Do plaque-related factors affect the diagnostic performance of an artificial intelligence coronary-assisted diagnosis system? Comparison with invasive coronary angiography.

OBJECTIVE: The aim of this study was to investigate the effects of plaque-related factors on the diagnostic performance of an artificial intelligence coronary-assisted diagnosis system (AI-CADS). METHODS: Patients who underwent coronary computed tomography angiography (CCTA) and invasive coronary angiography (ICA) were retrospectively included in this study. The degree of stenosis in each vessel was collected from CCTA and ICA, and the information on plaque-related factors (plaque length, plaque type, and coronary artery calcium score (CAC)) of the vessels with plaques was collected from CCTA. RESULTS: In total, 1224 vessels in 306 patients (166 men; 65.7 ± 10.1 years) were analyzed. Of these, 391 vessels in 249 patients showed significant stenosis using ICA as the gold standard. Using per-vessel as the unit, the area under the curves of coronary stenosis ≥ 50% for AI-CADS, doctor, and AI-CADS + doctor was 0.764, 0.837, and 0.853, respectively. The accuracies in interpreting the degree of coronary stenosis were 56.0%, 68.1%, and 71.2%, respectively. Seven hundred fifty vessels showed plaques on CCTA; plaque type did not affect the interpretation results by AI-CADS (chi-square test: p = 0.0093; multiple logistic regression: p = 0.4937). However, the interpretation results for plaque length (chi-square test: p < 0.0001; multiple logistic regression: p = 0.0061) and CACs (chi-square test: p < 0.0001; multiple logistic regression: p = 0.0001) were significantly different. CONCLUSION: AI-CADS has an ability to distinguish ≥ 50% coronary stenosis, but additional manual interpretation based on AI-CADS is necessary. The plaque length and CACs will affect the diagnostic performance of AI-CADS. KEY POINTS: • AI-CADS can help radiologists quickly assess CCTA and improve diagnostic confidence. • Additional manual interpretation on the basis of AI-CADS is necessary. • The plaque length and CACs will affect the diagnostic performance of AI-CADS.

Radiomic-based machine learning model for the accurate prediction of prostate cancer risk stratification.

OBJECTIVES: To precisely predict prostate cancer (PCa) risk stratification, we constructed a machine learning (ML) model based on magnetic resonance imaging (MRI) radiomic features. METHODS: Between August 2016 and May 2021, patients with histologically proven PCa who underwent pre-operative MRI and prostate-specific antigen screening were included. The patients were grouped into different risk categories as defined by the European Association of Urology-European Association of Nuclear Medicine-European Society for Radiotherapy and Oncology-European Society of Urogenital Radiology-International Society of Geriatric Oncology guidelines. Using Artificial Intelligence Kit software, PCa regions of interest were delineated and radiomic features were extracted. Subsequently, predictable models were built by utilising five traditional ML approaches: support vector machine, logistic regression, gradient boosting decision tree, k-nearest neighbour and random forest (RF) classifiers. The classification capacity of the developed models was assessed by area under the receiver operating characteristic curve (AUC) analysis. RESULTS: A total of 213 patients were enrolled, including 16 low-risk, 65 intermediate-risk, and 132 high-risk PCa patients. The risk stratification of PCa could be revealed by MRI radiomic features, and second-order features accounted for most of the selected features. Among the five established ML models, the RF model showed the best overall predictive performance (AUC = 0.87). After further analysis of the subgroups based on the RF model, the prediction of the high-risk group was the best (AUC = 0.89). CONCLUSION: This study demonstrated that the MR radiomics-based ML method could be a promising tool for predicting PCa risk stratification precisely. ADVANCES IN KNOWLEDGE: The ML models have valuable prospect for accurate PCa risk assessment, which might contribute to customize treatment and surveillance strategies.

Peritumoral Radiomics Strategy Based on Ensemble Learning for the Prediction of Gleason Grade Group of Prostate Cancer.

RATIONALE AND OBJECTIVES: To develop and evaluate a peritumoral radiomic-based machine learning model to differentiate low-Gleason grade group (L-GGG) and high-GGG (H-GGG) prostate lesions. MATERIALS AND METHODS: In this retrospective study, a total of 175 patients with prostate cancer (PCa) confirmed by puncture biopsy were recruited and included 59 patients with L-GGG and 116 patients with H-GGG. The original PCa regions of interest (ROIs) were delineated on T2-weighted (T2WI), diffusion-weighted imaging (DWI), and apparent diffusion coefficient (ADC) maps, and then centra-tumoral and peritumoral ROIs were defined. Features were meticulously extracted from each ROI to establish radiomics models, employing distinct sequence datasets. Peritumoral radiomics models were specifically developed for both the peripheral zone (PZ) and transitional zone (TZ), utilizing dedicated PZ and TZ datasets, respectively. The performances of the models were evaluated by using the receiver operating characteristic (ROC) curve and precision-recall curve. RESULTS: The classification model with combined peritumoral features based on T2 + DWI + ADC sequence dataset demonstrated superior performance compared to the original tumor and centra-tumoral classification models. It achieved an area under the ROC curve (AUC) of 0.850 [95% confidence interval, 0.849, 0.860] and an average accuracy of 0.950. The combined peritumoral model outperformed the regional peritumoral models with AUC of 0.85 versus 0.75 for PZ lesions and 0.88 versus 0.69 for TZ lesions, respectively. The peritumoral classification models exhibit greater efficacy in predicting PZ lesions as opposed to TZ lesions. CONCLUSION: The peritumoral radiomics features showed excellent performance in predicting GGG in PCa patients and might be a valuable addition to the non-invasive assessment of PCa aggressiveness.

