Selected Radiomics Features Research Articles

A contrast-enhanced CT-based whole-spleen radiomics signature for early prediction of oxaliplatin-related thrombocytopenia in patients with gastrointestinal malignancies: a retrospective study.

Thrombocytopenia is a common adverse event of oxaliplatin-based chemotherapy. Grade 2 or higher oxaliplatin-related thrombocytopenia may result in dose reduction, discontinuation or delay initiation of chemotherapy and may adversely affect the therapeutic efficacy and even overall survival of patients. Early recognition of patients at risk of developing grade 2 or higher thrombocytopenia is critical. However, to date there is no well-established method to early identify patients at high risk. The aims of this study were to develop and validate a contrast-enhanced CT-based whole-spleen radiomics signature for early prediction of grade 2 or higher thrombocytopenia in patients with gastrointestinal malignancies treated with oxaliplatin-based chemotherapy and to explore the incremental value of combining the radiomics signature and conventional clinical factors for risk prediction. A total of 119 patients with gastrointestinal malignancies receiving oxaliplatin-based chemotherapy from March 2017 to December 2020 were retrospectively included and randomly divided into a training cohort (n=85) and a validation cohort (n=34). Grade 2 or higher thrombocytopenia occurred in 26.1% of patients (22 and nine patients in the training and validation cohort, respectively) with a median time interval of 101 days from the start of chemotherapy. The whole-spleen radiomics features were extracted on the portal venous phase of the first follow-up CT images. The least absolute shrinkage and selection operator (LASSO) algorithm was applied to select radiomics features and to build the radiomics signature for the prediction of grade 2 or higher thrombocytopenia. A clinical model that included clinical factors only and a clinical-radiomics model that incorporated clinical factors and radiomics signature were constructed. The performances of both models were evaluated and compared in the training, validation and the whole cohorts. The radiomics signature yielded favorable performance in predicting grade 2 or higher thrombocytopenia, with the area under the curve (AUC), sensitivity and specificity being 0.865, 81.8% and 84.1% in the training cohort and 0.747, 77.8% and 80.0% in the validation cohort. The AUCs of the clinical-radiomics model in the training and validation cohorts reached 0.913 (95% CI [0.720-0.935]) and 0.867 (95% CI [0.727-1.000]), greater than the AUCs of the clinical model. Integrated discrimination improvement (IDI) index showed that incorporating radiomic signature into conventional clinical factors significantly improved the predictive accuracy by 17.0% (95% CI [4.9%-29.1%], p = 0.006) in the whole cohort. Contrast-enhanced CT-based whole-spleen radiomics signature might serve as an early predictor for grade 2 or higher thrombocytopenia during oxaliplatin-based chemotherapy in patients with gastrointestinal malignancies and provide incremental value over conventional clinical factors.

A Novel Multi-Objective Based Feature Selection Method for Response Prediction

Accurate response prediction is essential towards personalized treatment in radiation therapy. Excessive imaging features, extracted from medical images, pose a great challenge in radiomic analyses. Feature selection is an essential step to remove redundant and irrelevant features for model construction. We proposed a novel multi-objective based radiomic feature selection method (MRMOPSO), where the number of features, sensitivity, and specificity are jointly considered as optimization objectives for feature selection. The MRMOPSO innovated by three aspects: 1) Fisher score initialize the feature population to speed up the convergence; 2) Min-redundancy particle generation operations to reduce the redundancy between radiomic features, a truncation strategy was also introduced; 3) Particle selection operation guided by elitism strategies to improve local search ability of the algorithm. We evaluated the effectiveness of the proposed MRMOPSO method using a cohort of oropharyngeal cancer patients from The Cancer Imaging Archive (TCIA). 357 patients were used for model training and additional 64 patients were used for independent evaluation. The proposed methods were compared with (a) classical feature selection methods, i.e., Lasso, minimal-redundancy-maximal-relevance criterion (mRMR), F-score, and mutual information (MI), (b) single-objective feature selection methods, i.e., genetic algorithm (GA), particle swarm optimization algorithm (PSO) and (c) multi-objective feature selection methods, i.e., multiple objective particle swarm optimization (MOPSO), nondominated sorting genetic algorithm II (NSGA II). The other feature selection methods yielded AUCs, sensitivity, specificity of (0.48-0.71), (0.49-0.86), (0.33-0.67), respectively. The MRMOPSO achieved significantly highly AUC of 0.84 with smaller number of selected features on the independent dataset (Table 1). Additionally, the MRMOPSO remarkably improved the sensitivity (0.81), specificity (0.81) and achieved an excellent balance between sensitively and specificity. We demonstrated a novel multi-objective based radiomic feature selection method. The proposed algorithm effectively reduced feature dimension, and achieved superior AUC with simultaneous improved sensitivity and specificity, for radiomic response prediction.

