Background and purpose: Differentiating central gland prostate cancer (CGPCa) from benign prostatic hyperplasia (BPH) has always been one of the difficult clinical problems. Traditional examination methods can not be used to distinguish them well. Monoexponential and biexponential diffusion weighted imaging (DWI) models can quantitatively reflect the information of tissue diffusion and blood perfusion. This study aimed to investigate the value of parameters derived from monoexponential and biexponential DWI models (parameter derived from monoexponential DWI: ADC

total

; parameters derived from biexponential DWI: ADC slow

ADC fast and f) for distinguishing CGPCa from BPH. Methods: We retrospectively analyzed the megnatic resonance imaging (MRI) and clinical data of 36 patients with CGPCa and 48 patients with BPH ［25 patients with stromal hyperplasia (SH) and 23 patients with glandular hyperplasia (GH)］proved by prostate biopsy at Changhai Hospital

Naval Medical University

Second Military Medical University. MRI examinations included multiple-b-value (0

50

100

150

200

500

800

1 000

1 500

2 000 s/mm

2

) DWI. Drawing the region of interest (ROI)

we could measure the values of ADC

total

ADC

slow

ADC

fast

and f value for the three pathological types. ANOVA and LSD-t test were used to compare the differences of the parameters. By using the receiver operating characteristic (ROC) curves

the diagnostic performances of these parameters were calculated in differentiating CGPCa from BPH. Results: Among the three groups

there were significant differences in both ADC

total

and ADC

slow

(P0.000 1)

while there was no difference in ADC

fast

(P=0.685). There was significant difference in f value between CGPCa and SH (P=0.002) only. In differentiating CGPCa from BPH

the areas under the ROC curves for ADC

total

and ADC

slow

were the same (0.998)

and the diagn

ostic performances of ADC

total

and ADC

slow

were high. The areas under the ROC curve for f value was 0.674. There were significant differences in the areas under the ROC curves between ADC

total

ADC

slow

and f value (P0.000 1). Conclusion: Parameters derived from monoexponential DWI (ADC

total

) and biexponential DWI (ADC

slow

) both can reflect the diffusion of water molecules near the cell structure in tissues

which are of great value in distinguishing CGPCa from BPH.