International Journal of Molecular Medical Science, 2025, Vol.15, No.5, 205-213 http://medscipublisher.com/index.php/ijmms 208 identification. The platform capable of measuring multiple indicators can simultaneously detect various markers such as Aβ42, t-tau, p-tau, NfL, and GFAP, thereby more comprehensively judging the pathological state of AD (Kim et al., 2020). Studies have found that the combination of blood markers with genetic information, demographic data, or imaging data can lead to better classification results, which is particularly important in the early stage of disease or preclinical stage (Ghazi et al., 2024). For instance, multimodal blood tests designed based on gene expression, proteins and clinical features, combined with machine learning methods, have been proven to have a high diagnostic accuracy, capable of identifying susceptible populations before symptoms appear (Gunes et al., 2022; AlMansoori et al., 2024). This combined approach helps address the heterogeneity of AD and also provides assistance for more personalized and precise diagnosis (Bhalala et al., 2024; Dhauria et al., 2024; Zu et al., 2024). Figure 1 Main stages and features of Alzheimer’s disease (Adopted from Siedlecki-Wullich et al., 2021) Image caption: Pathological changes and cognitive symptoms are represented as blue and brown lines, respectively. Pathological hallmarks currently used as biomarkers (Aβ and tau) are shown in blue rectangles, while key global pathological changes are indicated with arrows. Cognitive symptoms are summarized as MCI (mild cognitive impairment) and dementia stages. miRNA-based signatures for potential diagnosis of MCI and AD stages are indicated as references (Adopted from Siedlecki-Wullich et al., 2021) 4.3 Multi-mode integration facilitates early diagnosis and dynamic monitoring Combining biomarkers from different sources, such as those related to body fluids, imaging and genetics, can help achieve early detection and long-term follow-up of the disease course of Alzheimer's disease. The multimodal analysis framework integrates advanced neuroimaging techniques, blood and cerebrospinal fluid markers, and computational models to clarify the pathophysiological mechanism of AD more accurately, thereby promoting disease staging, prognosis determination, and follow-up after treatment (Jamal et al., 2025). Incorporating machine learning and artificial intelligence technologies can further enhance the predictive ability of this integrated model, supporting the judgment of risks based on individual circumstances and real-time disease monitoring (Gunes et al., 2022; Lou and Xu, 2025). Nowadays, in both scientific research and clinical
RkJQdWJsaXNoZXIy MjQ4ODYzNA==