An MR-based radiomics nomogram including information from the peritumoral region to predict deep myometrial invasion in stage I endometrioid adenocarcinoma: a preliminary study.

To develop and validate an MR-based radiomics nomogram combining different imaging sequences (ADC mapping and T2 weighted imaging (T2WI)), different tumor regions (combined intra- and peritumoral regions), and different parameters (clinical features, tumor morphological features, and radiomics features) while considering different MR field strengths in predicting deep myometrial invasion (MI) in Stage I endometrioid adenocarcinoma (EEA). A total of 202 patients were retrospectively analyzed and divided into two cohorts (training cohort, 1.5 T MR, n = 131; validation cohort, 3.0 T MR, n = 71). Axial ADC mapping and T2WI were conducted. Radiomics features were extracted from intra- and peritumoral regions. Least absolute shrinkage and selection operator regression, univariate analysis, and multivariate logistic regression were used to select radiomics features and tumor morphological and clinical parameters. The area under the receiver operator characteristic curve (AUC) was calculated to evaluate the performance of the prediction model and radiomics nomogram. Ten radiomics features, 4 morphological parameters and 1 clinical characteristic were selected. The radiomics nomogram achieved good discrimination between the superficial and deep MI cohorts. The AUC was 0.927 (95% confidence interval [CI]: 0.865, 0.967) in the training cohort and 0.921 (95% CI: 0.872, 0.948) in the validation cohort. The specificity and sensitivity were 92.0 and 78.9% in the training cohort and 83.0 and 77.8% in the validation cohort, respectively. The radiomics nomogram showed good performance in predicting the depth of MI in Stage I EEA before surgery and might be useful for surgical patient management. An MR-based radiomics nomogram was useful for predicting deep MI in Stage I EEA patients (AUCtrain = 0.927, AUCvalidation = 0.921). The intra- and peritumoral radiomics features complemented each other. The nomogram was developed and validated with different MR field strengths, suggesting that the model demonstrates good generalizability.

Radiomics Features on Magnetic Resonance Images Can Predict C5aR1 Expression Levels and Prognosis in High-Grade Glioma.

The complement component C5a receptor 1 (C5aR1) regulates cancer immunity. This retrospective study aimed to assess its prognostic value in high-grade glioma (HGG) and predict C5aR1 expression using a radiomics approach. Among 298 patients with HGG, 182 with MRI data were randomly divided into training and test groups for radiomics analysis. We examined the association between C5aR1 expression and prognosis through Kaplan-Meier and Cox regression analyses. We used maximum relevance-minimum redundancy and recursive feature elimination algorithms for radiomics feature selection. We then built a support vector machine (SVM) and a logistic regression model, investigating their performances using receiver operating characteristic, calibration curves, and decision curves. C5aR1 expression was elevated in HGG and was an independent prognostic factor (hazard ratio = 3.984, 95% CI: 2.834-5.607). Both models presented with >0.8 area under the curve values in the training and test datasets, indicating efficient discriminatory ability, with SVM performing marginally better. The radiomics score calculated using the SVM model correlated significantly with overall survival (p < 0.01). Our results highlight C5aR1's role in HGG development and prognosis, supporting its potential as a prognostic biomarker. Our radiomics model can noninvasively and effectively predict C5aR1 expression and patient prognosis in HGG.

Preoperative prediction of pancreatic neuroendocrine tumor grade based on 68Ga-DOTATATE PET/CT

ObjectiveTo establish a prediction model for preoperatively predicting grade 1 and grade 2/3 tumors in patients with pancreatic neuroendocrine tumors (PNETs) based on 68Ga-DOTATATE PET/CT.MethodsClinical data of 41 patients with PNETs were included in this study. According to the pathological results, they were divided into grade 1 and grade 2/3. 68Ga-DOTATATE PET/CT images were collected within one month before surgery. The clinical risk factors and significant radiological features were filtered, and a clinical predictive model based on these clinical and radiological features was established. 3D slicer was used to extracted 107 radiomic features from the region of interest (ROI) of 68Ga-dotata PET/CT images. The Pearson correlation coefficient (PCC), recursive feature elimination (REF) based five-fold cross validation were adopted for the radiomic feature selection, and a radiomic score was computed subsequently. The comprehensive model combining the clinical risk factors and the rad-score was established as well as the nomogram. The performance of above clinical model and comprehensive model were evaluated and compared.ResultsAdjacent organ invasion, N staging, and M staging were the risk factors for PNET grading (p < 0.05). 12 optimal radiomic features (3 PET radiomic features, 9 CT radiomic features) were screen out. The clinical predictive model achieved an area under the curve (AUC) of 0.785. The comprehensive model has better predictive performance (AUC = 0.953).ConclusionWe proposed a comprehensive nomogram model based on 68Ga-DOTATATE PET/CT to predict grade 1 and grade 2/3 of PNETs and assist personalized clinical diagnosis and treatment plans for patients with PNETs.

