORADS scoring system. Data were divided into two main groups-non-COVID-19 and COVID-19 pneumonia-and compared statistically with chi-squared tests and multiple regression analysis of independent variables. Results: RSNA and CORADS classifications of CT scan images were able to successfully differentiate between positive and negative COVID-19 pneumonia patients. Statistically significant differences were found between the two patient groups in various categories including the percentage of involvement, number of lesions, distribution preference, dominant pattern, nodule, tree in bud, interstitial changes, crazy paving, reverse halo vascular enlargement, peribronchial wall thickness, air trapping, pleural retraction, pleural/pericardial effusion, cavitation, and mediastinal/hilar lymphadenopathy (p < 0.01). Multiple linear regression analysis of independent variables found a significant effect in reverse halo sign (ß = 0.097, p < 0.05) and pleural effusion (ß = 10.631, p < 0.05) on COVID-19 pneumonia patients. Conclusion: The presence of reverse halo and absence of pleural effusion was found to be characteristic of COVID-19 pneumonia and therefore a reliable diagnostic tool to differentiate it from non-COVID-19 pneumonia.

Measurement of iliotibial band diameter in iliotibial band friction syndrome and comparison with an asymptomatic population.

BACKGROUND: Iliotibial band friction syndrome (ITBFS) is an overuse injury of the lateral aspect of the knee. This syndrome classically affects the active young population. PURPOSE: To determine the diameter of the ITB using magnetic resonance imaging (MRI) in patients clinically diagnosed with ITBFS, compare the results with asymptomatic patients, and assess the inter-observer agreement between a senior and a junior radiologist with different levels of experience in musculoskeletal imaging. MATERIAL AND METHODS: From April 2014 to October 2019, 78 knee MRI scans of 78 patients were included in the study group who were referred from the orthopedic clinic with a clinical diagnosis of ITBFS. In the control group, there were 114 knee MRI scans of 114 patients who had knee MRI for various reasons and had no radiological abnormality on the performed knee MRI. The ITB diameters, cut-off values, and interclass correlation coefficient (ICC) were calculated. RESULTS: Mean thickness of the ITB was higher in the study group compared to the control group in measurements done by both the senior and junior radiologists and this was statistically significant (P < 0.001). Cut-off values of the diameters of the ITB were calculated as 2.385 for the senior radiologist and 2.420 for the junior radiologist. ICC of 0.80 was determined, which showed excellent agreement among interpreters. CONCLUSION: ITB thickness in the study group was significantly higher than in the control group. There was also excellent agreement among the two observers. Measurement of ITB thickness on axial plane knee MRI is one of the reliable criteria for ITBFS.

Liver and pancreas: 'Castor and Pollux' regarding the relationship between hepatic steatosis and pancreas exocrine insufficiency.

BACKGROUND: Pancreatic exocrine insufficiency (PEI) is found in 30-50% of diabetes mellitus (DM). Insulin resistance is triggering factor in both DM and nonalcoholic fatty liver disease (NAFLD). Therefore, we aimed to investigate frequency of PEI in NAFLD, and relationship of fecal pancreatic elastase (PE) levels with liver histology and pancreatic fat. METHODS: Ninety-seven biopsy proven NAFLD patients and 50 controls were enrolled. Pancreas exocrine functions were measured by PE. Magnetic resonance imaging-estimated proton density fat fraction (MRI-PDFF) was used to quantify fat. RESULTS: NAFLD patients had significantly lower PE levels than controls (297 [204-517] vs. 500 [298-678] µg/g, p < 0.01). PEI (PE < 200 µg/g) ratio of NAFLD patients (22.7%, n = 22) was higher than PEI ratio of controls (6%, n = 3) (p = 0.011). Among diabetic (n = 35) NAFLD patients, 9 (25.7%) exhibited PEI, compared to 13 (21%) of non-diabetics. There was no significant difference in patients with and without DM in terms of PEI (p = 0.592). Among NASH (n = 68) patients 16 (23.5%) exhibited PEI, compared to (20.7%) of non-NASH (p = 0.76). Multiple analysis revealed NAFLD as a predictor of PEI independent of age, sex and DM (OR = 4.892, p = 0,021). Mean pancreas MRI-PDFF was significantly higher in diabetics (13.7% ± 3.6% vs. 8.7% ± 5.1%, p = 0.001). There was no significant pancreas MRI-PDFF difference between NASH and non-NASH (P = 0.95). Mean pancreas MRI-PDFF was significantly higher in patients with PEI (13.7% ± 3.4% vs. 8.9% ± 5.2%, P < 0.01). CONCLUSION: This is the first study demonstrating the high frequency of PEI in NAFLD independent of DM. Moreover, increasing pancreatic steatosis appears to be associated with higher frequency of PEI in NAFLD.