MRI Radiomics-Based Machine Learning Models for Ki67 Expression and Gleason Grade Group Prediction in Prostate Cancer.

PURPOSE: The Ki67 index and the Gleason grade group (GGG) are vital prognostic indicators of prostate cancer (PCa). This study investigated the value of biparametric magnetic resonance imaging (bpMRI) radiomics feature-based machine learning (ML) models in predicting the Ki67 index and GGG of PCa. METHODS: A total of 122 patients with pathologically proven PCa who had undergone preoperative MRI were retrospectively included. Radiomics features were extracted from T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), and apparent diffusion coefficient (ADC) maps. Then, recursive feature elimination (RFE) was applied to remove redundant features. ML models for predicting Ki67 expression and GGG were constructed based on bpMRI and different algorithms, including logistic regression (LR), support vector machine (SVM), random forest (RF), and K-nearest neighbor (KNN). The performances of different models were evaluated with receiver operating characteristic (ROC) analysis. In addition, a joint analysis of Ki67 expression and GGG was performed by assessing their Spearman correlation and calculating the diagnostic accuracy for both indices. RESULTS: The ML model based on LR and ADC + T2 (LR_ADC + T2, AUC = 0.8882) performed best in predicting Ki67 expression, and ADC_wavelet-LHH_firstorder_Maximum had the highest feature weighting. The SVM_DWI + T2 (AUC = 0.9248) performed best in predicting GGG, and DWI_wavelet HLL_glcm_SumAverage had the highest feature weighting. The Ki67 and GGG exhibited a weak positive correlation (r = 0.382, p < 0.001), and LR_ADC + DWI had the highest diagnostic accuracy in predicting both (0.6230). CONCLUSION: The proposed ML models are suitable for predicting both Ki67 expression and GGG in PCa. This algorithm could be used to identify indolent or invasive PCa with a noninvasive, repeatable, and accurate diagnostic method.

Radiomics-Based Machine Learning Models for Predicting P504s/P63 Immunohistochemical Expression: A Noninvasive Diagnostic Tool for Prostate Cancer.

Objective: To develop and validate a noninvasive radiomic-based machine learning (ML) model to identify P504s/P63 status and further achieve the diagnosis of prostate cancer (PCa). Methods: A retrospective dataset of patients with preoperative prostate MRI examination and P504s/P63 pathological immunohistochemical results between June 2016 and February 2021 was conducted. As indicated by P504s/P63 expression, the patients were divided into label 0 (atypical prostatic hyperplasia), label 1 (benign prostatic hyperplasia, BPH) and label 2 (PCa) groups. This study employed T2WI, DWI and ADC sequences to assess prostate diseases and manually segmented regions of interest (ROIs) with Artificial Intelligence Kit software for radiomics feature acquisition. Feature dimensionality reduction and selection were performed by using a mutual information algorithm. Based on screened features, P504s/P63 prediction models were established by random forest (RF), gradient boosting decision tree (GBDT), logistic regression (LR), adaptive boosting (AdaBoost) and k-nearest neighbor (KNN) algorithms. The performance was evaluated by the area under the ROC curve (AUC) and accuracy. Results: A total of 315 patients were enrolled. Among the 851 radiomic features, the 32 top features were derived from T2WI, in which the gray-level run length matrix (GLRLM) and gray-level cooccurrence matrix (GLCM) features accounted for the largest proportion. Among the five models, the RF algorithm performed best in general evaluations (microaverage AUC=0.920, macroaverage AUC=0.870) and provided the most accurate result in further sublabel prediction (the accuracies of label 0, 1, and 2 were 0.831, 0.831, and 0.932, respectively). In comparative sequence analyses, T2WI was the best single-sequence candidate (microaverage AUC=0.94 and macroaverage AUC=0.78). The merged datasets of T2WI, DWI, and ADC yielded optimal AUCs (microaverage AUC=0.930 and macroaverage AUC=0.900). Conclusions: The radiomic-based RF classifier has the potential to be used to evaluate the presurgical P504s/P63 status and further diagnose PCa noninvasively and accurately.