Aim New statistical method was applied in data analysis of orthogonal experiments to optimize the preparation of liposome. Method Particle size, zeta potential, encapsulation efficiency and physical stability of liposomes were selected by orthogonal design as evaluating indicators. Through three statistical methods （direct observation, variance analysis and stepwise multiple regression）, the optimized preparing conditions were acquired and validated by experiment. Results All of the four indicators were different by these analyses. The validation experiments indicated that the optimized conditions by stepwise multiple regressions were better than that by traditional analysis. Conclusion Experiment results suggested that multiple regressions could avoid the weakness of direct observation and variance analysis, but more work should be done in preparing liposomes.
A new statistical method, the fuzzy analytical method, was introduced in the optimization processes of liposome preparation. It took the full advantage of the information from orthogonal experiments to obtain the optimal liposome preparation conditions by considering all the evaluation indexes. Liposomes were made by the modified reverse-phase evaporation method and the properties of liposomes including size, encapsulation efficiency and physical stability were selected as the evaluation indexes to indicate the quality of liposomes. The experimental data of these properties were analyzed by three different methods including direct observation, variance analysis and fuzzy analytical method. The optimal preparation conditions obtained from these methods were validated with further experiments. The results of all possible combinations of levels for all factors in orthogonal experiments were acquired by the fuzzy analytical method. All evaluation indexes were taken into account and the optimal preparation condition was obtained. The optimal preparation conditions from direct observation and fuzzy analytical method were different and further validation studies indicated that the optimal conditions obtained from the fuzzy analytical method were in agreement with that from traditional statistical analysis. Fuzzy analytical method avoided the problem resulted from the traditional method, in which different levels of the same factor were obtained when considering different evaluation indexes. More information could be obtained from the fuzzy analytical method and the blind area within the experimental range was eliminated. As a result, fuzzy analytical method can be applied in the optimization processes of liposome preparation.
