Automatic lymph node segmentation is the cornerstone for advances in computer vision tasks for early detection and staging of cancer. Traditional segmentation methods are constrained by manual delineation and variability in operator proficiency, limiting their ability to achieve high accuracy. The introduction of deep learning technologies offers new possibilities for improving the accuracy of lymph node image analysis. This study evaluates the application of deep learning in lymph node segmentation and discusses the methodologies of various deep learning architectures such as convolutional neural networks, encoder-decoder networks, and transformers in analyzing medical imaging data across different modalities. Despite the advancements, it still confronts challenges like the shape diversity of lymph nodes, the scarcity of accurately labeled datasets, and the inadequate development of methods that are robust and generalizable across different imaging modalities. To the best of our knowledge, this is the first study that provides a comprehensive overview of the application of deep learning techniques in lymph node segmentation task. Furthermore, this study also explores potential future research directions, including multimodal fusion techniques, transfer learning, and the use of large-scale pre-trained models to overcome current limitations while enhancing cancer diagnosis and treatment planning strategies.

Spinal medical image segmentation is critical for diagnosing and treating spinal disorders. However, ambiguity in anatomical boundaries and interfering factors in medical images often cause segmentation errors. Current deep learning models cannot fully capture the intrinsic data properties, leading to unstable feature spaces. To tackle the above problems, we propose Verdiff-Net, a novel diffusion-based segmentation framework designed to improve segmentation accuracy and stability by learning the underlying data distribution. Verdiff-Net integrates a multi-scale fusion module (MSFM) for fine feature extraction and a noise semantic adapter (NSA) to refine segmentation masks. Validated across four multi-modality spinal datasets, Verdiff-Net achieves a high Dice coefficient of 93%, demonstrating its potential for clinical applications in precision spinal surgery.

Medical ultrasound (US) image segmentation faces significant challenges due to limited labeled data and characteristic imaging artifacts, including speckle noise and low-contrast boundaries. While semi-supervised learning (SSL) approaches have emerged to address data scarcity, existing methods suffer from suboptimal unlabeled data utilization and lack robust feature representation mechanisms. In this article, we propose Switch, a novel SSL framework with two key innovations: 1) a multiscale switch (MSS) strategy that employs hierarchical patch mixing to achieve uniform spatial coverage; and 2) a frequency-domain switch (FDS) with contrastive learning that performs amplitude switching in Fourier space for robust feature representations. Our framework integrates these components within a teacher–student architecture to effectively leverage both labeled and unlabeled data. Comprehensive evaluation across six diverse US datasets (lymph nodes, breast lesions, thyroid nodules, and prostate) demonstrates consistent superiority over state-of-the-art (SOTA) methods. At a 5% labeling ratio, Switch achieves remarkable improvements: 80.04% Dice on LN-INT, 85.52% Dice on DDTI, and 83.48% Dice on Prostate datasets, with our semi-supervised approach even exceeding fully supervised baselines. The method maintains parameter efficiency (1.8 M parameters) while delivering superior performance, validating its effectiveness for resource-constrained medical imaging applications. The source code is publicly available at https://github.com/jinggqu/Switch

This study addresses the limited mechanistic understanding behind medical imaging for tumor microenvironment (TME) assessment. We developed a novel framework that analyzes tumor imaging heterogeneity index (TIHI)-correlated genes to uncover underlying TME biology and therapeutic vulnerabilities. DCE-MRI and mRNA data from 987 high-risk breast cancer patients in the I-SPY2 trial, together with mRNA data from 508 patients in GSE25066, were analyzed. TIHI-associated genes were identified via Pearson correlation, clustered via weighted gene co-expression network analysis (WGCNA), and subgroups were defined via non-negative matrix factorization (NMF). The clinical relevance of the image-to-gene comprehensive (I2G-C) subtype defined by subgroups was assessed using logistic regression and Cox analysis. I2G-C comprised four clusters with distinct immune and replication/repair functions. It further stratified receptor, PAM50, and RPS5 subtypes. The “immune+/replication+” was more likely to achieve pathological complete response (pCR) (OR = 2.587, p < 0.001), while the “immune−/replication−” was the least likely to achieve pCR (OR = 0.402, p < 0.001). The “immune+/replication+” showed sensitivity to pembrolizumab (OR = 10.192, p < 0.001) and veliparib/carboplatin (OR = 5.184, p = 0.006), while “immune-/replication-” responded poorly to pembrolizumab (OR = 0.086, p < 0.001). Additionally, “immune+/replication-” had the best distant recurrence-free survival (DRFS), whereas “immune-/replication+” had the worst (log-rank p = 6 × 10⁻⁴, HR = 5.45). By linking imaging heterogeneity directly to molecular subtypes and therapeutic response, this framework provides a robust, non-invasive surrogate for genomic profiling and a strategic tool for personalized neoadjuvant therapy selection.

Objective This study aims to develop a functional-based multi-omics model for early prediction of radiation pneumonitis (RP) by extracting radiomic and dosiomic features from functionally defined lung regions, using generated perfusion (Q) and ventilation (V) from pre-radiotherapy planning computed tomography (CT). Approach We retrospectively analyzed data from 121 patients with locally advanced non-small cell lung cancer (NSCLC) treated with curative-intent IMRT between 2015 and 2019, including pre-treatment CT and dose maps. Q and V maps were generated from CT with deep learning-based and supervoxel-based approaches, respectively. Regions of interest (ROIs) combined the planning target volume (PTV) with each of three functional lung regions—high functional lung (HFL), low functional lung (LFL), and whole lung (WL)—defined by thresholds on Q and V maps. Radiomic and dosiomic features were extracted from CT and dose distributions within each ROI. For each ROI, For each ROI, three methods—radiomics (R), dosiomics (D), and dual-omics (RD)—were constructed. 13 machine learning algorithms were trained and evaluated using 10-fold cross-validation, and model performance was assessed by the average area under the receiver operating characteristic curve (AUC), accuracy, precision, recall, and F1 score. RP was defined as CTCAE grade ≥ 2. Main results Of the 35 selected features, 20 were from HFL. In dual-omics models, using HFL features improved predictive performance for RP (AUC 0.879±0.105) compared to WL (AUC 0.778 ± 0.100). In HFL, the RD method outperformed both R (AUC 0.786± 0.076) and D (AUC 0.791 ± 0.107) methods. Decision curve analysis showed the dual-omics model based on HFL provided the highest net benefit across threshold probabilities. Significance This study is the first to systematically demonstrate that features extracted from CT-derived HFL capture important functional differences and provide strong predictive value for RP. Compared to conventional methods, integrating radiomics, dosiomics, and CT-based functional information further improves predictive performance.