Magnetic Resonance Imaging Radiomic Features Research Articles

Magnetic Resonance-Guided Cancer Therapy Radiomics and Machine Learning Models for Response Prediction.

Magnetic resonance imaging (MRI) is known for its accurate soft tissue delineation of tumors and normal tissues. This development has significantly impacted the imaging and treatment of cancers. Radiomics is the process of extracting high-dimensional features from medical images. Several studies have shown that these extracted features may be used to build machine-learning models for the prediction of treatment outcomes of cancer patients. Various feature selection techniques and machine models interrogate the relevant radiomics features for predicting cancer treatment outcomes. This study aims to provide an overview of MRI radiomics features used in predicting clinical treatment outcomes with machine learning techniques. The review includes examples from different disease sites. It will also discuss the impact of magnetic field strength, sample size, and other characteristics on outcome prediction performance.

Prediction of bone invasion of oral squamous cell carcinoma using a magnetic resonance imaging-based machine learning model.

Radiomics, a recently developed image-processing technology, holds potential in medical diagnostics. This study aimed to propose a machine-learning (ML) model and evaluate its effectiveness in detecting oral squamous cell carcinoma (OSCC) and predicting bone metastasis using magnetic resonance imaging (MRI). MRI radiomic features were extracted and analyzed to identify malignant lesions. A total of 86 patients (44 with benign lesions without bone invasion and 42 with malignant lesions with bone invasion) were included. Data and clinical information were managed using the RadCloud Platform (Huiying Medical Technology Co., Ltd., Beijing, China). The study employed a hand-crafted radiomics model, with the dataset randomly split into training and validation sets in an 8:2 ratio using 815 random seeds. The results revealed that the ML method support vector machine (SVM) performed best for detecting bone invasion (AUC = 0.999) in the test set. Radiomics tumor features derived from MRI are useful to predicting bone invasion from oral squamous cell carcinoma with high accuracy. This study introduces an ML model utilizing SVM and radiomics to predict bone invasion in OSCC. Despite the promising results, the small sample size necessitates larger multicenter studies to validate and expand these findings.

Magnetic resonance imaging radiomics-based prediction of severe inflammatory response in locally advanced rectal cancer patients after neoadjuvant radiochemotherapy.

To predict severe inflammatory response after neoadjuvant radiochemotherapy in locally advanced rectal cancer (RC) patients using magnetic resonance imaging (MRI) radiomics models. This retrospective study included patients who underwent radical surgery for RC cancer after neoadjuvant radiochemotherapy between July 2017 and December 2019 at XXX Hospital. MRI radiomics features were extracted from T2WI images before (pre-nRCT-RF) and after (post-nRCT-RF) neoadjuvant radiochemotherapy, and the variation of radiomics features before and after neoadjuvant radiochemotherapy (delta-RF) were calculated. Eight, eight, and five most relevant features were identified for pre-nRCT-RF, post-nRCT-RF, and delta-RF, respectively. Eighty-six patients were included and randomized 3:1 to the training and test set (n = 65 and n = 21, respectively). The prediction model based on delta-RF had areas under the curve (AUCs) of 0.80 and 0.85 in the training and test set, respectively. A higher rate of difficult operations was observed in patients with severe inflammation (65.5% vs. 42.9%, P = 0.045). The prediction model based on MRI delta-RF may be a useful tool for predicting severe inflammatory response after neoadjuvant radiochemotherapy in locally advanced RC patients.

CT and MRI radiomic features of lung cancer (NSCLC): comparison and software consistency

