This study introduces a hyperspectral PRM-SRS microscopy platform that integrates a Penalized Reference Matching (PRM) algorithm with Stimulated Raman Scattering (SRS) microscopy to enable high-resolution, label-free visualization and identification of lipid subtypes in various biological samples. The PRM-SRS method enhances chemical specificity, subcellular resolution, and fast data processing, allowing for the differentiation of lipid subtypes in different organs and species. The platform was used to visualize high-density lipoprotein particles in human kidney tissue, a high cholesterol to phosphatidylethanolamine ratio in mouse hippocampal granule cells, and subcellular distributions of sphingosine and cardiolipin in human brain tissue. Current lipidomic technologies, such as shotgun lipidomics, can quickly identify hundreds of lipids from small samples but are destructive and rely on mass spectrometry or nuclear magnetic resonance. Conventional matrix-assisted laser desorption/ionization (MALDI)-MS imaging enables label-free lipid imaging but has limited resolution and can destroy the sample. Other optical techniques rely on fluorescent markers, which may alter lipid distribution. Label-free optical imaging is essential for accurate lipid subtype differentiation. SRS microscopy has demonstrated non-destructive 3D imaging with subcellular resolution. Recent work has even demonstrated quantitative mass concentration measurements of lipids, proteins, and water. For label-free SRS imaging, multiple subcellular organelles and their chemical compositions can be visualized and mapped out through hyperspectral imaging or training of a deep learning model. The PRM-SRS method was developed to enhance the specificity for distinguishing lipid species accurately. It was applied to analyze different lipid subtypes in various organs and species, with a library of 38 biomolecules for potential detection. The method is efficient and can process a 512 pixels × 512 pixels × 76 hyperspectral image stack within one minute. Future studies will focus on improving detection sensitivity to enhance the signal-to-noise ratio for examining molecules of low abundance. This method provides quantitative and qualitative insights into the roles of lipid species in multiple biological processes and can augment other unmixing techniques. The PRM-SRS method was validated using Drosophila fat body tissues to detect and compare cholesterol levels in young and old flies. The results showed significantly higher similarity scores to the cholesterol reference spectra in old flies, indicating elevated cholesterol content. The method was also used to detect cardiolipin changes in cells, showing significant decreases of CL signals in shPGS1 cells compared with control cells. The PRM-SRS method was further applied to human kidney tissue to characterize lipid subtypes, revealing distributions of distinct lipid subtypes in the glomerulus and surrounding structures. The method was also used to analyze lipid subtypes in mouse brain samples, showing changes in the Cholesterol/PE ratio in subregions of granule cell nuclei in old brains. The PRM-SThis study introduces a hyperspectral PRM-SRS microscopy platform that integrates a Penalized Reference Matching (PRM) algorithm with Stimulated Raman Scattering (SRS) microscopy to enable high-resolution, label-free visualization and identification of lipid subtypes in various biological samples. The PRM-SRS method enhances chemical specificity, subcellular resolution, and fast data processing, allowing for the differentiation of lipid subtypes in different organs and species. The platform was used to visualize high-density lipoprotein particles in human kidney tissue, a high cholesterol to phosphatidylethanolamine ratio in mouse hippocampal granule cells, and subcellular distributions of sphingosine and cardiolipin in human brain tissue. Current lipidomic technologies, such as shotgun lipidomics, can quickly identify hundreds of lipids from small samples but are destructive and rely on mass spectrometry or nuclear magnetic resonance. Conventional matrix-assisted laser desorption/ionization (MALDI)-MS imaging enables label-free lipid imaging but has limited resolution and can destroy the sample. Other optical techniques rely on fluorescent markers, which may alter lipid distribution. Label-free optical imaging is essential for accurate lipid subtype differentiation. SRS microscopy has demonstrated non-destructive 3D imaging with subcellular resolution. Recent work has even demonstrated quantitative mass concentration measurements of lipids, proteins, and water. For label-free SRS imaging, multiple subcellular organelles and their chemical compositions can be visualized and mapped out through hyperspectral imaging or training of a deep learning model. The PRM-SRS method was developed to enhance the specificity for distinguishing lipid species accurately. It was applied to analyze different lipid subtypes in various organs and species, with a library of 38 biomolecules for potential detection. The method is efficient and can process a 512 pixels × 512 pixels × 76 hyperspectral image stack within one minute. Future studies will focus on improving detection sensitivity to enhance the signal-to-noise ratio for examining molecules of low abundance. This method provides quantitative and qualitative insights into the roles of lipid species in multiple biological processes and can augment other unmixing techniques. The PRM-SRS method was validated using Drosophila fat body tissues to detect and compare cholesterol levels in young and old flies. The results showed significantly higher similarity scores to the cholesterol reference spectra in old flies, indicating elevated cholesterol content. The method was also used to detect cardiolipin changes in cells, showing significant decreases of CL signals in shPGS1 cells compared with control cells. The PRM-SRS method was further applied to human kidney tissue to characterize lipid subtypes, revealing distributions of distinct lipid subtypes in the glomerulus and surrounding structures. The method was also used to analyze lipid subtypes in mouse brain samples, showing changes in the Cholesterol/PE ratio in subregions of granule cell nuclei in old brains. The PRM-S