Purpose: To assess the feasibility of applying diffusion kurtosis imaging(DKI)for differentiating head and neck tumor benignity and malignancy and for tumor heterogeneity discrimination Methods:A total of 34 subjects(16 with benign and 18 with malignant tumors)were included in the study.Conventional MRI,DWI and DKI were performed prior to surgery and pathologic verification.The diffusion coefficient(D)and excessive kurtosis(K)were obtained from diffusion-weighted images at b-values of 0,500,1000,and 1500 s/mm2 on 3T magnetic resonance imaging(MRI).The combination of D and K values showing the maximum density of the probable density function was estimated.The ADC was obtained(0 and 1000 s/mm2).Malignant tumors were classified into three subgroups(epithelial malignant tumor,soft tissue malignant tumors and lymphoid and haematopoietic malignant tumors)according to tumor histologic origin.Multivariate logistic regression modeling was performed to assess the combination of D and K model versus ADC for differentiating tumor benignity from malignancy.Fisher stepwise discriminant analysis was performed to distinguish tumor heterogeneity.Results: The Mean D,K and ADC were significantly different between benign and malignant tumors(P=0.0011,0.0237 and 0.0004,respectively).Using above three parameters for differentiation tumor benignity and malignancy had the sensitivity/specificity of 88.89%/81.25%,83.33%/68.75%,77.78%/75%.Furthermore,MD,MK and mean ADC all yielded high AUC for differentiating tumor benignity from malignancy(0.792,0.715,0.788,respectively).After stepwise variable selection,only MD showed discriminative function in distinguishing tumor heterogeneity in head and neck(p=0.034).Conclusion:Utilization of DKI parameters can help to differentiate tumor benignity from malignancy.MD could be expected to distinguish tumor heterogeneity in head and neck.