Related Experiment Video For Diagnosis
Updated: Feb 20, 2026
Detection and Isolation of Cancer in Prostate Biopsies Using Stimulated Raman Histology and Artificial Intelligence
Published on: June 10, 2025
[Use of prostate health index in diagnosing prostate cancer in Chinese men]
1Department of Urology, Fudan University Shanghai Cancer Center; Department of Oncology, Shanghai Medical College, Fudan University, Shanghai 200032, China.
Abstract:
Objective: To investigate the value of prostate health index (PHI) in the diagnosis of prostate cancer in patients with total prostate specific antigen (tPSA) <20 μg/L. Methods: Totally 1 135 patients with tPSA<20 μg/L and prostate biopsy indications at Department of Urology, Fudan University Shanghai Cancer Center from March 2013 to April 2016 were enrolled in this study. They were tested for serum tPSA, free prostate specific antigen and prostate specific antigen isoform 2, from which PHI was calculated. Diagnostic efficacy of PHI and tPSA were evaluated using receiver operating characteristic (ROC) curve analysis. The detection rates of prostate cancer were calculated in different ranges of PHI. Subgroup analysis of 716 patients, who were aged 50 or above with tPSA in the range of 4 to 10 μg/L and digital rectal examination negative, was performed. Results: In the biopsied objects with tPSA<20 μg/L, PHI was significantly higher in prostate cancer patients than that in non-cancer patients (48.4(37.4) vs. 26.5(16.9), U=52 674.00, P=0.000), PHI was also significantly higher in high-grade prostate cancer patients than that of low-grade prostate cancer patients (44.5(30.8) vs. 56.4(42.5), U=23 314.00, P=0.000). The area under the curve (AUC) of PHI for diagnosing prostate cancer was significantly higher than that of tPSA (0.771 vs. 0.627, P=0.000). When PHI was in the range of <27, 27 to <36, 36 to <55 and ≥55, the probability of prostate cancer was 9.4% (95%CI: 7.0% to 12.2%), 16.3% (95%CI: 12.2% to 20.8%), 31.0% (95%CI: 25.9% to 37.3%) and 66.4% (95%CI: 58.9% to 74.2%), respectively. Subgroup analysis showed that the AUC of PHI in diagnosing prostate cancer was significantly higher than that of tPSA (0.764 vs. 0.569, P=0.000). When PHI was in the range of <27, 27 to <36, 36 to <55 and ≥55, the probability of prostate cancer was 8.1% (95%CI: 5.4% to 11.3%), 14.0% (95%CI: 9.1% to 19.9%), 30.8% (95%CI: 23.6% to 38.7%) and 78.8% (95%CI: 66.7% to 88.9%), respectively. Conclusion: PHI is superior to tPSA in the diagnosis of prostate cancer in Chinese men with tPSA<20 μg/L.
