Coronary artery segmentation from intravascular optical coherence tomography using deep capsules.

The segmentation and analysis of coronary arteries from intravascular optical coherence tomography (IVOCT) is an important aspect of diagnosing and managing coronary artery disease. Current image processing methods are hindered by the time needed to generate expert-labelled datasets and the potential for bias during the analysis. Therefore, automated, robust, unbiased and timely geometry extraction from IVOCT, using image processing, would be beneficial to clinicians. With clinical application in mind, we aim to develop a model with a small memory footprint that is fast at inference time without sacrificing segmentation quality. Using a large IVOCT dataset of 12,011 expert-labelled images from 22 patients, we construct a new deep learning method based on capsules which automatically produces lumen segmentations. Our dataset contains images with both blood and light artefacts (22.8 %), as well as metallic (23.1 %) and bioresorbable stents (2.5 %). We split the dataset into a training (70 %), validation (20 %) and test (10 %) set and rigorously investigate design variations with respect to upsampling regimes and input selection. We show that our developments lead to a model, DeepCap, that is on par with state-of-the-art machine learning methods in terms of segmentation quality and robustness, while using as little as 12 % of the parameters. This enables DeepCap to have per image inference times up to 70 % faster on GPU and up to 95 % faster on CPU compared to other state-of-the-art models. DeepCap is a robust automated segmentation tool that can aid clinicians to extract unbiased geometrical data from IVOCT.

Conventional Transbronchial Needle Aspiration: The Original Guard Who Still Has a Role in Mediastinal Lymph Node Sampling.

INTRODUCTION: Conventional transbronchial needle aspiration (C-TBNA) is the originally described method for sampling mediastinal lymph nodes (MLN). After the advent of endobronchial ultrasound, the practice and reports of C-TBNA have dwindled. We report a large series of C-TBNA from the Indian subcontinent, highlighting aspects such as pathological spectrum, yield and complications, and reiterating its relevance in MLN sampling. METHODS: The study population included 400 consecutive patients over 6.8 years who had C-TBNA done for MLN ≥1 cm in size. C-TBNA was done using a 19-G needle, with conscious sedation. A maximum of 7 passes per node were done. Rapid-on-site evaluation was done in >95% cases. Lymph nodes sampled were labeled "adequate" if lymphocytes were present, and "diagnostic" if a definitive diagnosis was made. RESULTS: The study included 228 males and 172 females, mean age 49.4±14.7 years. The "adequacy" rate was 383/400 (95.75%), and "diagnostic" yield was 347/400 (86.75%). C-TBNA was the sole diagnostic modality in 215/400 (53.75%) patients. The diagnoses included tuberculosis (43%), sarcoidosis (25.5%) and malignancy (18.25%). Complications were rare. CONCLUSIONS: This is one of the largest studies of C-TBNA in literature, and one of the few studies to define accurate pathologic diagnosis of enlarged MLN in India. This is also the one of the largest series to define the yield of TBNA with rapid-on-site evaluation in MLN sampling. Currently, in many parts of the world, C-TBNA is still the most common MLN sampling procedure, from an availability, expertise, economic, and safety perspective.