BackgroundRadiomics is a quantitative approach that allows the extraction of mineable data from medical images. Despite the growing clinical interest, radiomics studies are affected by variability stemming from analysis choices. We aimed to investigate the agreement between two open-source radiomics software for both contrast-enhanced computed tomography (CT) and contrast-enhanced magnetic resonance imaging (MRI) of lung cancers and to preliminarily evaluate the existence of radiomic features stable for both techniques.MethodsContrast-enhanced CT and MRI images of 35 patients affected with non-small cell lung cancer (NSCLC) were manually segmented and preprocessed using three different methods. Sixty-six Image Biomarker Standardisation Initiative-compliant features common to the considered platforms, PyRadiomics and LIFEx, were extracted. The correlation among features with the same mathematical definition was analyzed by comparing PyRadiomics and LIFEx (at fixed imaging technique), and MRI with CT results (for the same software).ResultsWhen assessing the agreement between LIFEx and PyRadiomics across the considered resampling, the maximum statistically significant correlations were observed to be 94% for CT features and 95% for MRI ones. When examining the correlation between features extracted from contrast-enhanced CT and MRI using the same software, higher significant correspondences were identified in 11% of features for both software.ConclusionsConsidering NSCLC, (i) for both imaging techniques, LIFEx and PyRadiomics agreed on average for 90% of features, with MRI being more affected by resampling and (ii) CT and MRI contained mostly non-redundant information, but there are shape features and, more importantly, texture features that can be singled out by both techniques.Relevance statementIdentifying and selecting features that are stable cross-modalities may be one of the strategies to pave the way for radiomics clinical translation.Key points• More than 90% of LIFEx and PyRadiomics features contain the same information.• Ten percent of features (shape, texture) are stable among contrast-enhanced CT and MRI.• Software compliance and cross-modalities stability features are impacted by the resampling method.Graphical

MRI radiomics nomogram integrating postoperative adjuvant treatments in recurrence risk prediction for patients with early-stage cervical cancer

Background and purposeAdjuvant treatments are valuable to decrease the recurrence rate and improve survival for early-stage cervical cancer patients (ESCC), Therefore, recurrence risk evaluation is critical for the choice of postoperative treatment. A magnetic resonance imaging (MRI) based radiomics nomogram integrating postoperative adjuvant treatments was constructed and validated externally to improve the recurrence risk prediction for ESCC. Material and methods212 ESCC patients underwent surgery and adjuvant treatments from three centers were enrolled and divided into the training, internal validation, and external validation cohorts. Their clinical data, pretreatment T2-weighted images (T2WI) were retrieved and analyzed. Radiomics models were constructed using machine learning methods with features extracted and screen from sagittal and axial T2WI. A nomogram for recurrence prediction was build and evaluated using multivariable logistic regression analysis integrating radiomic signature and adjuvant treatments. ResultsA total of 8 radiomic features were screened out of 1020 extracted features. The extreme gradient boosting (XGboost) model based on MRI radiomic features performed best in recurrence prediction with an area under curve (AUC) of 0.833, 0.822 in the internal and external validation cohorts, respectively. The nomogram integrating radiomic signature and clinical factors achieved an AUC of 0.806, 0.718 in the internal and external validation cohorts, respectively, for recurrence risk prediction for ESCC. ConclusionIn this study, the nomogram integrating T2WI radiomic signature and clinical factors is valuable to predict the recurrence risk, thereby allowing timely planning for effective treatments for ESCC with high risk of recurrence.

Value of magnetic resonance imaging radiomics features in predicting histologic grade of invasive ductal carcinoma of the breast.

