專題演講 主講人：游鎮瑋教授 (輔仁大學統計資訊學系)
題 目：Innovative Integrative Prognostic Modeling in Cancer Precision Medicine: ARIMA-CNN for Temporal Dynamics and Genomic Immune Signatures
主講人：游鎮瑋教授 (輔仁大學統計資訊學系)
時 間：113年12月20日(星期五)上午10:10-11:00
(上午09:50-10:10茶會於綜合一館428室舉行)
地 點：綜合一館427室
使用Google Meet線上直播，
演講開始前20分鐘可進入會議，請點選下列連結後按下「加入」即可https://meet.google.com/pie-jmyd-cra
摘要
Hepatocellular carcinoma (HCC) poses a significant global health challenge due to its high incidence and mortality rates. Our study investigates the prognostic significance of chemokine (C-C motif) ligand 5 (CCL5) and various immune gene signatures in HCC using an innovative combination of Autoregressive Integrated Moving Average (ARIMA) and Convolutional Neural Network (CNN) models. Time series data from The Cancer Genome Atlas (TCGA) were utilized to apply an ARIMA model that captures the temporal dynamics of CCL5 expression. Residuals from this model were integrated with immune signature expression data, including lymphocytes and macrophages, to extract features using a CNN model. Our results demonstrate that CNN-extracted features provide a more statistically significant association with patient survival compared to traditional median split methods. Specifically, CNN-extracted features for CD8 T cells and effective T cells showed a hazard ratio (HR) of 0.7324 (p = 0.0008), highlighting their critical role in the immune response against tumors. Furthermore, clusters of immune genes based on non-parametric correlations revealed distinct survival patterns. The cluster, including B cells, Th2 cells, T cells, and NK cells, showed a moderate protective effect (HR: 0.8714, p = 0.1093) with a significant log-rank p-value (0.0233). Conversely, the cluster with granulocytes, Tregs, macrophages, and MDSCs did not show significant associations with survival, underscoring the complexity of immune regulation within the tumor immune microenvironment. These findings highlight the importance of considering both temporal dynamics and synergistic interactions among immune genes for accurate prognostic evaluation. Our study's integrated approach offers a novel framework for identifying potential biomarkers and developing personalized immunotherapy strategies, potentially revolutionizing HCC management. The innovative use of ARIMA-CNN models to analyze the dynamic impact of multiple immune genes represents a significant advancement over traditional methods, providing a more comprehensive understanding of their prognostic value. This hybrid approach, which captures both linear and nonlinear relationships within the data, holds promise for future research and clinical applications, offering new avenues for effectively managing and treating HCC.
