A Nomogram Based on a Multiparametric Ultrasound Radiomics Model for Discrimination Between Malignant and Benign Prostate Lesions

Abstract

ObjectivesTo evaluate the potential of a clinical-based model, a multiparametric ultrasound-based radiomics model, and a clinical-radiomics combined model for predicting prostate cancer (PCa).MethodsA total of 112 patients with prostate lesions were included in this retrospective study. Among them, 58 patients had no prostate cancer detected by biopsy and 54 patients had prostate cancer. Clinical risk factors related to PCa (age, prostate volume, serum PSA, etc.) were collected in all patients. Prior to surgery, patients received transrectal ultrasound (TRUS), shear-wave elastography (SWE) and TRUS-guided prostate biopsy. We used the five-fold cross-validation method to verify the results of training and validation sets of different models. The images were manually delineated and registered. All modes of ultrasound radiomics were retrieved. Machine learning used the pathology of “12+X” biopsy as a reference to draw the benign and malignant regions of interest (ROI) through the application of LASSO regression. Three models were developed to predict the PCa: a clinical model, a multiparametric ultrasound-based radiomics model and a clinical-radiomics combined model. The diagnostic performance and clinical net benefit of each model were compared by receiver operating characteristic curve (ROC) analysis and decision curve.ResultsThe multiparametric ultrasound radiomics reached area under the curve (AUC) of 0.85 for predicting PCa, meanwhile, AUC of B-mode radiomics and SWE radiomics were 0.74 and 0.80, respectively. Additionally, the clinical-radiomics combined model (AUC: 0.90) achieved greater predictive efficacy than the radiomics model (AUC: 0.85) and clinical model (AUC: 0.84). The decision curve analysis also showed that the combined model had higher net benefits in a wide range of high risk threshold than either the radiomics model or the clinical model.ConclusionsClinical-radiomics combined model can improve the accuracy of PCa predictions both in terms of diagnostic performance and clinical net benefit, compared with evaluating only clinical risk factors or radiomics score associated with PCa.

Radiomics analysis based on CT’s greater omental caking for predicting pathological grading of pseudomyxoma peritonei

“…Wise Multimodal Research Platform (https:/keyan.deepwise.com; Beijing Deepwise & League of PHD Technology Co., Ltd, 193 Beijing, China) was used for the CT’s feature extraction. Approximately 1218 features were extracted from the CT lesion’s ROI extracted and categorized into seven categories: first order features, shape based features, gray-scale co-occurrence Matrix (GLCM) features, gray-level size zone matrix (GLSZM) features, gray-level run length matrix (GLRLM) features, gray-level distance-zone matrix (GLDM), and neighboring gray level dependence matrix 16 .…”

Deep Learning With Radiomics for Disease Diagnosis and Treatment: Challenges and Potential

“…Apart from the three methods already described, unsupervised techniques are another effective means to reduce data dimensionality ( 68 , 69 ). Mapping the characteristic set to a lower-dimensional space by linear or non-linear transformation minimizes information loss.…”

“…Cross-validation is the most prevalent scheme. K - fold cross-validation divides the samples into k disjoint subsets, where k – 1 is the training set, and the remaining is the test set ( 68 ). The leave-one-out cross-validation method ensures that the value of k is equal to the number of examples and only one is used for testing at a time ( 87 ).…”

Prostate Cancer Aggressiveness Prediction Using CT Images

“…The addition of radiomic features to ultrasound images may provide the ability to diagnose PCa without any of these issues. The power of radiomic features to distinguish clinically significant PCa based on ultrasound images has been addressed by Wildeboer et al [13], Liang et al [14] with promising results. Liang et al [14] also added clinical parameters as age, prostate volume, PSA and others.…”

“…The power of radiomic features to distinguish clinically significant PCa based on ultrasound images has been addressed by Wildeboer et al [13], Liang et al [14] with promising results. Liang et al [14] also added clinical parameters as age, prostate volume, PSA and others. Both studies provided the baseline for deeper analysis using ultrasound images revealing also the potential to use radiomics in an early stage of PCa evaluation.…”