Breast cancer has the second highest mortality rate of all cancers and occurs mainly in women. To investigate the relationship between magnetic resonance imaging (MRI) radiomics features and histological grade of invasive ductal carcinoma (IDC) of the breast and to evaluate its diagnostic efficacy. The two conventional MRI quantitative indicators, i.e. the apparent diffusion coefficient (ADC) and the initial enhancement rate, were collected from 112 patients with breast cancer. The breast cancer lesions were manually segmented in dynamic contrast-enhanced MRI (DCE-MRI) and ADC images, the differences in radiomics features between Grades I, II and III IDCs were compared and the diagnostic efficacy was evaluated. The ADC values (0.77 ± 0.22 vs 0.91 ± 0.22 vs 0.92 ± 0.20, F= 4.204, p< 0.01), as well as the B_sum_variance (188.51 ± 67.803 vs 265.37 ± 77.86 vs 263.74 ± 82.58, F= 6.040, p< 0.01), L_energy (0.03 ± 0.02 vs 0.13 ± 0.11 vs 0.12 ± 0.14, F= 7.118, p< 0.01) and L_sum_average (0.78 ± 0.32 vs 16.34 ± 4.23 vs 015.45 ± 3.74, F= 21.860, p< 0.001) values of patients with Grade III IDC were significantly lower than those of patients with Grades I and II IDC. The B_uniform (0.15 ± 0.12 vs 0.11 ± 0.04 vs 0.12 ± 0.03, F= 3.797, p< 0.01) and L_SRE (0.85 ± 0.07 vs 0.78 ± 0.03 vs 0.79 ± 0.32, F= 3.024, p< 0.01) values of patients with Grade III IDC were significantly higher than those of patients with Grades I and II IDC. All differences were statistically significant (p< 0.05). The ADC radiomics signature model had a higher area-under-the-curve value in identifying different grades of IDC than the ADC value model and the DCE radiomics signature model (0.869 vs 0.711 vs 0.682). The accuracy (0.812 vs 0.647 vs 0.710), specificity (0.731 vs 0.435 vs 0.342), positive predictive value (0.815 vs 0.663 vs 0.669) and negative predictive value (0.753 vs 0.570 vs 0.718) of the ADC radiomics signature model were all significantly better than the ADC value model and the DCE radiomics signature model. ADC values and breast MRI radiomics signatures are significant in identifying the histological grades of IDC, with the ADC radiomics signatures having greater value.

Preoperative prediction of perineural invasion of rectal cancer based on a magnetic resonance imaging radiomics model: A dual-center study.

Perineural invasion (PNI) has been used as an important pathological indicator and independent prognostic factor for patients with rectal cancer (RC). Preoperative prediction of PNI status is helpful for individualized treatment of RC. Recently, several radiomics studies have been used to predict the PNI status in RC, demonstrating a good predictive effect, but the results lacked generalizability. The preoperative prediction of PNI status is still challenging and needs further study. To establish and validate an optimal radiomics model for predicting PNI status preoperatively in RC patients. This retrospective study enrolled 244 postoperative patients with pathologically confirmed RC from two independent centers. The patients underwent pre-operative high-resolution magnetic resonance imaging (MRI) between May 2019 and August 2022. Quantitative radiomics features were extracted and selected from oblique axial T2-weighted imaging (T2WI) and contrast-enhanced T1WI (T1CE) sequences. The radiomics signatures were constructed using logistic regression analysis and the predictive potential of various sequences was compared (T2WI, T1CE and T2WI + T1CE fusion sequences). A clinical-radiomics (CR) model was established by combining the radiomics features and clinical risk factors. The internal and external validation groups were used to validate the proposed models. The area under the receiver operating characteristic curve (AUC), DeLong test, net reclassification improvement (NRI), integrated discrimination improvement (IDI), calibration curve, and decision curve analysis (DCA) were used to evaluate the model performance. Among the radiomics models, the T2WI + T1CE fusion sequences model showed the best predictive performance, in the training and internal validation groups, the AUCs of the fusion sequence model were 0.839 [95% confidence interval (CI): 0.757-0.921] and 0.787 (95%CI: 0.650-0.923), which were higher than those of the T2WI and T1CE sequence models. The CR model constructed by combining clinical risk factors had the best predictive performance. In the training and internal and external validation groups, the AUCs of the CR model were 0.889 (95%CI: 0.824-0.954), 0.889 (95%CI: 0.803-0.976) and 0.894 (95%CI: 0.814-0.974). Delong test, NRI, and IDI showed that the CR model had significant differences from other models (P < 0.05). Calibration curves demonstrated good agreement, and DCA revealed significant benefits of the CR model. The CR model based on preoperative MRI radiomics features and clinical risk factors can preoperatively predict the PNI status of RC noninvasively, which facilitates individualized treatment of RC patients.