Radiomics features on computed tomography reflect thrombus histological age prior to endovascular treatment of acute ischemic stroke

PurposeTo investigate the role of radiomics features in thrombus age identification and establish a CT-based radiomics model for predicting thrombus age of large vessel occlusion stroke patients. MethodsWe retrospectively reviewed patients with middle cerebral artery occlusion receiving mechanical thrombectomy from July 2020 to March 2022 at our center. The retrieved clots were stained with Hematoxylin and Eosin (H&E) and determined as fresh or older thrombi based on coagulation age. Clot-derived radiomics features were selected by least absolute shrinkage and selection operator (LASSO) regression analysis, by which selected radiomics features were integrated into the Rad-score via the corresponding coefficients. The prediction performance of Rad-score in thrombus age was evaluated with the area under the curve (AUC) of receiver operating characteristic (ROC) curve analysis. ResultsA total of 104 patients were included in our analysis, with 52 in training and 52 in validation cohort. Older thrombi were characterized with delayed procedure time, worse functional outcome and marginally associated with more attempts of device. We extracted 982 features from NCCT images. Following T test and LASSO analysis in training cohort, six radiomics features were selected, based on which the Rad-score was generated by the linear combination of features. The Rad-score showed satisfactory performance in distinguishing fresh with older thrombi, with the AUC of 0.873 (95 %CI: 0.777-0.956) and 0.773 (95 %CI: 0.636-0.910) in training and validation cohort, respectively. ConclusionThis study established and validated a CT-based radiomics model that could accurately differentiate fresh with older thrombi for stroke patients receiving mechanical thrombectomy.

CT-based non-invasive identification of the most common gene mutation status in patients with non-small cell lung cancer.

Epidermal growth factor receptor (EGFR) and Kirsten rat sarcoma viral oncogene homolog (KRAS) are mutually exclusive, and they are two important genes that are most prone to mutation in patients with non-small cell lung cancer. This retrospective study investigated the ability of radiomics to predict the mutation status of EGFR and KRAS in patients with non-small cell lung cancer (NSCLC) and guide precision medicine. Computed tomography images of 1045 NSCLC patients from five different institutions were collected, and 1204 imaging features were extracted. In the training set (EGFR: 678, KRAS: 246), Max-Relevance and Min-Redundancy and least absolute shrinkage and selection operator logistic regression were used to screen radiomics features. The combination of selected radiomics features and clinical factors was used to establish the combined models in identifying EGFR and KRAS mutation status, respectively, through stepwise logistic regression. Then, on two independent external validation sets (EGFR: 203/164, KRAS: 123/95), the performance of each model was evaluated separately, and then the overall performance of predicting the two mutation states was calculated. In the EGFR and KRAS groups, radiomics signatures comprised 14 and 10 radiomics features, respectively. They were mutually exclusive between the tumors with positive EGFR mutation and those with positive KRAS mutation in imaging phenotype. For the EGFR group, the area under the curve (AUC) of the combined model in the two validation sets was 0.871 (95% CI: 0.821-0.926) and 0.861 (95% CI: 0.802-0.911), respectively, whereas the AUC of the combined model in the two validation sets was 0.798 (95% CI: 0.739-0.850) and 0.778 (95% CI: 0.735-0.821), respectively, for the KRAS group. Considering both EGFR and KRAS, the overall precision, recall, and F1-score of the combined model in the two validation sets were 0.704, 0.844, and 0.768, as well as 0.754, 0.693, and 0.722, respectively. Our study demonstrates the potential of radiomics in the non-invasive identification of EGFR and KRAS mutation status, which may guide patients with non-small cell lung cancer to choose the most appropriate personalized treatment. This method can be used when biopsy will bring unacceptable risk to patients with NSCLC.

CT radiomics model for predicting the Ki-67 proliferation index of pure-solid non-small cell lung cancer: a multicenter study.

