Aim In order to evaluate the instability of atherosclerotic plaque more comprehensively and objectively, the computational method of plaque vulnerability index based on lipid accumulation, which was put forward by Shiomi was improved in this study. Methods The experimental was divided into the control group (C57BL/6 mice) and the model group (LDLR－/－ mice and ApoE－/－ mice), with 6 mice in each group. The control group was given normal diet, while the model group was given high-fat diet. After 25 weeks, the mice were sacrificed and the aortic root and aortic arch were isolated and sectioned after OCT embedding. Serial sections were stained by hematoxylin and eosin (H&E), picrosirius red, oil red O, alcian blue, and quantitatively studied the amount of endothelial cell and smooth muscle cell by immunofluorescence, as well as matrix metalloproteinase and macrophage with immunohistochemistry. The calculation formula of plaque vulnerability index proposed by Shiomi was perfected through adding the factors affecting the plaque instability to the formula numerator and the stability index of the plaque to the denominator. The stability of aorta root and innominate artery plaque in LDLR－/－ mice was compared by two formulas, and the reliability of the optimized formula was evaluated. Finally, the stability of atherosclerotic plaques in different parts of aorta of ApoE－/－ and LDLR－/－ mice was evaluated by the optimized formula, and the difference of atherosclerosis progression between two genetically engineered mice was compared. Results The plaque damage area, lipid deposition, the distribution of collagen, proteoglycan and matrix metalloproteinase, macrophage infiltration, the endothelial cell coverage and smooth muscle cell coverage of fibrous cap in the model group were significantly changed compared with the control group. On this basis, the formula for calculating plaque instability based on multi index pathological staining was optimized. Through calculation, it was found that Plaque vulnerability index of innominate artery was significantly larger than that of aortic root in ApoE－/－ and LDLR－/－ mice. In the same site, plaque vulnerability index in ApoE－/－ mice was significantly higher than that in LDLR－/－ mice. Conclusion Optimized calculation formula of plaque vulnerability index based on multi index pathological staining covers a number of important factors that affect plaque instability, and can accurately assess plaque instability. The method is simple, comprehensive, reliable and of great practical value.