Role of Computer Aided Diagnosis (CAD) in the detection of pulmonary nodules on 64 row multi detector computed tomography

Unsupervised contrastive learning based transformer for lung nodule detection

Objective.Early detection of lung nodules with computed tomography (CT) is critical for the longer survival of lung cancer patients and better quality of life. Computer-aided detection/diagnosis (CAD) is proven valuable as a second or concurrent reader in this context. However, accurate detection of lung nodules remains a challenge for such CAD systems and even radiologists due to not only the variability in size, location, and appearance of lung nodules but also the complexity of lung structures. This leads to a high false-positive rate with CAD, compromising its clinical efficacy.Approach.Motivated by recent computer vision techniques, here we present a self-supervised region-based 3D transformer model to identify lung nodules among a set of candidate regions. Specifically, a 3D vision transformer is developed that divides a CT volume into a sequence of non-overlap cubes, extracts embedding features from each cube with an embedding layer, and analyzes all embedding features with a self-attention mechanism for the prediction. To effectively train the transformer model on a relatively small dataset, the region-based contrastive learning method is used to boost the performance by pre-training the 3D transformer with public CT images.Results.Our experiments show that the proposed method can significantly improve the performance of lung nodule screening in comparison with the commonly used 3D convolutional neural networks.Significance.This study demonstrates a promising direction to improve the performance of current CAD systems for lung nodule detection.

Comparison of Maximum Intensity Projection and Volume Rendering in Detecting Pulmonary Nodules on Multidetector Computed Tomography

Introduction Lung cancer is the most common cancer overall, and the foremost cause of cancer-related mortality. Almost all lung cancers evolve from pulmonary nodules. As multidetector CT (MDCT) scanners are now widely available, there is an increased rate of detection of pulmonary nodules. It is of utmost importance to evaluate pulmonary nodules to rule out the possibility of neoplastic diseases. With advancements in technology, there are various manual and automatic analytic software providing a wide range of post-processing techniques. Maximum intensity projection (MIP) and volume rendering (VR) techniques have been analyzed previously regarding pulmonary nodules but there is a scarcity of data in terms of low-density nodules. This study aims to delineate the comparison and supremacy of both techniques in terms of low-density nodules. Methodology The current prospective study was conducted from June 2019 to June 2020 in the Radiology Department at Dr. Ziauddin Hospital, Karachi. Chest CT scans were performed on 16 slice MDCT (Alexion 16 Multi-slice, Toshiba Medical System Corporation, Houston, TX). A consultant radiologist of six years experience and a postgraduate trainee of three years experience analyzed each patient on a workstation (Vitrea 6.2.0, Vital Images, Minnetonka, MN). SPSS 23.0 (SPSS Inc., Chicago, IL) was incorporated for data analysis. Data were expressed in the median and interquartile range (IQR). Data collected for this study were analyzed using analyzing the median difference in nodule count using Wilcoxon's signed-rank test. A p-value of <0.05 was considered significant. Results After informed consent, 236 patients were recruited for the study. MIP outperformed VR in terms of nodule detection and low-density nodules at each evaluated slab thicknesses (p<0.001). A 10-mm MIP was superior to all other techniques in terms of detection of pulmonary nodules and low-density nodules (p<0.001). MIP was also considered an easier technique as there was excellent inter-rater reliability and agreement. Conclusion This study is robust evidence regarding the supremacy of MIP. MIP outperformed VR on every slab thicknesses. The 10-mm MIP technique was superior to all others evaluated and was recorded to be an easier analyzing technique.

Computer-Aided Detection of Pulmonary Nodules in Computed Tomography Using ClearReadCT

This study evaluates the accuracy of a computer-aided detection (CAD) application for pulmonary nodular lesions (PNL) in computed tomography (CT) scans, the ClearReadCT (Riverain Technologies). The study was retrospective for 106 biopsied PNLs from 100 patients. Seventy-five scans were Contrast-Enhanced (CECT) and 25 received no enhancer (NECT). Axial reconstructions in soft-tissue and lung kernel were applied at three different slice thicknesses, 0.75 mm (CECT/NECT n = 25/6), 1.5 mm (n = 18/9) and 3.0 mm (n = 43/18). We questioned the effect of (1) enhancer, (2) kernel and (3) slice thickness on the CAD performance. Our main findings are: (1) Vessel suppression is effective and specific in both NECT and CECT. (2) Contrast enhancement significantly increased the CAD sensitivity from 60% in NECT to 80% in CECT, P = 0.025 Fischer's exact test. (3) The CAD sensitivity was 84% in 3 mm slices compared to 68% in 0.75 mm slices, P > 0.2 Fischer's exact test. (4) Small lesions of low attenuation were detected with higher sensitivity. (5) Lung kernel reconstructions increased the false positive rate without affecting the sensitivity (P > 0.05 McNemar's test). In conclusion, ClearReadCT showed an optimized sensitivity of 84% and a positive predictive value of 67% in enhanced lung scans with thick, soft kernel reconstructions. NECT, thin slices and lung kernel reconstruction were associated with inferior performance.

Differential diagnostic value of 64-slice spiral computed tomography in solitary pulmonary nodule

The present study aimed to evaluate the diagnostic value of 64-slice spiral multivariate computed tomography (CT) combined with dynamic contrast-enhanced scanning for benign and malignant solitary pulmonary nodules (SPNs). A total of 93 patients with SPN as diagnosed by CT were included. All these patients were subjected to routine and dynamic enhancement CT scanning. After reconstruction, the morphological characteristics following dynamic enhancement were analyzed, and compared for the benign and malignant SPN cases. The incidences of lobulation, spicular sign, pleural indentation and vacuole sign in the malignant SPN group were significantly higher compared with the benign SPN group. During the dynamic enhancement scanning, the CT values at all the time points for the inflammatory and malignant SPN groups were significantly higher than the benign SPN group. No significant differences were observed in the dynamic enhancement CT values at 30, 60, 90 and 120 sec between the inflammatory, and malignant SPN groups. However, in the inflammatory SPN group, the dynamic enhancement CT values at 300 and 540 sec were significantly lower than the malignant SPN group. Notably, the diagnostic accordance rate for the morphological signs combined with dynamic enhancement diagnosis was significantly higher than the morphological signs alone. The 64-slice spiral CT morphological signs combined with dynamic enhancement detection can be more effective for the differential diagnosis of benign and malignant SPN, which may provide potent evidence for the early clinical treatment.