Magnetic resonance imaging radiomic features stability in brain metastases: Impact of image preprocessing, image-, and feature-level harmonization

Background and purposeMagnetic resonance imaging (MRI) scans are highly sensitive to acquisition and reconstruction parameters which affect feature stability and model generalizability in radiomic research. This work aims to investigate the effect of image pre-processing and harmonization methods on the stability of brain MRI radiomic features and the prediction performance of radiomic models in patients with brain metastases (BMs). Materials and methodsTwo T1 contrast enhanced brain MRI data-sets were used in this study. The first contained 25 BMs patients with scans at two different time points and was used for features stability analysis. The effect of gray level discretization (GLD), intensity normalization (Z-score, Nyul, WhiteStripe, and in house-developed method named N-Peaks), and ComBat harmonization on features stability was investigated and features with intraclass correlation coefficient >0.8 were considered as stable. The second data-set containing 64 BMs patients was used for a classification task to investigate the informativeness of stable features and the effects of harmonization methods on radiomic model performance. ResultsApplying fixed bin number (FBN) GLD, resulted in higher number of stable features compare to fixed bin size (FBS) discretization (10 ± 5.5 % higher). `Harmonization in feature domain improved the stability for non-normalized and normalized images with Z-score and WhiteStripe methods. For the classification task, keeping the stable features resulted in good performance only for normalized images with N-Peaks along with FBS discretization. ConclusionsTo develop a robust MRI based radiomic model we recommend using an intensity normalization method based on a reference tissue (e.g N-Peaks) and then using FBS discretization.

Automated graded prognostic assessment for patients with hepatocellular carcinoma using machine learning

BackgroundAccurate mortality risk quantification is crucial for the management of hepatocellular carcinoma (HCC); however, most scoring systems are subjective.PurposeTo develop and independently validate a machine learning mortality risk quantification method for HCC patients using standard-of-care clinical data and liver radiomics on baseline magnetic resonance imaging (MRI).MethodsThis retrospective study included all patients with multiphasic contrast-enhanced MRI at the time of diagnosis treated at our institution. Patients were censored at their last date of follow-up, end-of-observation, or liver transplantation date. The data were randomly sampled into independent cohorts, with 85% for development and 15% for independent validation. An automated liver segmentation framework was adopted for radiomic feature extraction. A random survival forest combined clinical and radiomic variables to predict overall survival (OS), and performance was evaluated using Harrell’s C-index.ResultsA total of 555 treatment-naïve HCC patients (mean age, 63.8 years ± 8.9 [standard deviation]; 118 females) with MRI at the time of diagnosis were included, of which 287 (51.7%) died after a median time of 14.40 (interquartile range, 22.23) months, and had median followed up of 32.47 (interquartile range, 61.5) months. The developed risk prediction framework required 1.11 min on average and yielded C-indices of 0.8503 and 0.8234 in the development and independent validation cohorts, respectively, outperforming conventional clinical staging systems. Predicted risk scores were significantly associated with OS (p < .00001 in both cohorts).ConclusionsMachine learning reliably, rapidly, and reproducibly predicts mortality risk in patients with hepatocellular carcinoma from data routinely acquired in clinical practice.Clinical relevance statementPrecision mortality risk prediction using routinely available standard-of-care clinical data and automated MRI radiomic features could enable personalized follow-up strategies, guide management decisions, and improve clinical workflow efficiency in tumor boards.Key Points• Machine learning enables hepatocellular carcinoma mortality risk prediction using standard-of-care clinical data and automated radiomic features from multiphasic contrast-enhanced MRI.• Automated mortality risk prediction achieved state-of-the-art performances for mortality risk quantification and outperformed conventional clinical staging systems.• Patients were stratified into low, intermediate, and high-risk groups with significantly different survival times, generalizable to an independent evaluation cohort.

