Thyroid ultrasound diagnosis improvement via multi-view self-supervised learning and two-stage pre-training

Thyroid nodule classification and segmentation in ultrasound images are crucial for computer-aided diagnosis; however, they face limitations owing to insufficient labeled data. In this study, we proposed a multi-view contrastive self-supervised method to improve thyroid nodule classification and segmentation performance with limited manual labels. Our method aligns the transverse and longitudinal views of the same nodule, thereby enabling the model to focus more on the nodule area. We designed an adaptive loss function that eliminates the limitations of the paired data. Additionally, we adopted a two-stage pre-training to exploit the pre-training on ImageNet and thyroid ultrasound images. Extensive experiments were conducted on a large-scale dataset collected from multiple centers. The results showed that the proposed method significantly improves nodule classification and segmentation performance with limited manual labels and outperforms state-of-the-art self-supervised methods. The two-stage pre-training also significantly exceeded ImageNet pre-training.

Deep Learning for the Classification of Small (≤2 cm) Pulmonary Nodules on CT Imaging: A Preliminary Study

RATIONALE AND OBJECTIVES

We aimed to present a deep learning-based malignancy prediction model (CT-lungNET) that is simpler and faster to use in the diagnosis of small (≤2 cm) pulmonary nodules on nonenhanced chest CT and to preliminarily evaluate its performance and usefulness for human reviewers.

MATERIALS AND METHODS

A total of 173 whole nonenhanced chest CT images containing 208 pulmonary nodules (94 malignant and 11 benign nodules) ranging in size from 5 mm to 20 mm were collected. Pathologically confirmed nodules or nodules that remained unchanged for more than 1 year were included, and 30 benign and 30 malignant nodules were randomly assigned into the test set. We designed CT-lungNET to include three convolutional layers followed by two fully-connected layers and compared its diagnostic performance and processing time with those of AlexNET by using the area under the receiver operating curve (AUROC). An observer performance test was conducted involving eight human reviewers of four different groups (medical students, physicians, radiologic residents, and thoracic radiologists) at test 1 and test 2, referring to the CT-lungNET's malignancy prediction rate with pairwise comparison receiver operating curve analysis.

RESULTS

CT-lungNET showed an improved AUROC (0.85; 95% confidence interval: 0.74-0.93), compared to that of the AlexNET (0.82; 95% confidence interval: 0.71-0.91). The processing speed per one image slice for CT-lungNET was about 10 times faster than that for AlexNET (0.90 vs. 8.79 seconds). During the observer performance test, the classification performance of nonradiologists was increased with the aid of CTlungNET, (mean AUC improvement: 0.13; range: 0.03-0.19) but not significantly so in the radiologists group (mean AUC improvement: 0.02; range: -0.02 to 0.07).

CONCLUSION

CT-lungNET was able to provide better classification results with a significantly shorter amount of processing time as compared to AlexNET in the diagnosis of small pulmonary nodules on nonenhanced chest CT. In this preliminary observer performance test, CT-lungNET may have a role acting as a second reviewer for less experienced reviewers, resulting in enhanced performance in the diagnosis of early lung cancer.

Insights for Management of Ground-Glass Opacities From the National Lung Screening Trial

BACKGROUND

In the National Lung Screening Trial (NLST), screen-detected cancers that would not have been identified by the Lung Computed Tomographic Screening Reporting and Data System (Lung-RADS) nodule management guidelines were frequently ground-glass opacities (GGOs). Lung-RADS suggests that GGOs with diameter less than 20 mm return for annual screening, and GGOs greater than or equal to 20 mm receive 6-month follow-up. We examined whether this 20-mm threshold gives consistent management of GGOs compared with solid nodules.

METHODS

First, we calculated diameter-specific malignancy probabilities for GGOs and solid nodules in the NLST. Using the solid-nodule malignancy risks as benchmarks, we suggested risk-based management categories for GGOs based on their probability of malignancy. Second, we compared lung-cancer mortality between GGOs and solid nodules in the same risk-based category.

RESULTS

Using the Lung-RADS v1.0 classifications, malignancy probability is higher for GGOs than solid nodules within the same category. A risk-based classification of GGOs would assign annual screening for GGOs 4 to 5 mm (0.4% malignancy risk); 6-month follow-up for GGOs 6 to 7 mm (1.1%), 8 to 14 mm (3.0%), and 15 to 19 mm (5.2%); and 3-month follow-up for greater than or equal to 20 mm (10.9%). This reclassification would have assigned similarly fatal cancers to 3-month follow-up (hazard ratio = 2.0 for lung-cancer death in GGOs versus solid-nodule cancers, 95% confidence interval: 0.4-8.7), but for 6-month follow-up, mortality was lower in GGO cancers (hazard ratio = 0.18, 95% confidence interval: 0.05-0.67).

CONCLUSIONS

If Lung-RADS categories for GGOs were based on malignancy probability, then 6- to 19-mm GGOs would receive 6-month follow-up and greater than or equal to 20-mm GGOs would receive 3-month follow-up. Such risk-based management for GGOs could improve the sensitivity of Lung-RADS, especially for large GGO cancers. However, small GGO cancers were less aggressive than their solid-nodule counterparts.

An Appraisal of Nodule Diagnosis for Lung Cancer in CT Images

As "the second eyes" of radiologists, computer-aided diagnosis systems play a significant role in nodule detection and diagnosis for lung cancer. In this paper, we aim to provide a systematic survey of state-of-the-art techniques (both traditional techniques and deep learning techniques) for nodule diagnosis from computed tomography images. This review first introduces the current progress and the popular structure used for nodule diagnosis. In particular, we provide a detailed overview of the five major stages in the computer-aided diagnosis systems: data acquisition, nodule segmentation, feature extraction, feature selection and nodule classification. Second, we provide a detailed report of the selected works and make a comprehensive comparison between selected works. The selected papers are from the IEEE Xplore, Science Direct, PubMed, and Web of Science databases up to December 2018. Third, we discuss and summarize the better techniques used in nodule diagnosis and indicate the existing future challenges in this field, such as improving the area under the receiver operating characteristic curve and accuracy, developing new deep learning-based diagnosis techniques, building efficient feature sets (fusing traditional features and deep features), developing high-quality labeled databases with malignant and benign nodules and promoting the cooperation between medical organizations and academic institutions.