This study aimed to explore the efficacy of the computed tomography (CT) radiomics model for predicting the Ki-67 proliferation index (PI) of pure-solid non-small cell lung cancer (NSCLC). This retrospective study included pure-solid NSCLC patients from five centers. The radiomics features were extracted from thin-slice, non-enhanced CT images of the chest. The minimum redundancy maximum relevance (mRMR) and least absolute shrinkage and selection operator (LASSO) were used to reduce and select radiomics features. Logistic regression analysis was employed to build predictive models to determine Ki-67-high and Ki-67-low expression levels. Three prediction models were established: the clinical model, the radiomics model, and the nomogram model combining the radiomics signature and clinical features. The prediction efficiency of different models was evaluated using the area under the curve (AUC). A total of 211 NSCLC patients with pure-solid nodules or masses were included in the study (N=117 for the training cohort, N=49 for the internal validation cohort, and N=45 for the external validation cohort). The AUC values for the clinical models in the training, internal validation, and external validation cohorts were 0.73 (95% CI: 0.64-0.82), 0.75 (95% CI:0.62-0.89), and 0.72 (95% CI: 0.57-0.86), respectively. The radiomics models showed good predictive ability in diagnosing Ki-67 expression levels in the training cohort (AUC, 0.81 [95% CI: 0.73-0.89]), internal validation cohort (AUC, 0.81 [95% CI: 0.69-0.93]) and external validation cohort (AUC, 0.78 [95% CI: 0.64-0.91]). Compared to the clinical and radiomics models, the nomogram combining both radiomics signatures and clinical features had relatively better diagnostic performance in all three cohorts, with the AUC of 0.83 (95% CI: 0.76-0.90), 0.83 (95% CI: 0.71-0.94), and 0.81 (95% CI: 0.68-0.93), respectively. The nomogram combining the radiomics signature and clinical features may be a potential non-invasive method for predicting Ki-67 expression levels in patients with pure-solid NSCLC.

Performance of Radiomics Based Multi-Parameter Models Derived from Lung CT in Differentiating Benign Non-Inflammatory, Inflammatory and Malignant Pulmonary Nodule

Pulmonary nodules are a significant clinical issue that require accurate and efficient diagnosis. This study constructed a machine learning model, combining radiomics features of chest CT sequences with three types of microvascular density (MVD) values, to differentiate benign, inflammatory and malignant pulmonary nodules. A total of 100 patients with lung nodules on CT images and corresponding pathological results were retrospectively included in the study. The MVD values and radiomics features were calculated and extracted based on the segmented nodules. Univariate correlation analysis and principal component analysis were performed to select radiomics features. Combined MVD values and selected radiomics features, we conducted a logistic regression classification model. The area under the curve (AUC) was applied to show model performance. Our model reached an AUC value of 0.867 when tested on independent datasets. The performance of the model to differentiate benign, malignant and inflammatory nodules reached AUC values of only 0.908, 0.833, and 0.730, respectively. We conducted a prediction model that shows promising results in distinguishing three different types of lung nodules.

A Clinic-Radiomics Model for Predicting the Incidence of Persistent Organ Failure in Patients with Acute Necrotizing Pancreatitis.

Persistent organ failure (POF) is the leading cause of death in patients with acute necrotizing pancreatitis (ANP). Although several risk factors have been identified, there remains a lack of efficient instruments to accurately predict the incidence of POF in ANP. Retrospectively, the clinical and imaging data of 178 patients with ANP were collected from our database, and the patients were divided into training (n = 125) and validation (n = 53) cohorts. Through computed tomography image acquisition, the volume of interest segmentation, and feature extraction and selection, a pure radiomics model in terms of POF prediction was established. Then, a clinic-radiomics model integrating the pure radiomics model and clinical risk factors was constructed. Both primary and secondary endpoints were compared between the high- and low-risk groups stratified by the clinic-radiomics model. According to the 547 selected radiomics features, four models were derived from features. A clinic-radiomics model in the training and validation sets showed better predictive performance than pure radiomics and clinical models. The clinic-radiomics model was evaluated by the ratios of intervention and mechanical ventilation, intensive care unit (ICU) stays, and hospital stays. The results showed that the high-risk group had significantly higher intervention rates, ICU stays, and hospital stays than the low-risk group, with the confidence interval of 90% (p < 0.1 for all). This clinic-radiomics model is a useful instrument for clinicians to evaluate the incidence of POF, facilitating patients' and their families' understanding of the ANP prognosis.

Towards reproducible radiomics research: introduction of a database for radiomics studies

