Automatic Segmentation of Lumen Intima Layer in Longitudinal Mode Ultrasound Images.

We propose an automated method for the segmentation of lumen intima layer of the common carotid artery in longitudinal mode ultrasound images. The method is hybrid, in the sense that a coarse segmentation is first achieved by optimizing a locally defined contrast function of an active oblong considering its five degrees-of-freedom, and subsequently the fine segmentation and delineation of the carotid artery are achieved by post-processing the portion of the ultrasound image spanned by the annulus region of the optimally fitted active oblong. The post-processing includes median filtering and Canny edge detection to retain the lumen intima representative points followed by a smooth curve fitting technique to delineate the lumen intima boundary. The algorithm has been validated on 84 longitudinal mode carotid artery ultrasound images provided by the Signal Processing laboratory, Brno university. The proposed technique results in an average accuracy and Dice similarity index of 98.9% and 95.2%, respectively.

Deep learning, computer-aided radiography reading for tuberculosis: a diagnostic accuracy study from a tertiary hospital in India.

In general, chest radiographs (CXR) have high sensitivity and moderate specificity for active pulmonary tuberculosis (PTB) screening when interpreted by human readers. However, they are challenging to scale due to hardware costs and the dearth of professionals available to interpret CXR in low-resource, high PTB burden settings. Recently, several computer-aided detection (CAD) programs have been developed to facilitate automated CXR interpretation. We conducted a retrospective case-control study to assess the diagnostic accuracy of a CAD software (qXR, Qure.ai, Mumbai, India) using microbiologically-confirmed PTB as the reference standard. To assess overall accuracy of qXR, receiver operating characteristic (ROC) analysis was used to determine the area under the curve (AUC), along with 95% confidence intervals (CI). Kappa coefficients, and associated 95% CI, were used to investigate inter-rater reliability of the radiologists for detection of specific chest abnormalities. In total, 317 cases and 612 controls were included in the analysis. The AUC for qXR for the detection of microbiologically-confirmed PTB was 0.81 (95% CI: 0.78, 0.84). Using the threshold that maximized sensitivity and specificity of qXR simultaneously, the software achieved a sensitivity and specificity of 71% (95% CI: 66%, 76%) and 80% (95% CI: 77%, 83%), respectively. The sensitivity and specificity of radiologists for the detection of microbiologically-confirmed PTB was 56% (95% CI: 50%, 62%) and 80% (95% CI: 77%, 83%), respectively. For detection of key PTB-related abnormalities 'pleural effusion' and 'cavity', qXR achieved an AUC of 0.94 (95% CI: 0.92, 0.96) and 0.84 (95% CI: 0.82, 0.87), respectively. For the other abnormalities, the AUC ranged from 0.75 (95% CI: 0.70, 0.80) to 0.94 (95% CI: 0.91, 0.96). The controls had a high prevalence of other lung diseases which can cause radiological manifestations similar to PTB (e.g., 26% had pneumonia, 15% had lung malignancy, etc.). In a tertiary hospital in India, qXR demonstrated moderate sensitivity and specificity for the detection of PTB. There is likely a larger role for CAD software as a triage test for PTB at the primary care level in settings where access to radiologists in limited. Larger prospective studies that can better assess heterogeneity in important subgroups are needed.

Diagnostic reliability of the Thyroid Imaging Reporting and Data System (TI-RADS) in routine practice.

PURPOSE: To evaluate the diagnostic reliability of Thyroid Imaging Reporting and Data System (TI-RADS) classifications described by American College of Radiology (ACR) and Kwak et al. by calculating the risk of malignancy, to assess the role of TI-RADS in reducing fine-needle aspiration cytology (FNAC) of benign lesions. MATERIAL AND METHODS: This was a prospective study during the period from December 2017 to August 2018. Thyroid nodules were classified using ACR TI-RADS and TI-RADS proposed by Kwak et al. The TI-RADS categorisations were compared to the final diagnosis obtained by cytopathological/histopathological analysis. The risk of malignancy for each category was calculated. Sensitivity, specificity, and positive and negative predictive values for individual suspicious ultrasound features were also assessed. RESULTS: We evaluated a total of 127 thyroid nodules. The risk of malignancy was 0% in ACR TR1, 0% in ACR TR2, 6.9% in ACR TR3, 29.2% in ACR TR4, and 80% in ACR TR5 categories. The risk of malignancy for TI-RADS according to Kwak et al. were 0%, 0%, 21.5%, 32.4%, 100% for TI-RADS 2, 3, 4A, 4B, and 4C categories, respectively. Kwak TI-RADS 2 and 3 had higher sensitivity in predicting benignity compared to ACR TR1 and 2 (35.4% vs. 25.9%). CONCLUSIONS: We found TI-RADS classification to be a reliable, non-invasive, and practical method for assessing thyroid nodules in routine practice. TI-RADS can safely avert avoidable FNACs in a significant proportion of benign thyroid lesions.