Covid-19: Correlation of Early Chest Computed Tomography Findings With the Course of Disease.

OBJECTIVE: To investigate the role of chest computed tomography (CT) examinations acquired early after initial onset of symptoms in predicting disease course in coronavirus disease 2019. METHODS: Two hundred sixty-two patients were categorized according to intensive care unit (ICU) admission, survival, length of hospital stay, and reverse transcriptase-polymerase chain reaction positivity. Mean time interval between the onset of symptoms and CT scan was 5.2 ± 2.3 days. Groups were compared using Student t test, Mann-Whitney U, and Fisher exact tests. RESULTS: In the ICU (+) and died groups, crazy paving (64% and 57.1%), bronchus distortion (68% and 66.7%), bronchiectasis-bronchiolectasis (80% and 76.2%), air trapping (52% and 52.4%) and mediastinal-hilar lymph node enlargement (52% and 52.4%) were significantly more encountered (P < 0,05). These findings were correlated with longer hospital stays (P < 0.05). There were no differences between reverse transcriptase-polymerase chain reaction-positive and -negative patients except bronchiectasis-bronchiolectasis. CONCLUSION: Computed tomography examinations performed early after the onset of symptoms may help in predicting disease course and planning of resources, such as ICU beds.

Cardiac Hydatid Cysts: Computed Tomography and Magnetic Resonance Imaging Findings of the 5 Cases.

OBJECTIVE: In this article, we aimed to review the computed tomography (CT) and magnetic resonance imaging (MRI) findings of cardiac hydatid disease. CONCLUSIONS: Regarding the diagnostic work-up of cardiac hydatid disease, CT and MRI are complimentary techniques. Whereas CT effectively depicts wall calcifications, MRI reveals the exact anatomic location and nature of the internal and external structures and is the technique used for the early diagnosis, assessment, and follow-up of patients with cardiac hydatid disease.

Q-elastosonography of solid thyroid nodules: assessment of diagnostic efficacy and interobserver variability in a large patient cohort.

OBJECTIVE: Evaluation of the diagnostic efficacy and interobserver agreement of Q-elastography in the differentiation of benign from malignant thyroid nodules. METHODS: A total of 344 thyroid nodules in 288 patients were examined with grey-scale and colour Doppler ultrasound (CDUS) and Q-elastography by two independent operators. Nodules with hypoechogenicity, poorly defined margins, microcalcifications, and intralesional vascularity were classified as suspicious. Diagnostic performances of CDUS features and Q-elastography for predicting thyroid malignancy were estimated using ROC analysis. Cytology or histopathology was the reference standard. Interobserver agreement in the evaluation of CDUS and Q-elastography was assessed using Cohen's k-statistic. RESULTS: Q-elastography showed excellent diagnostic performance for the prediction of thyroid malignancy, with sensitivity of 93 % and specificity of 92 % for operator 1 (best cutoff at 2.02), and sensitivity of 84 % and specificity of 79 % for operator 2 (best cutoff at 1.86). Performance of Q-elastography was superior to that of CDUS. Reproducibility of the findings was excellent for both Q-elastography and CDUS features as assessed with Cohen's k, which was highest for strain ratio measurements (0.95) and lowest for the echogenicity score (0.83). CONCLUSIONS: Q-elastography showed excellent performance. It is a valid and reproducible diagnostic method as well as a promising tool for identifying suspicious solid thyroid nodules needing cytological assessment and surgery. KEY POINTS: • Elastography is an additional tool for optimal characterisation of malignant thyroid nodules. • The use of semiquantitative elastographic evaluation increases the diagnostic performance, • The interobserver agreement of quantitative elastography can be considered to be good.

Diffusion tensor imaging of cysts, hemangiomas, and metastases of the liver.

