文献类型：期刊文献

英文题名：MI2A: A Multimodal Information Interaction Architecture for Automated Diagnosis of Lung Nodules Using PET/CT Imaging

作者：Li, Kai[1]; Li, Tongtong[2]; Zhang, Lei[3]; Mao, Junfeng[4,5]; Shi, Xuerong[2]; Yao, Zhijun[2]; Fang, Lei[6]; Hu, Bin[2,7,8,9]

第一作者：Li, Kai

机构：[1] Lanzhou University, School of Mathematics and Statistics, Lanzhou, 730000, China; [2] Lanzhou University, Gansu Provincial Key Laboratory of Wearable Computing, School of Information Science and Engineering, Lanzhou, 730000, China; [3] Hexi University, College of Information Technology and Communication, Zhangye, 734000, China; [4] 940th Hospital of Joint Logistics Support Force of Chinese People’s Liberation Army, Department of Nuclear Medicine, Lanzhou, 730050, China; [5] Gansu University of Chinese Medicine, School of Basic Medical Sciences, Lanzhou, 730000, China; [6] Taikang Tongji [Wuhan] Hospital, Department of Nuclear Medicine, Wuhan, 430050, China; [7] Beijing Institute of Technology, School of Medical Technology, Beijing, 100081, China; [8] Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, CAS Center for Excellence in Brain Science and Intelligence Technology, Shanghai, 200031, China; [9] Chinese Academy of Sciences, Joint Research Center for Cognitive Neurosensor Technology of Lanzhou University, Institute of Semiconductors, Lanzhou, 730000, China

第一机构：Lanzhou University, School of Mathematics and Statistics, Lanzhou, 730000, China

年份：2025

外文期刊名：IEEE Sensors Journal

收录：EI(收录号：20252618692062);Scopus(收录号：2-s2.0-105009299019)

语种：英文

外文关键词：Biological organs - Computer aided diagnosis - Computerized tomography - Diseases - Lung cancer - Metabolism - Positron emission tomography

摘要：Lung cancer is one of the most common malignancies globally, with malignant nodules being an early indicator of the disease. Thus, accurate early diagnosis of lung nodules is imperative. Positron Emission Tomography-Computed Tomography (PET/CT) is a non-invasive imaging technique that provides both anatomical and metabolic information, playing a crucial role in the diagnosis of cancer. Existing deep learning-based multimodal fusion strategies often rely on the simple concatenation of features from two modalities, overlooking the intricate interactions between them. In this study, we proposed a multimodal information interaction framework named MI2A for the automated diagnosis of lung nodules using PET/CT imaging. Specifically, the lung parenchymal regions were cropped as regions of interest using a pre-trained U-Net model. Secondly, higher-order multimodal features from PET/CT scans were extracted and integrated using a custom-designed PET-CT Imaging Encoder (PCIE) module and a Cross-Attention Multimodal Encoder (CAME) module, respectively. Predictions were generated using multi-path pooling layers and a multi-layer perceptron (MLP) layer. Furthermore, an alignment loss function was designed to minimize the discrepancy between modality features during training. Finally, the proposed model was evaluated on an actual clinical dataset, achieving accuracy, precision, recall, specificity, and F1 scores of 0.9179, 0.8972, 0.8937, 0.9335, and 0.8954, respectively. In addition, the findings revealed that certain benign lesions, particularly those related to inflammatory or infectious conditions, displayed high metabolic activity, which is the main reason for limiting the model’s performance. This insight provides a promising direction for future research. ? 2001-2012 IEEE.