Distinguishing EGFR mutation molecular subtypes based on MRI radiomics features of lung adenocarcinoma brain metastases

ObjectiveTo explore the feasibility of identifying epidermal growth factor receptor (EGFR) mutation molecular subtypes in primary lesions based on the radiomics features of lung adenocarcinoma brain metastases using magnetic resonance imaging (MRI). MethodsWe retrospectively analyzed clinical, imaging, and genetic testing data of patients with lung adenocarcinoma with EGFR gene mutations who had brain metastases. Three-dimensional radiomics features were extracted from contrast-enhanced T1-weighted images. The volume of interest was delineated and normalized using Z-score, dimensionality reduction was performed using principal component analysis, feature selection using Relief, and radiomics model construction using adaptive boosting as a classifier. Data were randomly divided into training and testing datasets at an 8:2 ratio. Five-fold cross-validation was conducted in the training set to select the optimal radiomics features and establish a predictive model for distinguishing between exon 19 deletion (19Del) and exon 21 L858R point mutation (21L858R), the two most common EGFR gene mutations. The testing set was used for external validation of the models. Model performance was evaluated using receiver operating characteristic curve and decision curve analyses. ResultsOverall, 86 patients with 228 brain metastases were included. Patient age was identified as an independent predictor for distinguishing between 19Del and 21L858R. The area under the curve (AUC) values of the radiomics model in the training and testing datasets were 0.895 (95% confidence interval [CI]: 0.850−0.939) and 0.759 (95% CI: 0.0.614−0.903), respectively. The AUC for diagnosis of all cases using a combined model of age and radiomics was 0.888 (95% CI: 0.846−0.930), slightly higher than that of the radiomics model alone (0.866, 95% CI: 0.820−0.913), but without statistical significance (p=0.1626). In the decision curve analysis, both models demonstrated clinical net benefits. ConclusionsThe radiomics model based on MRI of lung adenocarcinoma brain metastases could distinguish between EGFR 19Del and 21L858R mutations in the primary lesion.

Prediction of Lymphovascular invasion status in breast cancer based on magnetic resonance imaging radiomics features

ObjectiveThis study intended to investigate the feasibility and effectiveness of using clinical magnetic resonance imaging (MRI) radiomics features to predict lymphovascular invasion (LVI) status in breast cancer (BC) patients. MethodsA total of 182 BC patients were retrospectively collected and randomly divided into a training set (n = 127) and a validation set (n = 55) in a 7:3 ratio. Based on pathological examination results, the training set was further divided into LVI group (n = 60) and non-LVI group (n = 67), and the validation set was divided into LVI group (n = 24) and non-LVI group (n = 31). General data and MRI examination indicators were compared. Multivariate logistic regression was utilized to analyze MRI radiomics features and clinically relevant indicators that were significant in the baseline information of patients in training set, independent risk factors were identified, and a logistic regression model was built. The accuracy of logistic model was validated using ROC curves in training and validation sets. ResultsAge, pathohistological classification, tumor length, tumor width, presence or absence of Magnetic Resonance Spectroscopy (MRS) cho peak, presence or absence of spicule sign, peritumoral enhancement, and peritumoral edema were statistically significant (P < 0.05) between the two groups. Multivariate logistic regression analysis presented that spicule and peritumoral edema were independent risk factors for LVI in BC patients (P < 0.05). The ROC curve illustrated that AUC of the logistic regression model in the training set was 0.807 (95%CI: 0.730–0.885) and that in the validation set was 0.837 (95%CI: 0.731–0.944). ConclusionRadiomics features of spicule sign and peritumoral edema were independent risk factors for LVI in BC patients. A logistic regression model based on these factors, along with age, could accurately predict LVI occurrence in BC patients, providing data support for diagnosis and modeling of LVI in BC patients.

Advanced magnetic resonance imaging for glioblastoma: Oncology-radiology integration.

Aggressive brain tumors like glioblastoma multiforme (GBM) pose a poor prognosis. While magnetic resonance imaging (MRI) is crucial for GBM management, distinguishing it from other lesions using conventional methods can be difficult. This study explores advanced MRI techniques better to understand GBM properties and their link to patient outcomes. We studied MRI scans of 157 GBM surgery patients from January 2020 to March 2024 to extract radiomic features and analyze the impact of fluid-attenuated inversion recovery (FLAIR) resection on survival using statistical methods, proportional hazards regression, and Kaplan-Meier survival analysis. Predictive models achieved high accuracy (area under the curve of 0.902) for glioma-grade prediction. FLAIR abnormality resection significantly improved survival, while diffusion-weighted image best-depicted tumor infiltration. Glioblastoma infiltration was best seen with advanced MRI compared to metastasis. Glioblastomas showed distinct features, including irregular shape, margins, and enhancement compared to metastases, which were oval or round, with clear edges and even contrast, and extensive peritumoral changes. Advanced radiomic and machine learning analysis of MRI can provide noninvasive glioma grading and characterization of tumor properties with clinical relevance. Combining advanced neuroimaging with histopathology may better integrate oncology and radiology for optimized glioblastoma management. However, further studies are needed to validate these findings with larger datasets and assess additional MRI sequences and radiomic features.

