Objectives: To compare the diagnostic performance of a newly developed artificial intelligence (AI) algorithm derived from the fusion of convolution neural networks (CNN) versus human observers in the estimation of malignancy risk in pulmonary nodules. Methods: The study population consists of 158 nodules from 158 patients. All nodules (81 benign and 77 malignant) were determined to be malignant or benign by a radiologist based on pathologic assessment and/or follow-up imaging. Two radiologists and an AI platform analyzed the nodules based on the Lung-RADS classification. The two observers also noted the size, location, and morphologic features of the nodules. An intraclass correlation coefficient was calculated for both observers and the AI; ROC curve analysis was performed to determine diagnostic performances. Results: Nodule size, presence of spiculation, and presence of fat were significantly different between the malignant and benign nodules (p < 0.001, for all three). Eighteen (11.3%) nodules were not detected and analyzed by the AI. Observer 1, observer 2, and the AI had an AUC of 0.917 ± 0.023, 0.870 ± 0.033, and 0.790 ± 0.037 in the ROC analysis of malignity probability, respectively. The observers were in almost perfect agreement for localization, nodule size, and lung-RADS classification [κ (95% CI)=0.984 (0.961–1.000), 0.978 (0.970–0.984), and 0.924 (0.878–0.970), respectively]. Conclusion: The performance of the fusion AI algorithm in estimating the risk of malignancy was slightly lower than the performance of the observers. Fusion AI algorithms might be applied in an assisting role, especially for inexperienced radiologists. Advances in knowledge: In this study, we proposed a fusion model using four state-of-art object detectors for lung nodule detection and discrimination. The use of fusion of deep learning neural networks might be used in a supportive role for radiologists when interpreting lung nodule discrimination.

We aimed to evaluate the diagnostic accuracy and safety of computed tomography (CT)-guided biopsy of pulmonary lesions with fine needle aspiration (FNA) using non-coaxial technique. We analyzed 442 patients who underwent CT-guided lung biopsy with FNA and non-coaxial technique to determine the diagnostic outcomes, complication rates, and independent risk factors for diagnostic failure and pneumothorax. Diagnostic accuracy, sensitivity, and specificity were 97.6%, 97.3%, and 100%, respectively. Age and >35 mm lesion size were significant risk factors for diagnostic failure. The rates of pneumothorax and chest tube placement were 19% and 2.9%, respectively. Middle and lower lobe location, lesion to pleura distance >7.5 mm, and >45° needle trajectory angle were significant risk factors for pneumothorax. CT-guided FNA of pulmonary lesions with non-coaxial technique is a safe and reliable method with a relatively low pneumothorax rate and an acceptably high diagnostic accuracy.

Dusunceli Atman E, et al. Korean J Radiol. 2015 Nov-Dec;16(6):1240-1252. http://dx.doi.org/10.3348/kjr.2015.16.6.1240

The purpose of this study was to investigate the relation between pancreatic steatosis and visceral adiposity. Furthermore, the study sought to explore the association between pancreatic steatosis, pancreas volume, hepatic steatosis, age, and sex in adults without prior history of pancreatic disease. The research also served to define a cut-off value of visceral fat tissue area (VFA) predicting fatty pancreas. CT scans of 98 living-liver donor transplant patients without prior history of pancreatic disease were evaluated for the presence of fatty pancreas. Pancreas volume, VFA, subcutaneous-total FA, VFA/TFA ratios of the patients with and without fatty pancreas were quantified with a semi-automated model on CT. Coexistence of hepatic steatosis was also recorded. VFA, TFA and VFA/TFA were significantly greater in the fatty group (p<0.001, p<0.001, p<0.001; respectively), and pancreatic steatosis was moderately correlated with VFA, VFA/TFA and TFA with the highest correlation coefficient with VFA (r=-0.715, r=-0.605, r=-0.573, respectively; p<0.001 for all). A cut-off value of VFA ≥ 107.2 cm2 estimates pancreatic steatosis with a sensitivity and specificity of 90% (95% CI=77-96%) and 87.9% (95% CI=77%-94%), respectively. Pancreas volume was higher in the fatty-group with a mean value of 86.5±17.3 mL (range; 58-119.2 mL, p=0.097). In multiple logistic regression analyses, pancreatic steatosis was significantly associated with VFA and the male sex (OR=58.2, 95% CI=12.2-277.1, p<0.001; OR=11.4, 95% CI=2.1-63.4, p<0.001; respectively). 77.5% of the fatty pancreas subjects had co-existing hepatic steatosis. Pancreatic steatosis is related to higher VFA, VFA/TFA and hepatic steatosis. A cut-off value of VFA ≥ 107.2 cm2 may predict pancreatic steatosis.