Purpose: In recent years, the continuous advancement of convolutional neural networks (CNNs) has led to the widespread integration of deep neural networks as a mainstream approach in clinical diagnostic support. Particularly, the utilization of CNN-based medical image segmentation has delivered favorable outcomes for aiding clinical diagnosis. Within this realm, network architectures based on the U-shaped structure and incorporating skip connections, along with their diverse derivatives, have gained extensive utilization across various medical image segmentation tasks. Nonetheless, two primary challenges persist. First, certain organs or tissues present considerable complexity, substantial morphological variations, and size discrepancies, posing significant challenges for achieving highly accurate segmentation. Second, the predominant focus of current deep neural networks on single-resolution feature extraction limits the effective extraction of feature information from complex medical images, thereby contributing to information loss via continuous pooling operations and contextual information interaction constraints within the U-shaped structure.

Approach: We proposed a five-layer pyramid segmentation network (PS5-Net), a multiscale segmentation network with diverse resolutions that is founded on the U-Net architecture. Initially, this network effectively leverages the distinct features of images at varying resolutions across different dimensions, departing from prior single-resolution feature extraction methods to adapt to intricate and variable segmentation scenarios. Subsequently, to comprehensively integrate feature information from diverse resolutions, a kernel selection module is proposed to assign weights to features across different dimensions, enhancing the fusion of feature information from various resolutions. Within the feature extraction network denoted as PS-UNet, we preserve the classical structure of the traditional U-Net while enhancing it through the incorporation of dilated convolutions.

Results: PS5-Net attains a Dice score of 0.9613 for liver segmentation on the CHLISC dataset and 0.8587 on the ISIC2018 dataset for skin lesion segmentation. Comparative analysis with diverse medical image segmentation methodologies in recent years reveals that PS5-Net has achieved the highest scores and substantial advancements.

Conclusions: PS5-Net effectively harnesses the rich semantic information available at different resolutions, facilitating a comprehensive and nuanced understanding of the input medical images. By capitalizing on global contextual connections, the network adeptly captures the intricate interplay of features and dependencies across the entire image, resulting in more accurate and robust segmentation outcomes. The experimental validation of PS5-Net underscores its superior performance in medical image segmentation tasks, offering promising prospects for enhancing diagnostic and analytical processes within clinical settings. These results highlight the potential of PS5-Net to significantly contribute to the advancement of medical imaging technologies and ultimately improve patient care through more precise and reliable image analysis.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC10876014PMC
http://dx.doi.org/10.1117/1.JMI.11.1.014008DOI Listing

Publication Analysis

Top Keywords

medical image
20
image segmentation
20
segmentation
12
segmentation network
12
neural networks
12
feature extraction
12
deep neural
8
segmentation tasks
8
single-resolution feature
8
medical images
8

Similar Publications

Atomic force microscopy (AFM) has reached a significant level of maturity in biology, demonstrated by the diversity of modes for obtaining not only topographical images but also insightful mechanical and adhesion data by performing force measurements on delicate samples with a controlled environment (e.g., liquid, temperature, pH).

View Article and Find Full Text PDF

Myocardial extracellular volume fraction (ECV) measured via MRI serves as a quantitative indicator of myocardial fibrosis. However, accurate measurement of ECV using MRI in the presence of AF is challenging. Meanwhile, CT could be a promising alternative tool for measuring ECV regardless of sinus rhythm or AF.

View Article and Find Full Text PDF

Retraction Note: Letter to editor: Comment on, "The role of secondary imaging in children aged under 24 months with proven skull fracture on initial computed tomography scan".

Neurosurg Rev

January 2025

Department of Cariology, Saveetha Dental College and Hospitals, Saveetha Institute of Medical and Technical Sciences (SIMATS), Saveetha University, Chennai, 600077, India.

View Article and Find Full Text PDF

Purpose Of Review: What is the pathophysiology and clinical findings as well as management of patients presenting with INOCA/MINOCA (Ischemia/Myocardial Infarction with Non-Obstructive Coronary Arteries).

Recent Findings: INOCA/MINOCA has a complex pathophysiology. In this review article, we aim to summarize the complex pathophysiology and clinical diagnosis, and review the current management options.

View Article and Find Full Text PDF

Applications of mass spectrometry imaging in botanical research.

Adv Biotechnol (Singap)

February 2024

State Key Laboratory of Traditional Chinese Medicine/School of Pharmaceutical Sciences, Guangzhou University of Chinese Medicine, Guangzhou, 510006, China.

Mass spectrometry imaging (MSI) serves as a valuable tool enabling researchers to scrutinize various compounds, peptides, and proteins within a sample, providing detailed insights at both elemental and molecular levels. This innovative technology transforms information obtained from a mass spectrometer- encompassing ionic strength, mass-to-charge ratio, and ionized molecule coordinates-within a defined region into a pixel-based model. Consequently, it reconstructs the spatial distribution of ions, allowing for a comprehensive understanding of molecular landscapes.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!