Investigating the Feasibility of Predicting KRAS Status, Tumor Staging, and Extramural Venous Invasion in Colorectal Cancer Using Inter-Platform Magnetic Resonance Imaging Radiomic Features

(1) Background: Colorectal cancer is the third most common type of cancer with a high mortality rate and poor prognosis. The accurate prediction of key genetic mutations, such as the KRAS status, tumor staging, and extramural venous invasion (EMVI), is crucial for guiding personalized treatment decisions and improving patients’ outcomes. MRI radiomics was assessed to predict the KRAS status and tumor staging in colorectal cancer patients across different imaging platforms to improve the personalized treatment decisions and outcomes. (2) Methods: Sixty colorectal cancer patients (35M/25F; avg. age 56.3 ± 12.9 years) were treated at an oncology unit. The MRI scans included T2-weighted (T2W) and diffusion-weighted imaging (DWI) or the apparent diffusion coefficient (ADC). The manual segmentation of colorectal cancer was conducted on the T2W and DWI/ADC images. The cohort was split into training and validation sets, and machine learning was used to build predictive models. (3) Results: The neural network (NN) model achieved 73% accuracy and an AUC of 0.71 during training for predicting the KRAS mutation status, while during testing, it achieved 62.5% accuracy and an AUC of 0.68. In the case of tumor grading, the support vector machine (SVM) model excelled with a training accuracy of 72.93% and an AUC of 0.7, and during testing, it reached an accuracy of 72% and an AUC of 0.69. (4) Conclusions: ML models using radiomics from ADC maps and T2-weighted images are effective for distinguishing KRAS genes, tumor grading, and EMVI in colorectal cancer. Standardized protocols are essential to improve MRI radiomics’ reliability in clinical practice.

Delta radiomics analysis for prediction of intermediary- and high-risk factors for patients with locally advanced cervical cancer receiving neoadjuvant therapy

This study aimed to assess the feasibility of using magnetic resonance imaging (MRI)-based Delta radiomics characteristics extrapolated from the Ax LAVA + C series to identify intermediary- and high-risk factors in patients with cervical cancer undergoing surgery following neoadjuvant chemoradiotherapy. A total of 157 patients were divided into two groups: those without any intermediary- or high-risk factors and those with one intermediary-risk factor (negative group; n = 75). Those with any high-risk factor or more than one intermediary-risk factor (positive group; n = 82). Radiomics characteristics were extracted using Ax-LAVA + C MRI sequences. The data was divided into training (n = 126) and test (n = 31) sets in an 8:2 ratio. The training set data features were selected using the Mann–Whitney U test and the Least Absolute Shrinkage and Selection Operator (LASSO) test. The best radiomics features were then analyzed to build a preoperative predictive radiomics model for predicting intermediary- and high-risk factors in cervical cancer. Three models—the clinical model, the radiomics model, and the combined clinic and radiomics model—were developed in this study utilizing the random forest Algorithm. The receiver operating characteristic (ROC) curve, decision curve analysis (DCA), accuracy, sensitivity, and specificity were used to assess the predictive efficacy and clinical benefits of each model. Three models were developed in this study to predict intermediary- and high-risk variables associated with postoperative pathology for patients who underwent surgery after receiving neoadjuvant radiation. In the training and test sets, the AUC values assessed using the clinical model, radiomics model, and combined clinical and radiomics models were 0.76 and 0.70, 0.88 and 0.86, and 0.91 and 0.89, respectively. The use of machine learning algorithms to analyze Delta Ax LAVA + C MRI radiomics features can aid in the prediction of intermediary- and high-risk factors in patients with cervical cancer receiving neoadjuvant therapy.

