ABSTRACT
Objective To investigate the value of ADC histogram of tumor volume measurement in the diagnosis of prostate cancer,and to screen out the best diagnostic parameter value.Methods 31 cases of prostate cancer and 35 cases of benign prostatic hyperplasia confirmed by biopsy or surgical pathology were analyzed retrospectively.DWI examination was performed on all patients before treatment,and b value of 0 and 1500 s/mm2 was selected.The total tumor ADC histogram parameters were measured respectively,including the average value of ADC (ADCmean ),the median ADC (ADCmedian ),the tenth percentile ADC (ADC10th),the twenty-fifth percentile ADC (ADC25th), the fiftieth percentile ADC (ADC50th),the seventy-fifth percentile ADC (ADC75th),the ninetieth percentile ADC (ADC90th), skewness and kurtosis.The histogram parameters of the two groups of patinents and their diagnostic efficacy were analyzed and compared.Results ADCmean ,ADCmedian ,ADC10th,ADC25th,ADC50th,ADC75th and ADC90th in the prostate cancer group were statistically lower than those of benign prostatic hyperplasia group (P <0.01),and there was no significant difference in the skewness and kurtosis between the two groups of diseases (P > 0.05 ).ADCmean ,ADCmedian ,ADC10th,ADC25th,ADC50th,ADC75th and ADC90th diagnosing prostate cancer in the area under ROC curves (AUC)were more than 0.78.ADC10th had the best diagnostic efficacy and its AUC was 0.82, with the optimal cut-off value for 0.27 × 10 -3 mm2/s,with sensitivity and specificity for 78.4％ and 83.3％.Conclusion The ADC histogram of the total tumor volume measurement is of great value in the diagnosis of prostate cancer,among which ADC10th is the most effective parameter.It can accurately distinguish between prostate cancer and prostatic hyperplasia nodules.
ABSTRACT
Objective To investigate the value of ADC histogram of tumor volume measurement in the diagnosis of prostate cancer,and to screen out the best diagnostic parameter value.Methods 31 cases of prostate cancer and 35 cases of benign prostatic hyperplasia confirmed by biopsy or surgical pathology were analyzed retrospectively.DWI examination was performed on all patients before treatment,and b value of 0 and 1500 s/mm2 was selected.The total tumor ADC histogram parameters were measured respectively,including the average value of ADC (ADCmean ),the median ADC (ADCmedian ),the tenth percentile ADC (ADC10th),the twenty-fifth percentile ADC (ADC25th), the fiftieth percentile ADC (ADC50th),the seventy-fifth percentile ADC (ADC75th),the ninetieth percentile ADC (ADC90th), skewness and kurtosis.The histogram parameters of the two groups of patinents and their diagnostic efficacy were analyzed and compared.Results ADCmean ,ADCmedian ,ADC10th,ADC25th,ADC50th,ADC75th and ADC90th in the prostate cancer group were statistically lower than those of benign prostatic hyperplasia group (P <0.01),and there was no significant difference in the skewness and kurtosis between the two groups of diseases (P > 0.05 ).ADCmean ,ADCmedian ,ADC10th,ADC25th,ADC50th,ADC75th and ADC90th diagnosing prostate cancer in the area under ROC curves (AUC)were more than 0.78.ADC10th had the best diagnostic efficacy and its AUC was 0.82, with the optimal cut-off value for 0.27 × 10 -3 mm2/s,with sensitivity and specificity for 78.4％ and 83.3％.Conclusion The ADC histogram of the total tumor volume measurement is of great value in the diagnosis of prostate cancer,among which ADC10th is the most effective parameter.It can accurately distinguish between prostate cancer and prostatic hyperplasia nodules.
ABSTRACT
OBJECTIVE: This study is performed to evaluate the role of gamma knife radiosurgery(GKS) for treatment of patients with large number of metastatic brain lesions. METHODS: The authors retrospectively reviewed the clinical outcome of 99 patients who underwent 124 GKS procedures for a total of 564 metastatic lesions. Sixty-three patients presented with 1 to 3 lesions (group A) and 36 patients presented with 4 or more lesions (group B) at initial GKS. The authors assessed survival according to a variety of possible prognostic factors. RESULTS: Median survival from the GKS for all cases was 28 weeks. The recursive partitioning analysis (RPA) class (p<0.0001) and the total tumor volume (p=0.0174) were significant factors influencing survival. The number of lesions, whole brain radiotherapy, primary tumor site, age, and sex were not significant. Median survival of group A (28 weeks) was not different from that of group B (28 weeks) (p=0.4027). Median survival was 66, 28, and 13 weeks for RPA class I, II, and III respectively in group A and 44, 36, and 11 weeks in group B. CONCLUSION: It is suggested that GKS provides a survival gain even in patients with a large number (4 or more) of metastatic lesions. Concerning the survival and the selection of patients for radiosurgery, RPA class and the total tumor volume should be considered as more important factors than the number of lesions. Multiplicity of the lesions alone should not be a factor for contraindication of radiosurgery.