BACKGROUND: Although diffusion-weighted magnetic resonance imaging imaging (DW-MRI) is commonly used to characterize hepatic lesions, the literature is sparse about the use of MR diffusion tensor imaging (DTI) in this regard. By using DTI, one is able to obtain not only apparent diffusion coefficients (ADCs) but also fractional anisotropy (Fa) values. PURPOSE: To evaluate DTI using ADC and Fa values in the imaging of hepatic cysts, hemangiomas, and metastases. MATERIAL AND METHODS: Sixty-six patients with 77 lesions were examined with DTI. There were 32 metastases, 13 cysts, and 32 hemangiomas. Two radiologists performed ADC and Fa measurements. Inter-observer agreement was evaluated using Bland-Altman plots. ADCs and Fa values were correlated using Pearson correlation. The differences were compared using ANOVA and Tukey tests. A ROC analysis was applied; sensitivities and specificities were calculated. RESULTS: The inter-observer agreement was very good. The correlation between ADC and Fa was negative, weak, and significant (r = -0.36). The mean ADC value of cysts (3.30 ± 0.8 × 10(-3) mm(2)/s) was significantly higher than that of hemangiomas (2.23 ± 0.5 × 10(-3) mm(2)/s) and metastases (1.62 ± 0.4 × 10(-3) mm(2)/s). The mean Fa value of cysts (0.2 ± 0.05) was significantly lower than hemangiomas (0.37 ± 0.1) and metastases (0.46 ± 0.1). The Az values for discriminating metastases from benign hepatic lesions for ADC and Fa value were 0.885 and 0.731, respectively. The sensitivity and specificity of ADC and Fa were 87.5% and 84.4%, and 78.1% and 57.8%, respectively. The Az value for discriminating cysts from hemangiomas for Fa was 0.96. The sensitivity and specificity were 90.6% and 92.3%, respectively. CONCLUSION: Fa values may play a supportive role in the imaging of liver lesions. Whereas metastases tend to have low ADCs and high Fa values, cysts have high ADCs and low Fa values and hemangiomas have high ADCs and high Fa values.

Immune checkpoint inhibitors: Assessment of the performance and the agreement of iRECIST, irRC, and irRECIST.

INTRODUCTION: Immunotherapy has become more widely accepted and used by medical oncologists. Radiologists face challenges in assessing tumor response and becoming more involved in the management of treatment. We aimed to assess the agreement between immune-related response criteria (irRC), immune-related RECIST (irRECIST), and immune RECIST (iRECIST) to correlate the response measured by them with overall survival (OS), and to determine the confirmation rate of progressive disease (PD). METHODS: A total of 43 patients (28 men, 15 women; average age = 54.6 ± 15.7 years) treated with immunotherapy were included in this study. Pairwise agreements between iRECIST, irRC, and irRECIST were calculated using Cohen's kappa statistics. The correlation of the criteria-based response and OS was evaluated using the Kaplan-Meier method and log-rank test. A confirmation rate with 95% confidence intervals (CI) was calculated in patients with PD. RESULTS: The kappa values between iRECIST and irRC, iRECIST and irRECIST, and irRC and irRECIST were 0.961 (almost perfect; P < 0.001), 0.961 (almost perfect; P < 0.001), and 0.922 (almost perfect; P < 0.001), respectively. The Kaplan-Meier method and log-rank test showed for each criterion a statistically significant correlation with OS (P < 0.05). The confirmation rates of PD for irRC, irRECIST, and iRECIST were 95% (19/20; 95% CI = 76.4-99.1%), 90% (18/20; 95% CI = 69.9-97.2%), and 90.5% (19/21; 95% CI = 71.1-97.4%), respectively. CONCLUSION: There was an almost perfect and statistically significant agreement between iRECIST, irRC, and irRECIST. The measurements performed with them significantly correlated with the OS; their confirmation rates were similar. iRECIST and irRECIST might be favored over irRC because of their relative ease of use.

Thorax computed tomography (CTX) guided ground truth annotation of CHEST radiographs (CXR) for improved classification and detection of COVID-19.