ObjectivesTo investigate the model-, code-, and data-sharing practices in the current radiomics research landscape and to introduce a radiomics research database.MethodsA total of 1254 articles published between January 1, 2021, and December 31, 2022, in leading radiology journals (European Radiology, European Journal of Radiology, Radiology, Radiology: Artificial Intelligence, Radiology: Cardiothoracic Imaging, Radiology: Imaging Cancer) were retrospectively screened, and 257 original research articles were included in this study. The categorical variables were compared using Fisher’s exact tests or chi-square test and numerical variables using Student’s t test with relation to the year of publication.ResultsHalf of the articles (128 of 257) shared the model by either including the final model formula or reporting the coefficients of selected radiomics features. A total of 73 (28%) models were validated on an external independent dataset. Only 16 (6%) articles shared the data or used publicly available open datasets. Similarly, only 20 (7%) of the articles shared the code. A total of 7 (3%) articles both shared code and data. All collected data in this study is presented in a radiomics research database (RadBase) and could be accessed at https://github.com/EuSoMII/RadBase.ConclusionAccording to the results of this study, the majority of published radiomics models were not technically reproducible since they shared neither model nor code and data. There is still room for improvement in carrying out reproducible and open research in the field of radiomics.Clinical relevance statementTo date, the reproducibility of radiomics research and open science practices within the radiomics research community are still very low. Ensuring reproducible radiomics research with model-, code-, and data-sharing practices will facilitate faster clinical translation.Key Points• There is a discrepancy between the number of published radiomics papers and the clinical implementation of these published radiomics models.• The main obstacle to clinical implementation is the lack of model-, code-, and data-sharing practices.• In order to translate radiomics research into clinical practice, the radiomics research community should adopt open science practices.

Radiomics in the characterization of lipid-poor adrenal adenomas at unenhanced CT: time to look beyond usual density metrics

ObjectivesIn this study, we developed a radiomic signature for the classification of benign lipid-poor adenomas, which may potentially help clinicians limit the number of unnecessary investigations in clinical practice. Indeterminate adrenal lesions of benign and malignant nature may exhibit different values of key radiomics features.MethodsPatients who had available histopathology reports and a non-contrast-enhanced CT scan were included in the study. Radiomics feature extraction was done after the adrenal lesions were contoured. The primary feature selection and prediction performance scores were calculated using the least absolute shrinkage and selection operator (LASSO). To eliminate redundancy, the best-performing features were further examined using the Pearson correlation coefficient, and new predictive models were created.ResultsThis investigation covered 50 lesions in 48 patients. After LASSO-based radiomics feature selection, the test dataset’s 30 iterations of logistic regression models produced an average performance of 0.72. The model with the best performance, made up of 13 radiomics features, had an AUC of 0.99 in the training phase and 1.00 in the test phase. The number of features was lowered to 5 after performing Pearson’s correlation to prevent overfitting. The final radiomic signature trained a number of machine learning classifiers, with an average AUC of 0.93.ConclusionsIncluding more radiomics features in the identification of adenomas may improve the accuracy of NECT and reduce the need for additional imaging procedures and clinical workup, according to this and other recent radiomics studies that have clear points of contact with current clinical practice.Clinical relevance statementThe study developed a radiomic signature using unenhanced CT scans for classifying lipid-poor adenomas, potentially reducing unnecessary investigations that scored a final accuracy of 93%.Key Points• Radiomics has potential for differentiating lipid-poor adenomas and avoiding unnecessary further investigations.• Quadratic mean, strength, maximum 3D diameter, volume density, and area density are promising predictors for adenomas.• Radiomics models reach high performance with average AUC of 0.95 in the training phase and 0.72 in the test phase.

The MRI radiomics signature can predict the pathologic response to neoadjuvant chemotherapy in locally advanced esophageal squamous cell carcinoma.

To investigate the MRI radiomics signatures in predicting pathologic response among patients with locally advanced esophageal squamous cell carcinoma (ESCC), who received neoadjuvant chemotherapy (NACT). Patients who underwent NACT from March 2015 to October 2019 were prospectively included. Each patient underwent esophageal MR scanning within one week before NACT and within 2-3weeks after completion of NACT, prior to surgery. Radiomics features extracted from T2-TSE-BLADE were randomly split into the training and validation sets at a ratio of 7:3. According to the progressive tumor regression grade (TRG), patients were stratified into two groups: good responders (GR, TRG 0 + 1) and poor responders (non-GR, TRG 2 + 3). We constructed the Pre/Post-NACT model (Pre/Post-model) and the Delta-NACT model (Delta-model). Kruskal-Wallis was used to select features, logistic regression was used to develop the final model. A total of 108 ESCC patients were included, and 3/2/4 out of 107 radiomics features were selected for constructing the Pre/Post/Delta-model, respectively. The selected radiomics features were statistically different between GR and non-GR groups. The highest areaunder the curve (AUC) was for the Delta-model, which reached 0.851 in the training set and 0.831 in the validation set. Among the three models,Pre-model showed the poorest performance in the training and validation sets (AUC, 0.466 and 0.596), and the Post-model showed better performance than the Pre-model in the training and validation sets (AUC, 0.753 and 0.781). MRI-based radiomics models can predict the pathological response after NACT in ESCC patients, with the Delta-model exhibiting optimal predictive efficacy. MRI radiomics features could be used as a useful tool for predicting the efficacy of neoadjuvant chemotherapy in esophageal carcinoma patients, especially in selecting responders among those patients who may be candidates to benefit from neoadjuvant chemotherapy. • The MRI radiomics features based on T2WI-TSE-BLADE could potentially predict the pathologic response to NACT among ESCC patients. • The Delta-model exhibited the best predictive ability for pathologic response, followed by the Post-model, which similarly had better predictive ability, while the Pre-model performed less well in predicting TRG.

