Detection-Guided Deep Learning-Based Model with Spatial Regularization for Lung Nodule Segmentation

TL;DR

This paper tackles lung nodule segmentation in CT by proposing a multitask deep learning model that jointly performs segmentation and classification. It embeds a ResNet-50 classifier with a ResU-Net segmentation network and employs feature combination blocks to share information, augmented by STD-based spatial regularization guided by classification priors. The authors demonstrate superior performance over several baselines on a small clinical dataset, with gains augmented by pre-training on LUNA16 and a fully fine-tuned transfer-learning strategy. The approach offers a practical path to robust, accurate nodule delineation in settings with limited annotated data, potentially aiding radiologists in diagnosis and treatment planning.

Abstract

Lung cancer ranks as one of the leading causes of cancer diagnosis and is the foremost cause of cancer-related mortality worldwide. The early detection of lung nodules plays a pivotal role in improving outcomes for patients, as it enables timely and effective treatment interventions. The segmentation of lung nodules plays a critical role in aiding physicians in distinguishing between malignant and benign lesions. However, this task remains challenging due to the substantial variation in the shapes and sizes of lung nodules, and their frequent proximity to lung tissues, which complicates clear delineation. In this study, we introduce a novel model for segmenting lung nodules in computed tomography (CT) images, leveraging a deep learning framework that integrates segmentation and classification processes. This model is distinguished by its use of feature combination blocks, which facilitate the sharing of information between the segmentation and classification components. Additionally, we employ the classification outcomes as priors to refine the size estimation of the predicted nodules, integrating these with a spatial regularization technique to enhance precision. Furthermore, recognizing the challenges posed by limited training datasets, we have developed an optimal transfer learning strategy that freezes certain layers to further improve performance. The results show that our proposed model can capture the target nodules more accurately compared to other commonly used models. By applying transfer learning, the performance can be further improved, achieving a sensitivity score of 0.885 and a Dice score of 0.814.

Figures (4)

• Figure 1: Examples of nodules from LUNA16. From left to right: isolated nodule, juxtapleural nodule, calcific nodule, ground-glass opacity (GGO) nodule and cavitary nodule.
• Figure 2: The architecture of the proposed multitask model. The orange blocks correspond to ResNet-50 for classification. The blue blocks correspond to an ResU-Net based network for segmentation. An input of size $128\times 128$ is used for demonstration.
• Figure 3: Structure of our combination block.
• Figure 4: Visual comparison of segmentation results for different methods, including four examples in the CCF dataset. The dice score of each result is shown on the upper left of each prediction.