|
Value of a model combining clinical indicators with qualitative and quantitative contrast-enhanced ultrasound parameters for differentiating benign and malignant prostatic lesions
|
L. Yu   , F. Shen , X. Chen , L. Han , Y. Qiu |
| Department of Ultrasound, Haining Central Hospital, Haining, China , qiuyun94028455@163.com |
|
|
Abstract: (11 Views) |
Background: Benign prostate lesions and cancer can be difficult to separate when the judgment depends on one test only, this study reviewed routine clinical data, visual CEUS signs, and measured CEUS blood-flow values, then combined useful items in a simple prediction model. Materials and Methods: We reviewed 120 patients from one hospital, all had pathological proof of a prostate lesion and underwent transrectal ultrasound plus CEUS between January 2021 and December 2025; clinical indexes, image signs, and curve-based CEUS values were checked in benign and malignant cases, related variables were screened with logistic regression, and ROC analysis together with bootstrap resampling was used for model checking. Results: The malignant cases showed an unfavorable PSA pattern and abnormal DRE more often than benign cases, on CEUS they also had more hyperenhancement, uneven enhancement, early wash-in, and perfusion defects; the curve data suggested quicker and stronger enhancement. In the adjusted analysis, f/tPSA, PSAD, early wash-in, and PI remained related to cancer. The model reached an AUC of 0.930, with sensitivity of 82.2% and specificity of 89.3%; the corrected AUC after bootstrap testing was 0.918. Conclusion: Putting clinical indexes together with visual and numerical CEUS information gave a useful and internally stable way to separate benign prostate lesions from malignant ones. |
|
| Keywords: Prostatic neoplasms, ultrasonography, contrast media, prostate-specific antigen, differential diagnosis. |
|
|
Full-Text [PDF 917 kb]
(3 Downloads)
|
|
Type of Study: Original Research |
Subject:
Radiation Biology
|
|
|
|
|
|
|
| Send email to the article author |
|
|
|
| Add your comments about this article |
|
|
|
