Enhancing Prostate Cancer Detection: Integrating Multiparametric Magnetic Resonance Imaging and 68Ga-prostate-specific Membrane Antigen Positron Emission Tomography/Computed Tomography with Machine Learning
Year: 2025
Type: article
Study design: Observational Study
Abstract: Purpose: Early and accurate detection of clinically significant prostate cancer (PCa) is crucial for effective patient management. Traditional diagnostic methods, including systematic biopsy guided by transrectal ultrasound, have limitations in detecting significant cancers. Multiparametric magnetic resonance imaging (mpMRI) has shown promise in overcoming these limitations, but it remains operator-dependent and may miss some significant cases. This study explores the integration of mpMRI with 68 Ga-prostate-specific membrane antigen positron emission tomography/computed tomography ( 68 Ga-PSMA PET/CT) to enhance diagnostic accuracy. Materials and Methods: This study included 15 patients with suspected PCa, who underwent mpMRI and 68 Ga-PSMA PET/CT imaging, followed by histopathological confirmation after radical prostatectomy. Imaging data were analyzed using advanced machine learning (ML) models, including support vector machine, random forest, logistic regression, and k-nearest neighbors, to assess diagnostic accuracy. Sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were calculated to compare the performance of the imaging modalities. Results: MpMRI demonstrated an AUC of 0.91, with a sensitivity of 77% and specificity of 89% at the selected threshold (Prostate Imaging-Reporting and Data System 4). PET/CT alone showed an AUC of 0.89, with higher sensitivity (90%) but lower specificity (72%). The combination of mpMRI and PET/CT did not significantly improve the overall diagnostic performance, as indicated by a net reclassification index of −3% ( P = 0.84). Conclusions: While 68 Ga-PSMA PET/CT improves sensitivity in detecting PCa, its lower specificity limits the incremental benefit when combined with mpMRI. ML models show potential in optimizing diagnostic approaches, but further research with larger cohorts is needed to confirm these findings and refine the integration of imaging modalities.
Source: Journal of Medical Physics
Language: English
Subfield: Pulmonary and Respiratory Medicine
Field: Medicine
Domain: Health Sciences
Open Access status: diamond
OpenAlex ID: W4411972662