Artificial intelligence-based lesion characterization and outcome prediction of prostate cancer on [<sup>18</sup>F]DCFPyL PSMA imaging.
Authors
Affiliations (13)
Affiliations (13)
- Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD, USA; Department of Radiology, Xiangya Hospital, Central South University, Changsha, Hunan, China; Department of Radiology, University of Colorado Denver, Aurora, CO, USA.
- Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD, USA.
- Department of Radiology, Xiangya Hospital, Central South University, Changsha, Hunan, China.
- Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD, USA; Department of Radiology, University of Colorado Denver, Aurora, CO, USA.
- Molecular Imaging Branch, National Cancer Institute, NIH, Bethesda, MD, USA.
- School of Informatics, Hunan University of Chinese Medicine, Changsha, China.
- Department of Radiology, Second Xiangya Hospital, Central South University, Changsha, Hunan, China.
- Department of Radiology, University of Colorado Denver, Aurora, CO, USA.
- Department of Medical Imaging and Intervention, Chang Gung Memorial Hospital at Linkou, Taoyuan, Taiwan.
- Department of Diagnostic Imaging, Rhode Island Hospital, Providence, RI, USA.
- Department of Radiology, University of North Carolina at Chapel Hill, NC, USA.
- Department of Radiology, UT Southwestern Medical Center, Dallas, TX, USA.
- Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD, USA; Department of Radiology, University of Colorado Denver, Aurora, CO, USA. Electronic address: [email protected].
Abstract
Prostate cancer remains a significant clinical challenge, particularly in characterizing lesions and predicting patient outcomes. With the growing availability of advanced imaging techniques like [18F]DCFPyL PET/CT, there is an urgent need for intelligent tools that can facilitate clinical decision-making. This study aimed to develop artificial intelligence (AI) models for lesion characterization and outcome prediction in prostate cancer (PCa) patients. PCa patients who underwent [18F]DCFPyL PET/CT imaging were divided into training and internal test sets (n = 238) and a prospective test set (n = 36). Lesions were scored using the PSMA-Reporting and Data System (RADS) and assessed for malignancy, treatment response, and survival outcomes. Single- and multi-modality deep learning models were trained for four tasks: PSMA-RADS scoring, malignancy classification, treatment response prediction, and survival prediction. The input concatenation model, which combined PET and CT modalities, demonstrated superior performance across all tasks. For the internal test set, the area under the receiver operating characteristic curves (AUROCs) were 0.81 (95 % CI: 0.80-0.81) for PSMA-RADS scoring, 0.79 (95 % CI: 0.78-0.80) for malignancy classification, and 0.74 (95 % CI: 0.73-0.77) for treatment response prediction. In the prospective test set, the AUROCs were 0.72 (95 % CI: 0.69-0.75) for PSMA-RADS scoring, 0.70 (95 % CI: 0.68-0.71) for malignancy classification, and 0.70 (95 % CI: 0.67-0.72) for treatment response prediction. The C-indices for survival predictions were 0.58 (95 % CI: 0.57-0.59) and 0.60 (95 % CI: 0.60-0.63) for the internal and prospective test sets, respectively. Our study highlights the potential of AI to improve lesion characterization and identify patients at high risk of disease progression.