Several data sets have been collected and various artificial intelligence models have been developed for COVID-19 classification and detection from both chest radiography (CXR) and thorax computed tomography (CTX) images. However, the pitfalls and shortcomings of these systems significantly limit their clinical use. In this respect, improving the weaknesses of advanced models can be very effective besides developing new ones. The inability to diagnose ground-glass opacities by conventional CXR has limited the use of this modality in the diagnostic work-up of COVID-19. In our study, we investigated whether we could increase the diagnostic efficiency by collecting a novel CXR data set, which contains pneumonic regions that are not visible to the experts and can only be annotated under CTX guidance. We develop an ensemble methodology of well-established deep CXR models for this new data set and develop a machine learning-based non-maximum suppression strategy to boost the performance for challenging CXR images. CTX and CXR images of 379 patients who applied to our hospital with suspected COVID-19 were evaluated with consensus by seven radiologists. Among these, CXR images of 161 patients who also have had a CTX examination on the same day or until the day before or after and whose CTX findings are compatible with COVID-19 pneumonia, are selected for annotating. CTX images are arranged in the main section passing through the anterior, middle, and posterior according to the sagittal plane with the reformed maximum intensity projection (MIP) method in the coronal plane. Based on the analysis of coronal MIP reconstructed CTX images, the regions corresponding to the pneumonia foci are annotated manually in CXR images. Radiologically classified posterior to anterior (PA) CXR of 218 patients with negative thorax CTX imaging were classified as COVID-19 pneumonia negative group. Accordingly, we have collected a new data set using anonymized CXR (JPEG) and CT (DICOM) images, where the PA CXRs contain pneumonic regions that are hidden or not easily recognized and annotated under CTX guidance. The reference finding was the presence of pneumonic infiltration consistent with COVID-19 on chest CTX examination. COVID-Net, a specially designed convolutional neural network, was used to detect cases of COVID-19 among CXRs. Diagnostic performances were evaluated by ROC analysis by applying six COVID-Net variants (COVIDNet-CXR3-A, -B, -C/COVIDNet-CXR4-A, -B, -C) to the defined data set and combining these models in various ways via ensemble strategies. Finally, a convex optimization strategy is carried out to find the outperforming weighted ensemble of individual models. The mean age of 161 patients with pneumonia was 49.31 ± 15.12, and the median age was 48 years. The mean age of 218 patients without signs of pneumonia in thorax CTX examination was 40.04 ± 14.46, and the median was 38. When working with different combinations of COVID-Net's six variants, the area under the curve (AUC) using the ensemble COVID-Net CXR 4A-4B-3C was .78, sensitivity 67%, specificity 95%; COVID-Net CXR 4a-3b-3c was .79, sensitivity 69% and specificity 94%. When diverse and complementary COVID-Net models are used together through an ensemble, it has been determined that the AUC values are close to other studies, and the specificity is significantly higher than other studies in the literature.

State-of-the-art Prostate Imaging.

Prostate cancer is one of the most common cancers in men. In addition to methods such as prostate-specific antigen test, digital rectal examination, and transrectal ultrasonography, magnetic resonance imaging has an important role for accurate and reproducible diagnosis. However, guidance in targeted biopsies and recent use in determining localization for treatment increase its importance. Due to technical difficulties, patient tolerance, and differences in interpretation, the prostate imaging reporting and data system recommends preparations for the patient and magnetic resonance imaging techniques. However, techniques continue to be developed to improve the diagnosis rate and image quality. In our article, patient preparation before imaging and techniques were tried to be discussed in detail. In addition, current approaches in biparametric magnetic resonance imaging and radiomics and new techniques such as T1 and T2 mapping will be mentioned.

The Significance of Preoperative Computed Tomography Features in the Prediction of Overall Survival in Gastric Cancer: A Retrospective Analysis.

Objectives: Computed tomography (CT) is a frequently used modality for staging in the preoperative evaluation of gastric cancer (GC). Our aim was to interpret the importance of preoperative CT features in predicting overall survival (OS) in patients operated for GC. Methods: One hundred and one patients with GC (33 women, 68 men; range of age: 29-82 years, median age: 61 years) who had abdominal CT prior to surgical resection were included in the study retrospectively. Two radiologists evaluated CT scans to record the longest dimension of the tumor, the localization of the lesion, the attenuation values of the tumor in the arterial and venous phases (Hounsfield units), invasion depth of the lesion (T stage), and the number of pathological lymph nodes (LNs) (N stage). Postoperative pathological results including resection (R0, R1), T stage, N stage, grade, and histopathological subtype were documented. All CT-provided results and clinicopathological features associated with OS were analyzed by univariate, multivariate, and receiver operator characteristic analysis. Results: Multivariate analysis revealed that none of the CT features were associated with the OS. After resection, the survival ratio was poor for the R1 and high-grade groups than for the R0 and low-grade groups (p=0.001 and p=0.005, respectively). N stage and the longest dimension of the tumor on CT imaging truly estimated R1 resection status (AUC, 0.697; sensitivity, 63%; and specificity, 88%, and AUC, 0.734; sensitivity, 18%; and specificity, 76%, respectively). Conclusion: R1 resection status is associated with poor OS in GC. CT features, including the tumor's longest dimension and the number of pathological LNs, can predict R1 resection status.

