Home > Journals > JOEUC > Article
A Generalized Deep Learning Framework for Robust Medical Image Segmentation
Open Access Journal

A Generalized Deep Learning Framework for Robust Medical Image Segmentation

Chuan-Wang Chang (National Chin-Yi University of Technology, Taiwan), Shi-Hong Qiu (National Chung Hsing University, Taiwan), Tien-Yi Liao (National Chin-Yi University of Technology, Taiwan), and Cheng-Mu Tsai (National Chung Hsing University, Taiwan)
Copyright: © 2026 | Pages: 31
DOI: 10.4018/JOEUC.413059

Abstract

Semantic segmentation is crucial for medical image analysis, enabling accurate delineation of anatomical structures and pathological regions. However, heterogeneity across imaging modalities such as ultrasound, fundus photography, X-ray, and computed tomography (CT) challenges model generalization. This study proposes GenMed-Net, a generalized segmentation framework based on a U-Net backbone integrating a ResNet50 encoder, a Spatial Channel Block Attention Module (SCBAM), and an Atrous Spatial Pyramid Pooling (ASPP) module. GenMed-Net is evaluated on four heterogeneous datasets: thyroid ultrasound, retinal fundus vessel images, pulmonary fibrosis chest X-rays, and the CC-CCII pneumonia chest CT dataset. The model achieves Dice Similarity Coefficients of 91.87%, 96.15%, 98.99%, and 89.11%, respectively, outperforming representative CNN- and Transformer-based methods. Visual and attention heatmap analyses further demonstrate improved lesion localization and strong cross-modality generalization, supporting its potential for clinical decision support.
Article Preview
Top

Introduction

With the rapid advancement of medical diagnostic technologies, particularly the breakthroughs in artificial intelligence (AI) and computer vision, the capacity of machines to process and interpret medical images has reached an unprecedented level of precision and efficiency. From early imaging modalities such as X-ray, computed tomography (CT), and magnetic resonance imaging (MRI) to ultrasound, medical images have become indispensable tools for clinical diagnosis, disease localization, treatment planning, and postoperative monitoring. These imaging technologies allow clinicians to observe both anatomical structures and pathological variations in a non-invasive manner, thereby enabling early disease detection and more accurate evaluation of treatment outcomes.

However, as imaging devices have gained sophistication, the volume, dimensionality, and heterogeneity of medical data have also expanded dramatically. Manual interpretation by radiologists is often time-consuming, subjective, and prone to inter-observer variability. The demand for high-throughput and objective analysis has thus accelerated the integration of AI-driven solutions in healthcare. In particular, computer-aided diagnosis (CAD) systems have emerged as essential tools that leverage image analysis and machine learning algorithms to assist radiologists in identifying abnormal patterns and quantifying pathological regions (Yanase et al., 2019; Yeasmin et al., 2024). CAD systems not only enhance diagnostic efficiency and reduce misdiagnosis rates but also alleviate the cognitive burden on physicians by providing consistent, data-driven insights.

Despite these advances, several critical challenges remain in CAD-based medical image analysis. First, acquiring large, high-quality annotated datasets is costly and time-intensive, as expert-level annotation requires significant medical expertise. Second, disease manifestations often vary greatly across imaging modalities, patient populations, and acquisition conditions. For instance, the appearance of lesions in CT differs substantially from that in ultrasound or MRI. Such variability leads to reduced generalizability when a model trained on one dataset is applied to another. Therefore, developing robust and generalizable segmentation frameworks capable of adapting to different imaging modalities is vital for clinical reliability and real-world deployment.

Within this context, semantic segmentation plays a fundamental role in the understanding of medical images. As one of the most crucial tasks in image analysis, semantic segmentation aims to assign a class label to every pixel in an image, thereby providing a detailed map of anatomical structures and pathological regions (Liu et al., 2019; Wu, 2017).This fine-grained localization is indispensable for quantitative medical assessment, such as tumor volume measurement, organ delineation, and lesion progression tracking. Traditional rule-based or handcrafted-feature methods, however, often fail to handle the complex textures, noise, and variability present in real-world medical data. In contrast, deep learning, especially convolutional neural network (CNN) technology, has shown remarkable success in extracting hierarchical and discriminative representations, significantly improving segmentation performance across numerous medical applications (Wang et al., 2022).

Nevertheless, a single segmentation model often struggles to achieve cross-modality generalization. The significant differences in image contrast, spatial resolution, and tissue morphology between modalities, such as CT, MRI, X-ray, and ultrasound, pose severe challenges for model robustness. A model optimized for one imaging domain may not perform effectively in another owing to differences in visual patterns and underlying noise distributions (Szegedy et al., 2016). Thus, constructing a unified, adaptive segmentation framework that maintains strong performance across multiple modalities has become a crucial step toward achieving generalized medical image understanding.

Complete Article List

Search this Journal:
Reset
Volume 38: 1 Issue (2026)
Volume 37: 1 Issue (2025)
Volume 36: 1 Issue (2024)
Volume 35: 3 Issues (2023)
Volume 34: 10 Issues (2022)
Volume 33: 6 Issues (2021)
Volume 32: 4 Issues (2020)
Volume 31: 4 Issues (2019)
Volume 30: 4 Issues (2018)
Volume 29: 4 Issues (2017)
Volume 28: 4 Issues (2016)
Volume 27: 4 Issues (2015)
Volume 26: 4 Issues (2014)
Volume 25: 4 Issues (2013)
Volume 24: 4 Issues (2012)
Volume 23: 4 Issues (2011)
Volume 22: 4 Issues (2010)
Volume 21: 4 Issues (2009)
Volume 20: 4 Issues (2008)
Volume 19: 4 Issues (2007)
Volume 18: 4 Issues (2006)
Volume 17: 4 Issues (2005)
Volume 16: 4 Issues (2004)
Volume 15: 4 Issues (2003)
Volume 14: 4 Issues (2002)
Volume 13: 4 Issues (2001)
Volume 12: 4 Issues (2000)
Volume 11: 4 Issues (1999)
Volume 10: 4 Issues (1998)
Volume 9: 4 Issues (1997)
Volume 8: 4 Issues (1996)
Volume 7: 4 Issues (1995)
Volume 6: 4 Issues (1994)
Volume 5: 4 Issues (1993)
Volume 4: 4 Issues (1992)
Volume 3: 4 Issues (1991)
Volume 2: 4 Issues (1990)
Volume 1: 3 Issues (1989)
View Complete Journal Contents Listing