Automated detection and segmentation of thoracic lymph nodes from CT using 3D foveal fully convolutional neural networks.

BACKGROUND: In oncology, the correct determination of nodal metastatic disease is essential for patient management, as patient treatment and prognosis are closely linked to the stage of the disease. The aim of the study was to develop a tool for automatic 3D detection and segmentation of lymph nodes (LNs) in computed tomography (CT) scans of the thorax using a fully convolutional neural network based on 3D foveal patches. METHODS: The training dataset was collected from the Computed Tomography Lymph Nodes Collection of the Cancer Imaging Archive, containing 89 contrast-enhanced CT scans of the thorax. A total number of 4275 LNs was segmented semi-automatically by a radiologist, assessing the entire 3D volume of the LNs. Using this data, a fully convolutional neuronal network based on 3D foveal patches was trained with fourfold cross-validation. Testing was performed on an unseen dataset containing 15 contrast-enhanced CT scans of patients who were referred upon suspicion or for staging of bronchial carcinoma. RESULTS: The algorithm achieved a good overall performance with a total detection rate of 76.9% for enlarged LNs during fourfold cross-validation in the training dataset with 10.3 false-positives per volume and of 69.9% in the unseen testing dataset. In the training dataset a better detection rate was observed for enlarged LNs compared to smaller LNs, the detection rate for LNs with a short-axis diameter (SAD) ≥ 20 mm and SAD 5-10 mm being 91.6% and 62.2% (p < 0.001), respectively. Best detection rates were obtained for LNs located in Level 4R (83.6%) and Level 7 (80.4%). CONCLUSIONS: The proposed 3D deep learning approach achieves an overall good performance in the automatic detection and segmentation of thoracic LNs and shows reasonable generalizability, yielding the potential to facilitate detection during routine clinical work and to enable radiomics research without observer-bias.

Endoscopic Lung Volume Reduction Using Endobronchial Valves in Patients with Severe Emphysema and Very Low FEV1.

BACKGROUND: Patients with a forced expiratory volume in 1 s (FEV1) below 20% of the predicted normal values (pred.) and either homogeneous emphysema or low diffusing capacity for carbon monoxide (DLCO) have a high risk for adverse events including death when undergoing surgical lung volume reduction. OBJECTIVES: We hypothesized that selected patients can benefit from endoscopic lung volume reduction (eLVR) despite a very low FEV1. METHODS: This study is a retrospective analysis of consecutive patients with severe airflow obstruction, an FEV1 ≤20% of pred., and low DLCO who were treated by eLVR with endobronchial valves (EBV) between June 2012 and January 2015. Pre- and postinterventional lung function parameters, the 6-min walking test (6-MWT) distance, adverse events, and follow-up were recorded. RESULTS: In 20 patients, there was an overall improvement in lung function with an increase in FEV1 (16.97-21.03% of pred.) and a decrease in residual volume (322-270% of pred.) and total lung capacity (144-129.06% of pred.). The 6-MWT distance improved (from 239 ± 77 to 267± 97 m overall, and from 184 ± 50 to 237 ± 101 m if patients developed an atelectasis of the target lobe). Pneumothorax occurred in 5 of the 20 patients (25%). 30-day mortality was 0%, and all patients survived to discharge. CONCLUSIONS: The patients benefitted moderately from EBV treatment despite an initially low FEV1. Some patients improved remarkably. EBV treatment in patients with an FEV1 ≤20% of pred. is generally feasible and safe. The greatest risk is pneumothorax with prolonged chest tube duration.