ABVS-Based Radiomics for Early Predicting the Efficacy of Neoadjuvant Chemotherapy in Patients with Breast Cancers.

Neoadjuvant chemotherapy (NAC) plays a significant role in breast cancer (BC) management; however, its efficacy varies among patients. Current evaluation methods may lead to delayed treatment alterations, and traditional imaging modalities often yield inaccurate results. Radiomics, an emerging field in medical imaging, offers potential for improved tumor characterization and personalized medicine. Nevertheless, its application in early and accurately predicting NAC response remains underinvestigated. This study aims to develop an automated breast volume scanner (ABVS)-based radiomics model to facilitate early detection of suboptimal NAC response, ultimately promoting personalized therapeutic approaches for BC patients. This retrospective study involved 248 BC patients receiving NAC. Standard guidelines were followed, and patients were classified as responders or non-responders based on treatment outcomes. ABVS images were obtained before and during NAC, and radiomics features were extracted using the PyRadiomics toolkit. Inter-observer consistency and hierarchical feature selection were assessed. Three machine learning classifiers, logistic regression, support vector machine, and random forest, were trained and validated using a five-fold cross-validation with three repetitions. Model performance was comprehensively evaluated based on discrimination, calibration, and clinical utility. Of the 248 BC patients, 157 (63.3%) were responders, and 91 (36.7%) were non-responders. Radiomics feature selection revealed 7 pre-NAC and 6 post-NAC ABVS features, with higher weights for post-NAC features (min >0.05) than pre-NAC (max <0.03). The three post-NAC classifiers demonstrated AUCs of approximately 0.9, indicating excellent discrimination. DCA curves revealed a substantial net benefit when the threshold probability exceeded 40%. Conversely, the three pre-NAC classifiers had AUCs between 0.7 and 0.8, suggesting moderate discrimination and limited clinical utility based on their DCA curves. The ABVS-based radiomics model effectively predicted suboptimal NAC responses in BC patients, with early post-NAC classifiers outperforming pre-NAC classifiers in discrimination and clinical utility. It could enhance personalized treatment and improve patient outcomes in BC management.

Development and validation of a radiomics-based prediction pipeline for the response to stereotactic radiosurgery therapy in brain metastases.

The first treatment strategy for brain metastases (BM) plays a pivotal role in the prognosis of patients. Among all strategies, stereotactic radiosurgery (SRS) is considered a promising therapy method. Therefore, we developed and validated a radiomics-based prediction pipeline to prospectively identify BM patients who are insensitive to SRS therapy, especially those who are at potential risk of progressive disease. A total of 337 BM patients (277, 30, and 30 in the training set, internal validation set, and external validation set, respectively) were enrolled in the study. 19,377 radiomics features (3 masks × 3 MRI sequences × 2153 features) extracted from 9 ROIs were filtered through LASSO and Max-Relevance and Min-Redundancy (mRMR) algorithms. The selected radiomics features were combined with 4 clinical features to construct a two-stage cascaded model for the prediction of BM patients' response to SRS therapy using SVM and an ensemble learning classifier. The performance of the model was evaluated by its accuracy, specificity, sensitivity, and AUC curve. Radiomics features were integrated with the clinical features of patients in our optimal model, which showed excellent discriminative performance in the training set (AUC: 0.95, 95% CI: 0.88-0.98). The model was also verified in the internal validation set and external validation set (AUC 0.93, 95% CI: 0.76-0.95 and AUC 0.90, 95% CI: 0.73-0.93, respectively). The proposed prediction pipeline could non-invasively predict the response to SRS therapy in patients with brain metastases thus assisting doctors to precisely designate individualized first treatment decisions. The proposed prediction pipeline combines the radiomics features of multi-modal MRI with clinical features to construct machine learning models that noninvasively predict the response of patients with brain metastases to stereotactic radiosurgery therapy, assisting neuro-oncologists to develop personalized first treatment plans. • The proposed prediction pipeline can non-invasively predict the response to SRS therapy. • The combination of multi-modality and multi-mask contributes significantly to the prediction. • The edema index also shows a certain predictive value.

Stacking Ensemble Learning-Based [18F]FDG PET Radiomics for Outcome Prediction in Diffuse Large B-Cell Lymphoma.