Prediction of carcinogenic human papillomavirus types in cervical cancer from multiparametric magnetic resonance images with machine learning-based radiomics models.

PURPOSE: This study aimed to evaluate the potential of machine learning-based models for predicting carcinogenic human papillomavirus (HPV) oncogene types using radiomics features from magnetic resonance imaging (MRI). METHODS: Pre-treatment MRI images of patients with cervical cancer were collected retrospectively. An HPV DNA oncogene analysis was performed based on cervical biopsy specimens. Radiomics features were extracted from contrast-enhanced T1-weighted images (CE-T1) and T2-weighted images (T2WI). A third feature subset was created as a combined group by concatenating the CE-T1 and T2WI subsets. Feature selection was performed using Pearson's correlation coefficient and wrapper- based sequential-feature selection. Two models were built with each feature subset, using support vector machine (SVM) and logistic regression (LR) classifiers. The models were validated using a five-fold cross-validation technique and compared using Wilcoxon's signed rank and Friedman's tests. RESULTS: Forty-one patients were enrolled in the study (26 were positive for carcinogenic HPV oncogenes, and 15 were negative). A total of 851 features were extracted from each imaging sequence. After feature selection, 5, 17, and 20 features remained in the CE-T1, T2WI, and combined groups, respectively. The SVM models showed 83%, 95%, and 95% accuracy scores, and the LR models revealed 83%, 81%, and 92.5% accuracy scores in the CE-T1, T2WI, and combined groups, respectively. The SVM algorithm performed better than the LR algorithm in the T2WI feature subset (P = 0.005), and the feature sets in the T2WI and the combined group performed better than CE-T1 in the SVM model (P = 0.033 and 0.006, respectively). The combined group feature subset performed better than T2WI in the LR model (P = 0.023). CONCLUSION: Machine learning-based radiomics models based on pre-treatment MRI can detect carcinogenic HPV status with discriminative accuracy.

Sonographic diagnosis of choledochocele.

We present a case of a 48-year-old female with a history of cholecystectomy and recurrent attacks of pancreatitis whose initial abdominal sonography (US) revealed multiple conglomerated stones in the descending part of the duodenum. Abdominal CT, MRI, and magnetic resonance cholangiopancreatography showed the same findings. The distended sacciform distal intramural segment of the common bile duct was protruding into the duodenum. The imaging findings explained the etiology of the patient's recurrent attacks of pancreatitis and led to surgical excision of the choledococele.

A Rare Presentation of Extramedullary Hematopoiesis: Perirenal Masses.

Extramedullary hematopoiesis (EMH) is a physiological compensatory mechanism that develops in response to ineffective or insufficient hematopoiesis. Although the liver and spleen are the most common sites, EMH may occur anywhere in the body. Peri-renal involvement in EMH is quite rare and mimics tumoral lesions. To our knowledge, 12 peri-renal EMH cases have been reported in the literature so far.  A 57-year woman was admitted to our hospital with non-specific abdominal pain. She had a history of chronic renal failure. On physical examination, diffuse abdominal tenderness and mild rigidity were detected. Blood tests revealed bicytopenia and elevated acute phase reactants. On non-contrast abdominal computed tomography (CT) scan, two peri-renal mass lesions of 4.5 cm and 4 cm were detected incidentally on the right side. Tumoral lesions were considered in the differential diagnosis, and magnetic resonance imaging (MRI) and positron emission tomography-computed tomography (PET-CT) were performed. Lesions showed mild hypermetabolic activity (SUVmax: 4,68) on PET-CT images. For the definitive diagnosis, an ultrasound-guided core needle biopsy was performed. The histopathological examination was consistent with EMH. This case aims to highlight the peri-renal involvement of EMH, which mimics renal and peri-renal malignancies.

Can peritoneal thickening and enhancement be used to determine perforation in patients with acute appendicitis?