Biological and clinical significance of radiomics features obtained from magnetic resonance imaging preceding pre-carbon ion radiotherapy in prostate cancer based on radiometabolomics.

We aimed to investigate the feasibility of metabolomics to explain the underlying biological implications of radiomics features obtained from magnetic resonance imaging (MRI) preceding carbon ion radiotherapy (CIRT) in patients with prostate cancer and to further explore the clinical significance of radiomics features on the prognosis of patients, based on their biochemical recurrence (BCR) status. Metabolomic results obtained using high-performance liquid chromatography coupled with tandem mass spectrometry of urine samples, combined with pre-RT radiomic features extracted from MRI images, were evaluated to investigate their biological significance. Receiver operating characteristic (ROC) curve analysis was subsequently conducted to examine the correlation between these biological implications and clinical BCR status. Statistical and metabolic pathway analyses were performed using MetaboAnalyst and R software. Correlation analysis revealed that methionine alteration extent was significantly related to four radiomic features (Contrast, Difference Variance, Small Dependence High Gray Level Emphasis, and Mean Absolute Deviation), which were significantly correlated with BCR status. The area under the curve (AUC) for BCR prediction of these four radiomic features ranged from 0.704 to 0.769, suggesting that the higher the value of these four radiomic features, the greater the decrease in methionine levels after CIRT and the lower the probability of BCR. Pre-CIRT MRI radiomic features were associated with CIRT-suppressed metabolites. These radiomic features can be used to predict the alteration in the amplitude of methionine after CIRT and the BCR status, which may contribute to the optimization of the CIRT strategy and deepen the understanding of PCa.

Feasibility analysis of magnetic resonance imaging-based radiomics features for preoperative prediction of nuclear grading of ductal carcinoma in situ.

The nuclear grading of ductal carcinoma in situ (DCIS) affects its clinical risk. The aim of this study was to investigate the possibility of predicting the nuclear grading of DCIS, by magnetic resonance imaging (MRI)-based radiomics features. And to develop a nomogram combining radiomics features and MRI semantic features to explore the potential role of MRI radiomic features in the assessment of DCIS nuclear grading. A total of 156 patients (159 lesions) with DCIS and DCIS with microinvasive (DCIS-MI) were enrolled in this retrospective study, with 112 lesions included in the training cohort and 47 lesions included in the validation cohort. Radiomics features were extracted from Dynamic contrast-enhanced MRI (DCE-MRI) phases 1st and 5th. After feature selection, radiomics signature was constructed and radiomics score (Rad-score) was calculated. Multivariate analysis was used to identify MRI semantic features that were significantly associated with DCIS nuclear grading and combined with Rad-score to construct a Nomogram. Receiver operating characteristic curves were used to evaluate the predictive performance of Rad-score and Nomogram, and decision curve analysis (DCA) was used to evaluate the clinical utility. In multivariate analyses of MRI semantic features, larger tumor size and heterogeneous enhancement pattern were significantly associated with high-nuclear grade DCIS (HNG DCIS). In the training cohort, Nomogram had an area under curve (AUC) of 0.879 and Rad-score had an AUC of 0.828. Similarly, in the independent validation cohort, Nomogram had an AUC value of 0.828 and Rad-score had an AUC of 0.772. In both the training and validation cohorts, Nomogram had a significantly higher AUC value than Rad-score (P<0.05). DCA confirmed that Nomogram had a higher net clinical benefit. MRI-based radiomic features can be used as potential biomarkers for assessing nuclear grading of DCIS. The nomogram constructed by radiomic features combined with semantic features is feasible in discriminating non-HNG and HNG DCIS.