This study aimed to develop an analytic approach based on [18F]FDG PET radiomics using stacking ensemble learning to improve the outcome prediction in diffuse large B-cell lymphoma (DLBCL). Methods: In total, 240 DLBCL patients from 2 medical centers were divided into the training set (n = 141), internal testing set (n = 61), and external testing set (n = 38). Radiomics features were extracted from pretreatment [18F]FDG PET scans at the patient level using 4 semiautomatic segmentation methods (SUV threshold of 2.5, SUV threshold of 4.0 [SUV4.0], 41% of SUVmax, and SUV threshold of mean liver uptake [PERCIST]). All extracted features were harmonized with the ComBat method. The intraclass correlation coefficient was used to evaluate the reliability of radiomics features extracted by different segmentation methods. Features from the most reliable segmentation method were selected by Pearson correlation coefficient analysis and the LASSO (least absolute shrinkage and selection operator) algorithm. A stacking ensemble learning approach was applied to build radiomics-only and combined clinical-radiomics models for prediction of 2-y progression-free survival and overall survival based on 4 machine learning classifiers (support vector machine, random forests, gradient boosting decision tree, and adaptive boosting). Confusion matrix, receiver-operating-characteristic curve analysis, and survival analysis were used to evaluate the model performance. Results: Among 4 semiautomatic segmentation methods, SUV4.0 segmentation yielded the highest interobserver reliability, with 830 (66.7%) selected radiomics features. The combined model constructed by the stacking method achieved the best discrimination performance. For progression-free survival prediction in the external testing set, the areas under the receiver-operating-characteristic curve and accuracy of the stacking-based combined model were 0.771 and 0.789, respectively. For overall survival prediction, the stacking-based combined model achieved an area under the curve of 0.725 and an accuracy of 0.763 in the external testing set. The combined model also demonstrated a more distinct risk stratification than the International Prognostic Index in all sets (log-rank test, all P < 0.05). Conclusion: The combined model that incorporates [18F]FDG PET radiomics and clinical characteristics based on stacking ensemble learning could enable improved risk stratification in DLBCL.

Multiparametric MRI radiomics in prostate cancer for predicting Ki-67 expression and Gleason score: a multicenter retrospective study

PurposeProstate cancer (PCa) with high Ki-67 expression and high Gleason Scores (GS) tends to have aggressive clinicopathological characteristics and a dismal prognosis. In order to predict the Ki-67 expression status and the GS in PCa, we sought to construct and verify MRI-based radiomics signatures.Methods and materialsWe collected T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), and apparent diffusion coefficient (ADC) images from 170 PCa patients at three institutions and extracted 321 original radiomic features from each image modality. We used support vector machine (SVM) and least absolute shrinkage and selection operator (LASSO) logistic regression to select the most informative radiomic features and built predictive models using up sampling and feature selection techniques. Using receiver operating characteristic (ROC) analysis, the discriminating power of this feature was determined. Subsequent decision curve analysis (DCA) assessed the clinical utility of the radiomic features. The Kaplan–Meier (KM) test revealed that the radiomics-predicted Ki-67 expression status and GS were prognostic factors for PCa survival.ResultThe hypothesized radiomics signature, which included 15 and 9 selected radiomics features, respectively, was significantly correlated with pathological Ki-67 and GS outcomes in both the training and validation datasets. Areas under the curve (AUC) for the developed model were 0.813 (95% CI 0.681,0.930) and 0.793 (95% CI 0.621, 0.929) for the training and validation datasets, respectively, demonstrating discrimination and calibration performance. The model's clinical usefulness was verified using DCA. In both the training and validation sets, high Ki-67 expression and high GS predicted by radiomics using SVM models were substantially linked with poor overall survival (OS).ConclusionsBoth Ki-67 expression status and high GS correlate with PCa patient survival outcomes; therefore, the ability of the SVM classifier-based model to estimate Ki-67 expression status and the Lasso classifier-based model to assess high GS may enhance clinical decision-making.

Deep learning radiomics of ultrasonography for differentiating sclerosing adenosis from breast cancer.

