Automatic Detection and Classification of Multiple Catheters in Neonatal Radiographs with Deep Learning

Abstract

We develop and evaluate a deep learning algorithm to classify multiple catheters on neonatal chest and abdominal radiographs. A convolutional neural network (CNN) was trained using a dataset of 777 neonatal chest and abdominal radiographs, with a split of 81%-9%-10% for training-validation-testing, respectively. We employed ResNet-50 (a CNN), pre-trained on ImageNet. Ground truth labelling was limited to tagging each image to indicate the presence or absence of endotracheal tubes (ETTs), nasogastric tubes (NGTs), and umbilical arterial and venous catheters (UACs, UVCs). The data set included 561 images containing 2 or more catheters, 167 images with only one, and 49 with none. Performance was measured with average precision (AP), calculated from the area under the precision-recall curve. On our test data, the algorithm achieved an overall AP (95% confidence interval) of 0.977 (0.679-0.999) for NGTs, 0.989 (0.751-1.000) for ETTs, 0.979 (0.873-0.997) for UACs, and 0.937 (0.785-0.984) for UVCs. Performance was similar for the set of 58 test images consisting of 2 or more catheters, with an AP of 0.975 (0.255-1.000) for NGTs, 0.997 (0.009-1.000) for ETTs, 0.981 (0.797-0.998) for UACs, and 0.937 (0.689-0.990) for UVCs. Our network thus achieves strong performance in the simultaneous detection of these four catheter types. Radiologists may use such an algorithm as a time-saving mechanism to automate reporting of catheters on radiographs.

Combining convolutional neural networks and spatial-channel “squeeze and excitation” block for multiple-label image classification

“…Let 𝐹 = [𝑓 1,1 , 𝑓 1,2 , … , 𝑓 𝑖,𝑗 … , 𝑓 𝐻,𝑊 ] represent the slice on the spatial dimension, where spatial location (𝑖, 𝑗). In this experiment, we adopted the original sSE as shown in (8) and applied BN after the sigmoid function as shown in (9) for this experiment.…”

“…The classification methods for tubes have been proposed using various methods, including rule-based [4], and decision trees [5]. Other classification methods based on deep learning (DL) such as, Alexnet [6], GoogLeNet [6], UNET [7], ResNet [8], and EfficientNet [9]. However, these methods are still not optimal in multiple classification tasks, these methods are still inefficient because image classification involves multiple tubes from the current intubation depending upon the patient.…”

Neonatal Catheter and Tube Placement and Radiographic Assessment Statistics in Relation to Important Anatomic Landmarks

“…Our dataset is drawn from that which we used in reference 29 (henceforth referred to as Paper I), which was originally designed for training a deep learning model for automatic catheter detection. 29 We obtained de-identified chest/ abdominal radiographs of neonates and young infants (acquired using standard NICU protocols) at our local institution from a Picture Archive and Communication System (PACS). The data were drawn at random from NICU cases in 2014 through 2016 in which the patients were recorded as less than 60 days of age at the time of the radiograph.…”

Classification of Catheters and Tubes on Chest Radiographs Using Light-Weight Deep Convolutional Neural Networks

“…However, there are few aims in analyzing multiple different types of catheters [7]. One study has constructed a multi-label CNN that accurately detects the presence or absence of four catheters of interest on neonatal chest radiographs on a small dataset [9]. To increase exibility for subsequent model expansion, we developed multiple models (single-label network) to identify the endotracheal tube (ETT), the central venous catheter (CVC), and the nasogastric tube (NGT), respectively on a large dataset of around 10,000 chest X-rays, presenting a robust and practical solution to this problem.…”