Could a new PSMA-derived radiomics model help detect intraprostatic lesions on PET/CT?



New research suggests that a radiomics model can provide significantly higher sensitivity, higher specificity and a 22% greater area under the curve (AUC) than nuclear medicine radiologists for the assessment of intraprostatic lesions by tomography by positron emission/computed tomography (PET/CT).

Recognizing that small lesion size, manual analysis of PET/CT data, and different levels of experience of nuclear medicine specialists can contribute to a significant number of missed intraprostatic lesions, researchers recently assessed the predictive viability of a radiomics model based on 68Ga-prostate specific membrane antigen (PSMA)-11 PET.

For the retrospective study, which has just been published in the European Journal of Nuclear Medicine and Molecular Imaging, the researchers built the radiomics model with nine features taken from PET imaging of patients with and without prostate cancer. According to the study, the training set included 61 patients with prostate cancer and 26 patients with benign prostate disease. The study authors noted that the subsequent test set consisted of 25 patients with prostate cancer and 13 patients with benign prostate disease.

Compared to visual assessment of PET imaging by nuclear medicine radiologists, researchers found that the radiomics model had higher AUC (85% vs 63%), higher specificity (77% vs 55% ) and higher sensitivity (84% versus 74%) for detecting intraprostatic lesions.

“Benign intraprostatic processes may be associated with relatively high expression levels of PSMA with significant overlap between low-volume malignancies and benign disease. intraprostatic lesions due to their small size or configuration, leading to potential false positives or false negatives,” wrote Feng Wang, MD, affiliated with the Nuclear Medicine Department of Nanjing First Hospital and Nanjing Medical University in China, and his colleagues.

“In this study, the detected radiomics model (prostate cancer) was based only on a set of numerical features from high-dimensional medical imaging data, regardless of the clinical situation, which could explain its greater sensitivity to reader ratings.”

Noting recent research that has shown the benefits of radiomic features learned from half-glandular segmentation of 68Ga-PSMA-11 PET imaging, Wang and colleagues said the use of whole prostate gland segmentation helped overcome the limitations of PET imaging for small prostate lesions.

“In the present study, further comparison between the radiomics model and the readers revealed that the high-dimensional features provided better disease characterization for (prostate cancer),” Wang and colleagues added.

Regarding study limitations, the study authors acknowledged that the small sample size and single-center design may have affected the reported sensitivity and specificity rates and limit the extrapolation of results. results to larger populations. Noting that they used a PET/CT system for the study and focused on PET imaging data, Wang and colleagues suggested that future research should examine multimodal imaging data and use multiple PET devices. /CT to determine the broader viability of the radiomics model.

Source link


Comments are closed.