The purpose of our study is to present a method combining radiomics with deep learning and clinical data for improved differential diagnosis of sclerosing adenosis (SA)and breast cancer (BC). A total of 97 patients with SA and 100 patients with BC were included in this study. The best model for classification was selected from among four different convolutional neural network (CNN) models, including Vgg16, Resnet18, Resnet50, and Desenet121. The intra-/inter-class correlation coefficient and least absolute shrinkage and selection operator method were used for radiomics feature selection. The clinical features selected were patient age and nodule size. The overall accuracy, sensitivity, specificity, Youden index, positive predictive value, negative predictive value, and area under curve (AUC) value were calculated for comparison of diagnostic efficacy. All the CNN models combined with radiomics and clinical data were significantly superior to CNN models only. The Desenet121+radiomics+clinical data model showed the best classification performance with an accuracy of 86.80%, sensitivity of 87.60%, specificity of 86.20% and AUC of 0.915, which was better than that of the CNN model only, which had an accuracy of 85.23%, sensitivity of 85.48%, specificity of 85.02%, and AUC of 0.870. In comparison, the diagnostic accuracy, sensitivity, specificity, and AUC value for breast radiologists were 72.08%, 100%, 43.30%, and 0.716, respectively. A combination of the CNN-radiomics model and clinical data could be a helpful auxiliary diagnostic tool for distinguishing between SA and BC.

An imaging-based machine learning model outperforms clinical risk scores for prognosis of cirrhotic variceal bleeding.

To develop and validate a machine learning model based on contrast-enhanced CT to predict the risk of occurrence of the composite clinical endpoint (hospital-based intervention or death) in cirrhotic patients with acute variceal bleeding (AVB). This retrospective study enrolled 330 cirrhotic patients with AVB between January 2017 and December 2020 from three clinical centers. Contrast-enhanced CT and clinical data were collected. Centers A and B were divided 7:3 into a training set and an internal test set, and center C served as a separate external test set. A well-trained deep learning model was applied to segment the liver and spleen. Then, we extracted 106 original features of the liver and spleen separately based on the Image Biomarker Standardization Initiative (IBSI). We constructed the Liver-Spleen (LS) model based on the selected radiomics features. The performance of LS model was evaluated by receiver operating characteristics and calibration curves. The clinical utility of models was analyzed using decision curve analyses (DCA). The LS model demonstrated the best diagnostic performance in predicting the composite clinical endpoint of AVB in patients with cirrhosis, with an AUC of 0.782 (95% CI 0.650-0.882) and 0.789 (95% CI 0.674-0.878) in the internal test and external test groups, respectively. Calibration curves and DCA indicated the LS model had better performance than traditional clinical scores. A novel machine learning model outperforms previously known clinical risk scores in assessing the prognosis of cirrhotic patients with AVB CLINICAL RELEVANCE STATEMENT: The Liver-Spleen model based on contrast-enhanced CT has proven to be a promising tool to predict the prognosis of cirrhotic patients with acute variceal bleeding, which can facilitate decision-making and personalized therapy in clinical practice. • The Liver-Spleen machine learning model (LS model) showed good performance in assessing the clinical composite endpoint of cirrhotic patients with AVB (AUC ≥ 0.782, sensitivity ≥ 80%). • The LS model outperformed the clinical scores (AUC ≤ 0.730, sensitivity ≤ 70%) in both internal and external test cohorts.

A radiomics nomogram based on contrast-enhanced CT for preoperative prediction of Lymphovascular invasion in esophageal squamous cell carcinoma.

To develop a radiomics nomogram based on contrast-enhanced computed tomography (CECT) for preoperative prediction of lymphovascular invasion (LVI) status of esophageal squamous cell carcinoma (ESCC). The clinical and imaging data of 258 patients with ESCC who underwent surgical resection and were confirmed by pathology from June 2017 to December 2021 were retrospectively analyzed.The clinical imaging features and radiomic features were extracted from arterial-phase CECT. The least absolute shrinkage and selection operator (LASSO) regression model was used for radiomics feature selection and signature construction. Multivariate logistic regression analysis was used to develop a radiomics nomogram prediction model. The receiver operating characteristic (ROC) curve and decision curve analysis (DCA) were used to evaluate the performance and clinical effectiveness of the model in preoperative prediction of LVI status. We constructed a radiomics signature based on eight radiomics features after dimensionality reduction. In the training cohort, the area under the curve (AUC) of radiomics signature was 0.805 (95% CI: 0.740-0.860), and in the validation cohort it was 0.836 (95% CI: 0.735-0.911). There were four predictive factors that made up the individualized nomogram prediction model: radiomic signatures, TNRs, tumor lengths, and tumor thicknesses.The accuracy of the nomogram for LVI prediction in the training and validation cohorts was 0.790 and 0.768, respectively, the specificity was 0.800 and 0.618, and the sensitivity was 0.786 and 0.917, respectively. The Delong test results showed that the AUC value of the nomogram model was significantly higher than that of the clinical model and radiomics model in the training and validation cohort(P<0.05). DCA results showed that the radiomics nomogram model had higher overall benefits than the clinical model and the radiomics model. This study proposes a radiomics nomogram based on CECT radiomics signature and clinical image features, which is helpful for preoperative individualized prediction of LVI status in ESCC.