Natural medicines (NMs) are crucial for treating human diseases. Efficiently characterizing their bioactive components in vivo has been a key focus and challenge in NM research. High-performance liquid chromatography-high-resolution mass spectrometry (HPLC-HRMS) systems offer high sensitivity, resolution, and precision for conducting in vivo analysis of NMs. However, due to the complexity of NMs, conventional data acquisition, mining, and processing techniques often fail to meet the practical needs of in vivo NM analysis. Over the past two decades, intelligent spectral data-processing techniques based on various principles and algorithms have been developed and applied for in vivo NM analysis. Consequently, improvements have been achieved in the overall analytical performance by relying on these techniques without the need to change the instrument hardware. These improvements include enhanced instrument analysis sensitivity, expanded compound analysis coverage, intelligent identification, and characterization of nontargeted in vivo compounds, providing powerful technical means for studying the in vivo metabolism of NMs and screening for pharmacologically active components. This review summarizes the research progress on in vivo analysis strategies for NMs using intelligent MS data processing techniques reported over the past two decades. It discusses differences in compound structures, variations among biological samples, and the application of artificial intelligence (AI) neural network algorithms. Additionally, the review offers insights into the potential of in vivo tracking of NMs, including the screening of bioactive components and the identification of pharmacokinetic markers. The aim is to provide a reference for the integration and development of new technologies and strategies for future in vivo analysis of NMs.