Multiple Resolution Residually Connected Feature Streams for Automatic Lung Tumor Segmentation From CT Images

Abstract

Volumetric lung tumor segmentation and accurate longitudinal tracking of tumor volume changes from computed tomography (CT) images are essential for monitoring tumor response to therapy. Hence, we developed two multiple resolution residually connected network (MRRN) formulations called incremental-MRRN and dense-MRRN. Our networks simultaneously combine features across multiple image resolution and feature levels through residual connections to detect and segment lung tumors. We evaluated our method on a total of 1210 non-small cell (NSCLC) lung tumors and nodules from three datasets consisting of 377 tumors from the open-source Cancer Imaging Archive (TCIA), 304 advanced stage NSCLC treated with anti- PD-1 checkpoint immunotherapy from internal institution MSKCC dataset, and 529 lung nodules from the Lung Image Database Consortium (LIDC). The algorithm was trained using the 377 tumors from the TCIA dataset and validated on the MSKCC and tested on LIDC datasets. The segmentation accuracy compared to expert delineations was evaluated by computing the Dice Similarity Coefficient (DSC), Hausdorff distances, sensitivity and precision metrics. Our best performing incremental-MRRN method produced the highest DSC of 0.74±0.13 for TCIA, 0.75±0.12 for MSKCC and 0.68±0.23 for the LIDC datasets. There was no significant difference in the estimations of volumetric tumor changes computed using the incremental-MRRN method compared with expert segmentation. In summary, we have developed a multi-scale CNN approach for volumetrically segmenting lung tumors which enables accurate, automated identification of and serial measurement of tumor volumes in the lung.

Inf-Net: Automatic COVID-19 Lung Infection Segmentation from CT Images

“…However, the similar visual appearances of nodules and background makes it difficult for extracting the nodule regions. To overcome this issue, several deep learning algorithms have been proposed to learn a powerful visual representations [28]- [30]. For instance, Wang et al [28] developed a central focused convolutional neural network to segment lung nodules from heterogeneous CT slices.…”

“…Jin et al [29] utilized GAN-synthesized data to improve the training of a discriminative model for pathological lung segmentation. Jiang et al [30] designed two deep networks to segment lung tumors from CT slices by adding multiple residual streams of varying resolutions. Wu et al [31] built an explainable COVID-19 diagnosis system by joint classification and segmentation.…”

Library of model implementations for sharing deep-learning image segmentation and outcomes models

“…Tables 1-3 list various models implemented for image segmentation, radiotherapy, and radiomics based outcomes prediction. Image segmentation models include the popular Deeplab [16] network for segmentation of prostate, heart sub structures, chewing and swallowing structures as well as specialized architectures such as MRRN [17] for segmentation of lung nodules. Radiotherapy and radiomics models derived with a sufficient number of patients as well as satisfactory performance on validation dataset were included.…”

Mixed-Supervised Dual-Network for Medical Image Segmentation

“…Image segmentation is an important application of medical image analysis. Recently, deep learning based methods [1,2,3,4] have achieved remarkable success in many medical image segmentation tasks, such as brain tumor and lung nodule segmentation. However, all these methods require a large amount of training data with high-quality dense annotations to train, which is very expensive and time-consuming to prepare.…”