Luigi Benecchi | PubMed Health

PubMed2006

Neuro-fuzzy system for prostate cancer diagnosis.

OBJECTIVES: To develop a neuro-fuzzy system to predict the presence of prostate cancer. Neuro-fuzzy systems harness the power of two paradigms: fuzzy logic and artificial neural networks. We compared the predictive accuracy of our neuro-fuzzy system with that obtained by total prostate-specific antigen (tPSA) and percent free PSA (%fPSA). METHODS: The data from 1030 men (both outpatients and hospitalized patients) were used. All men had a tPSA level of less than 20 ng/mL. Of the 1030 men, 195 (18.9%) had prostate cancer. A neuro-fuzzy system was developed using the coactive neuro-fuzzy inference system model. RESULTS: The mean area under the receiver operating characteristic curve for the neuro-fuzzy system output was 0.799 +/- 0.029 (95% confidence interval 0.760 to 0.835), for tPSA, it was 0.724 +/- 0.032 (95% confidence interval 0.681 to 0.765), and for %fPSA, 0.766 +/- 0.024 (95% confidence interval 0.725 to 0.804). Furthermore, pairwise comparison of the area under the curves evidenced differences among %fPSA, tPSA, and neuro-fuzzy system's output (tPSA versus neuro-fuzzy system's output, P = 0.008; %fPSA versus neuro-fuzzy system's output, P = 0.032). The comparison at 95% sensitivity showed that the neuro-fuzzy system had the best specificity (31.9%). CONCLUSIONS: This study presented a neuro-fuzzy system based on both serum data (tPSA and %fPSA) and clinical data (age) to enhance the performance of tPSA to discriminate prostate cancer. The predictive accuracy of the neuro-fuzzy system was superior to that of tPSA and %fPSA.

Aged↗

PubMed2006

Optimum PSA reflex-range.

OBJECTIVE: Aim of this study was to evaluate different decision strategies based on variations in the cut-off value of percent free PSA and in the range of total PSA values (reflex range) in which free PSA testing was applied. We compared these strategies to conventional total PSA testing by determining which strategies would provide a maximum decrease in unnecessary biopsies with a minimum number of additional undetected cancers. MATERIALS AND METHODS: This retrospective study was conducted with 807 patients who were referred to transrectal ultrasound biopsies for elevated serum PSA levels or for abnormal digital rectal examination. Overall 156 patients were affected by primary prostate cancer (CaP), 651 were controls without prostate cancer (benign prostatic hypertrophy, prostatic intraepithelial neoplasm, prostatitis or normal prostatic gland). RESULTS: Total PSA was significantly higher (F=4.93; p<0.0001) and percent free PSA was significantly lower in cancer patients than in controls (F=2.16; p<0.0001). Sensitivity, specificity and the positive likelihood ratio (LR+) of PSA and percent free PSA have been calculated for several total PSA intervals: PSA 4-10 ng/ml, 3-10 ng/ml, 3-20 ng/ml, 2-10 ng/ml, 2-20 ng/ml. These data suggest that in the reflex range 2-10 ng/ml there are the best results. On ROC comparison restricted to men with total PSA between 2 and 10 ng/ml, percent free PSA also had a higher area under the curve than total PSA (AUC 0.7452 for free percent PSA; 0.6267 for total PSA: p = 0.0059). CONCLUSIONS: The PSA revolution that occurred over the previous 2 decades has positively impacted the detection of prostate cancer. Percent free PSA improves specificity, at the beginning the percent free PSA was used only in the gray zone 4-10 ng/ml. Analyzing our data, we confirm that the usefulness of percent free PSA in prostate cancer diagnosis increases, enlarging the reflex range. Our best result is obtained in the reflex range 2-10 ng/ml using a percent free PSA cut-off of 22%.

Aged↗