College of Software, Xinjiang University, Urumqi, Xinjiang, People's Republic of China.
Key Laboratory of Software Engineering Technology, College of Software, Xin Jiang University, Urumqi, People's Republic of China.
Phys Med Biol. 2024 Mar 18;69(7). doi: 10.1088/1361-6560/ad25c1.
. Celiac disease (CD) has emerged as a significant global public health concern, exhibiting an estimated worldwide prevalence of approximately 1%. However, existing research pertaining to domestic occurrences of CD is confined mainly to case reports and limited case analyses. Furthermore, there is a substantial population of undiagnosed patients in the Xinjiang region. This study endeavors to create a novel, high-performance, lightweight deep learning model utilizing endoscopic images from CD patients in Xinjiang as a dataset, with the intention of enhancing the accuracy of CD diagnosis.. In this study, we propose a novel CNN-Transformer hybrid architecture for deep learning, tailored to the diagnosis of CD using endoscopic images. Within this architecture, a multi-scale spatial adaptive selective kernel convolution feature attention module demonstrates remarkable efficacy in diagnosing CD. Within this module, we dynamically capture salient features within the local channel feature map that correspond to distinct manifestations of endoscopic image lesions in the CD-affected areas such as the duodenal bulb, duodenal descending segment, and terminal ileum. This process serves to extract and fortify the spatial information specific to different lesions. This strategic approach facilitates not only the extraction of diverse lesion characteristics but also the attentive consideration of their spatial distribution. Additionally, we integrate the global representation of the feature map obtained from the Transformer with the locally extracted information via convolutional layers. This integration achieves a harmonious synergy that optimizes the diagnostic prowess of the model.. Overall, the accuracy, specificity, F1-Score, and precision in the experimental results were 98.38%, 99.04%, 98.66% and 99.38%, respectively.. This study introduces a deep learning network equipped with both global feature response and local feature extraction capabilities. This innovative architecture holds significant promise for the accurate diagnosis of CD by leveraging endoscopic images captured from diverse anatomical sites.
. 乳糜泻（CD）已成为一个重大的全球公共卫生问题，估计全球患病率约为 1%。然而，国内有关 CD 的研究主要局限于病例报告和有限的病例分析。此外，新疆地区还有大量未确诊的患者。本研究旨在利用新疆 CD 患者的内镜图像作为数据集，创建一个新颖的、高性能的、轻量级的深度学习模型，以提高 CD 诊断的准确性。. 在这项研究中，我们提出了一种新的 CNN-Transformer 混合架构，用于使用内镜图像诊断 CD。在这个架构中，多尺度空间自适应选择核卷积特征注意力模块在诊断 CD 方面表现出了显著的效果。在这个模块中，我们动态地捕获局部通道特征图中的显著特征，这些特征对应于 CD 受累区域内镜图像病变的不同表现，如十二指肠球部、十二指肠降段和末端回肠。这个过程有助于提取和强化不同病变的空间信息。这种策略不仅有助于提取不同病变的特征，而且还可以注意到它们的空间分布。此外，我们通过卷积层将来自 Transformer 的特征图的全局表示与局部提取的信息进行集成。这种集成实现了一个和谐的协同作用，优化了模型的诊断能力。. 总的来说，实验结果的准确率、特异性、F1-Score 和精度分别为 98.38%、99.04%、98.66%和 99.38%。. 本研究引入了一种具有全局特征响应和局部特征提取能力的深度学习网络。这种创新的架构有望通过利用来自不同解剖部位的内镜图像来准确诊断 CD。