BACKGROUND: Acute appendicitis (AA) is a common cause of abdominal pain in developed countries. In patients with suspected AA, computed tomography (CT) is considered as the gold standard with the highest sensitivity and specificity, and it is also an im-portant modality, especially in patients with complicated AA. In this study, we aimed to evaluate age and laboratory findings, as well as specific CT findings in differentiating between perforated and non-perforated appendicitis. METHODS: We retrospectively analyzed 252 patients diagnosed with AA and underwent appendectomy between November 2015 and December 2019 in Somalia Mogadishu Recep Tayyip Erdogan Education and Research Hospital. Patients under 18 years of age and those with no pre-operative CT scans were excluded from the study. The demographic, laboratory, CT findings, and pathological data of all patients were evaluated. RESULTS: This study included 80 patients, 32 (40%) classified as perforated appendicitis (Group-1) and 48 (60%) as non-perforated appendicitis (Group-2). The C-reactive protein value was found to be statistically higher in Group-1 than in Group-2 (177.5±118.9 and 100.2±87.3 mg / L, respectively; p=0.001). The appendix lumen diameter (p=0.002), appendix wall defect (p<0.001), peritoneal thickening and enhancement (p<0.001), ascites (p=0.031), intra-abdominal abscess (p=0.003), jejunal thickening (p=0.019), ileal thick-ening (p=0.008), and ileus (p=0.035) values were significantly higher in Group-1. In the binominal logistic regression analysis performed with statistically significant data, an appendiceal wall defect (OR: 0.069, 95% CI=0.014-0.327, p=0.001) and peritoneal thickening and enhancement (OR: 0.131, 95% CI=0.024-0.714, p=0.019) were identified as independent variables for perforated appendicitis. CONCLUSION: Among CT findings, appendix wall defects and peritoneal thickening and enhancement play an important role in detecting perforation.

Classification of retinoblastoma-1 gene mutation with machine learning-based models in bladder cancer.

Purpose: This study aims to evaluate the potential of machine learning algorithms built with radiomics features from computed tomography urography (CTU) images that classify RB1 gene mutation status in bladder cancer. Method: The study enrolled CTU images of 18 patients with and 54 without RB1 mutation from a public database. Image and data preprocessing were performed after data augmentation. Feature selection steps were consisted of filter and wrapper methods. Pearson's correlation analysis was the filter, and a wrapper-based sequential feature selection algorithm was the wrapper. Models with XGBoost, Random Forest (RF), and k-Nearest Neighbors (kNN) algorithms were developed. Performance metrics of the models were calculated. Models' performances were compared by using Friedman's test. Results: 8 features were selected from 851 total extracted features. Accuracy, sensitivity, specificity, precision, recall, F1 measure and AUC were 84%, 80%, 88%, 86%, 80%, 0.83 and 0.84, for XGBoost; 72%, 80%, 65%, 67%, 80%, 0.73 and 0.72 for RF; 66%, 53%, 76%, 67%, 53%, 0.60 and 0.65 for kNN, respectively. XGBoost model had outperformed kNN model in Friedman's test (p = 0.006). Conclusions: Machine learning algorithms with radiomics features from CTU images show promising results in classifying bladder cancer by RB1 mutation status non-invasively.

Predictors of malignancy in patients with solitary pulmonary nodules undergoing pulmonary resection.

BACKGROUND: The management of a solitary pulmonary nodule is a challenging issue in pulmonary disease. Although many factors have been defined as predictors for malignancy in solitary pulmonary nodules, the accurate diagnosis can only be established with the permanent histological diagnosis. OBJECTIVE: We tried to clarify the possible predictors of malignancy in solitary pulmonary nodules in patients who had definitive histological diagnosis. METHODS: We made a retrospective study to collect the data of patients with solitary pulmonary nodules who had histological diagnosis either before or after surgery. We made a statistical analysis of both the clinic and radiological features of these nodules with respect to malignancy both in contingency tables and with logistic regression analysis. RESULTS: We had a total of 223 patients with a radiological diagnosis of solitary pulmonary nodule. Age, smoking status and pack years of smoking, maximum standardized uptake value (SUVmax), and radiological features such as solid component, spiculation, pleural tag, lobulation, calcification, and higher density were significant predictors of malignancy in contingency tables. Age, smoking status and smoking (pack/year), SUVmax, and radiological features including spiculation, pleural tag, lobulation, calcification, and higher density were the significant predictors in univariate analysis. However, multivariate analysis revealed only SUVmax greater than 2.5 (p < 0.0001), spiculation (p = 0.009), and age older than 61 years (p = 0.015) as the significant predictors for malignancy. CONCLUSION: Age, SUVmax, and spiculation are the independent predictors of malignancy in patients with solitary pulmonary nodules.