BMNet: A New Region-Based Metric Learning Method for Early Alzheimer's Disease Identification With FDG-PET Images

Wenju Cui; Caiying Yan; Zhuangzhi Yan; Yunsong Peng; Yilin Leng; Chenlu Liu; Shuangqing Chen; Xi Jiang; Jian Zheng; Xiaodong Yang

doi:10.3389/fnins.2022.831533

BMNet: A New Region-Based Metric Learning Method for Early Alzheimer's Disease Identification With FDG-PET Images

Front Neurosci. 2022 Feb 24:16:831533. doi: 10.3389/fnins.2022.831533. eCollection 2022.

Authors

Wenju Cui^{1

2}, Caiying Yan³, Zhuangzhi Yan¹, Yunsong Peng^{2

4}, Yilin Leng^{1

2}, Chenlu Liu³, Shuangqing Chen³, Xi Jiang⁵, Jian Zheng², Xiaodong Yang²

Affiliations

¹ Institute of Biomedical Engineering, School of Communication and Information Engineering, Shanghai University, Shanghai, China.
² Medical Imaging Department, Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou, China.
³ Department of Radiology, The Affiliated Suzhou Hospital of Nanjing Medical University, Suzhou, China.
⁴ School of Biomedical Engineering, Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, China.
⁵ School of Life Sciences and Technology, The University of Electronic Science and Technology of China, Chengdu, China.

Abstract

18F-fluorodeoxyglucose (FDG)-positron emission tomography (PET) reveals altered brain metabolism in individuals with mild cognitive impairment (MCI) and Alzheimer's disease (AD). Some biomarkers derived from FDG-PET by computer-aided-diagnosis (CAD) technologies have been proved that they can accurately diagnosis normal control (NC), MCI, and AD. However, existing FDG-PET-based researches are still insufficient for the identification of early MCI (EMCI) and late MCI (LMCI). Compared with methods based other modalities, current methods with FDG-PET are also inadequate in using the inter-region-based features for the diagnosis of early AD. Moreover, considering the variability in different individuals, some hard samples which are very similar with both two classes limit the classification performance. To tackle these problems, in this paper, we propose a novel bilinear pooling and metric learning network (BMNet), which can extract the inter-region representation features and distinguish hard samples by constructing the embedding space. To validate the proposed method, we collect 898 FDG-PET images from Alzheimer's disease neuroimaging initiative (ADNI) including 263 normal control (NC) patients, 290 EMCI patients, 147 LMCI patients, and 198 AD patients. Following the common preprocessing steps, 90 features are extracted from each FDG-PET image according to the automatic anatomical landmark (AAL) template and then sent into the proposed network. Extensive fivefold cross-validation experiments are performed for multiple two-class classifications. Experiments show that most metrics are improved after adding the bilinear pooling module and metric losses to the Baseline model respectively. Specifically, in the classification task between EMCI and LMCI, the specificity improves 6.38% after adding the triple metric loss, and the negative predictive value (NPV) improves 3.45% after using the bilinear pooling module. In addition, the accuracy of classification between EMCI and LMCI achieves 79.64% using imbalanced FDG-PET images, which illustrates that the proposed method yields a state-of-the-art result of the classification accuracy between EMCI and LMCI based on PET images.

Keywords: FDG-PET images; bilinear pooling; early Alzheimer’s disease; embedding space; inter-region representation; metric learning; mild cognitive impairment.