Automatic and accurate segmentation of pulmonary nodules in CT images can help physicians perform more accurate quantitative analysis, diagnose diseases, and improve patient survival. In recent years, with the development of deep learning technology, pulmonary nodule segmentation methods based on deep neural networks have gradually replaced traditional segmentation methods. This paper reviews the recent pulmonary nodule segmentation algorithms based on deep neural networks. First, the heterogeneity of pulmonary nodules, the interpretability of segmentation results, and external environmental factors are discussed, and then the open-source 2D and 3D models in medical segmentation tasks in recent years are applied to the Lung Image Database Consortium and Image Database Resource Initiative (LIDC) and Lung Nodule Analysis 16 (Luna16) datasets for comparison, and the visual diagnostic features marked by radiologists are evaluated one by one. According to the analysis of the experimental data, the following conclusions are drawn: (1) In the pulmonary nodule segmentation task, the performance of the 2D segmentation models DSC is generally better than that of the 3D segmentation models. (2) 'Subtlety', 'Sphericity', 'Margin', 'Texture', and 'Size' have more influence on pulmonary nodule segmentation, while 'Lobulation', 'Spiculation', and 'Benign and Malignant' features have less influence on pulmonary nodule segmentation. (3) Higher accuracy in pulmonary nodule segmentation can be achieved based on better-quality CT images. (4) Good contextual information acquisition and attention mechanism design positively affect pulmonary nodule segmentation.
16 January 2024 in IEEE Journal of Biomedical and Health Informatics
Lung cancer is one of the deadliest cancers globally, and early diagnosis is crucial for patient survival. Pulmonary nodules are the main manifestation of early lung cancer, usually assessed using CT scans. Nowadays, computer-aided diagnostic systems are widely used to assist physicians in disease diagnosis. The accurate segmentation of pulmonary nodules is affected by internal heterogeneity and external data factors. In order to overcome the segmentation challenges of subtle, mixed, adhesion-type, benign, and uncertain categories of nodules, a new mixed manual feature network that enhances sensitivity and accuracy is proposed. This method integrates feature information through a dual-branch network framework and multi-dimensional fusion module. By training and validating with multiple data sources and different data qualities, our method demonstrates leading performance on the LUNA16, Multi-thickness Slice Image dataset, LIDC, and UniToChest, with Dice similarity coefficients reaching 86.89%, 75.72%, 84.12%, and 80.74% respectively, surpassing most current methods for pulmonary nodule segmentation. Our method further improved the accuracy, reliability, and stability of lung nodule segmentation tasks even on challenging CT scans. The code used in this study is published on GitHub and is available at the following URL(https://github.com/BITEWKRER/DBNet).
Content‐based medical image retrieval (CBMIR) systems attempt to search medical image database to narrow the semantic gap in medical image analysis. The efficacy of high‐level medical information representation using features is a major challenge in CBMIR systems. Features play a vital role in the accuracy and speed of the search process. In this paper, we propose a deep convolutional neural network‐ (CNN‐) based framework to learn concise feature vector for medical image retrieval. The medical images are decomposed into five components using empirical mode decomposition (EMD). The deep CNN is trained in a supervised way with multicomponent input, and the learned features are used to retrieve medical images. The IRMA dataset, containing 11,000 X‐ray images, 116 classes, is used to validate the proposed method. We achieve a total IRMA error of 43.21 and a mean average precision of 0.86 for retrieval task and IRMA error of 68.48 and F1 measure of 0.66 on classification task, which is the best result compared with existing literature for this dataset.
10 July 2026 in Biomedical Signal Processing and Control
The worldwide incidence of breast cancer continues to increase annually. Accurate classification of breast cancer subtypes is crucial for improving treatment precision. However, the diagnosis of breast cancer is time-consuming and labor-intensive. Although the introduction of Computer-Aided Diagnosis (CAD) systems can assist doctors in diagnosing breast cancer more effectively, establishing high-quality medical image datasets is challenging, significantly hampering the development of CAD systems. This paper proposes a meta-learning method suitable for the multi-classification of breast cancer histopathological images. This method effectively addresses the issue of limited data by adopting the concept of transfer learning and incorporates group convolution within the model to enhance efficiency and feature extraction capability. Experimental results demonstrate that this method achieves a classification accuracy of 96.30% on the BreakHis dataset across eight categories, surpassing state-of-the-art methods. Furthermore, we validated the generalization performance of this method on another publicly available dataset, Kather-CRC-2016.
Cells are capable of conveying rich information about their microenvironment. Automated cell nucleus segmentation technology not only alleviates the workload of pathologists but also provides accurate microenvironment data applied to biological and clinical research. Existing deep learning models show excellent performance given a large volume of labeled data, they still usually require some manual labeling for domain-specific training when faced with data from unknown fields. Unfortunately, obtaining precise annotations of histopathological images is an extremely challenging task that is not only highly dependent on expert knowledge but also takes a significant amount of time. In this study, we aim to develop a general cell nucleus model for segmentation that requires a smaller amount of data and has broader applicability. To this end, we designed a Dual Encoder U-net model based on Meta Transfer Learning (MTL), which can learn effectively with fewer training samples. Our network employs a Meta Transfer Learning optimization algorithm and introduces a feature fusion module to integrate deep and shallow feature information. Experimental results show that our proposed MTLDEUnet model not only enhances the generalization ability of the model but also achieves performance comparable to the current highly advanced models trained with a limited sample set.