Predicting the BRAF mutation with pretreatment MRI radiomics features for melanoma brain metastases receiving Gamma Knife radiosurgery

To develop a model using radiomics features extracted from magnetic resonance imaging (MRI) images of Gamma Knife radiosurgery (GKRS) to predict the BRAF mutation in patients with melanoma brain metastases (MBM). Data of 220 tumours were classified into two groups. One was a group whose BRAF mutation was identified, and the other group whose BRAF mutation was not identified. We extracted 1,962 radiomics features from gadolinium contrast-enhanced T1-weighted MRI treatment-planning images. Synthetic Minority Over-sampling TEchnique (SMOTE) was performed to address the unbalanced data-related issues. A single-layer neural network (NN) was used to build predictive models with radiomics features. The sensitivity, specificity, accuracy, and the area under the curve (AUC) were evaluated to assess the model performance. The prediction performance for the final evaluation without the SMOTE had an accuracy of 77.14%, a specificity of 82.44%, a sensitivity of 81.85%, and an AUC of 0.79. The application of SMOTE improved the prediction model to an accuracy of 83.1%, a specificity of 87.07%, a sensitivity of 78.82%, and an AUC of 0.82. The current study showed the feasibility of generating a highly accurate NN model for the BRAF mutation prediction. The prediction performance improved with SMOTE. The model assists physicians to obtain more accurate expectations of the treatment outcome without a genetic test.

Oligodendrocyte Transcription Factor 2 as a Potential Prognostic Biomarker of Glioblastoma: Kaplan-Meier Analysis and the Development of a Binary Predictive Model Based on Visually Accessible Rembrandt Image and Magnetic Resonance Imaging Radiomic Features.

Oligodendrocyte transcription factor 2 (OLIG2) is universally expressed in human glioblastoma (GB). Our study explores whether OLIG2 expression impacts GB patients' overall survival and establishes a machine learning model for OLIG2 level prediction in patients with GB based on clinical, semantic, and magnetic resonance imaging radiomic features. Kaplan-Meier analysis was used to determine the optimal cutoff value of the OLIG2 in 168 GB patients. Three hundred thirteen patients enrolled in the OLIG2 prediction model were randomly divided into training and testing sets in a ratio of 7:3. The radiomic, semantic, and clinical features were collected for each patient. Recursive feature elimination (RFE) was used for feature selection. The random forest (RF) model was built and fine-tuned, and the area under the curve was calculated to evaluate the performance. Finally, a new testing set excluding IDH-mutant patients was built and tested in a predictive model using the fifth edition of the central nervous system tumor classification criteria. One hundred nineteen patients were included in the survival analysis. Oligodendrocyte transcription factor 2 was positively associated with GB survival, with an optimal cutoff of 10% (P = 0.00093). One hundred thirty-four patients were eligible for the OLIG2 prediction model. An RFE-RF model based on 2 semantic and 21 radiomic signatures achieved areas under the curve of 0.854 in the training set, 0.819 in the testing set, and 0.825 in the new testing set. Glioblastoma patients with ≤10% OLIG2 expression tended to have worse overall survival. An RFE-RF model integrating 23 features can predict the OLIG2 level of GB patients preoperatively, irrespective of the central nervous system classification criteria, further guiding individualized treatment.

#6073 MAGNETIC RESONANCE RADIOMIC ANALYSIS FOR THE PREDICTION OF GRAFT FAILURE IN PATIENTS WITH KIDNEY TRANSPLANTATION

Abstract Background and Aims Prediction of future graft failure with non-invasive investigations is a relevant objective for transplant clinicians. We previously demonstrated a promising discrimination capacity for interstitial fibrosis / tubular atrophy (IFTA) &amp;gt;50% in kidney biopsy of a machine learning (ML) based magnetic resonance imaging (MRI) radiomic algorithm. Aim of the present study is to evaluate accuracy of MRI radiomic-based ML algorithms in predicting future graft failure. Method Single-center retrospective observational cohort study on a previously characterized cohort of kidney transplant recipients who underwent graft biopsy and graft MRI imaging within 6 months from biopsy, both on clinical indication, at the “Azienda Ospedaliero-Universitaria di Modena”, Italy, from 1/1/2012 to 1/3/2021. The primary outcome was to identify the best combination of MRI radiomic features for the prediction of graft failure during follow-up. Segmentation of renal parenchyma, cortex and medulla on MRI sequences was performed using the 3DSlicer software. Radiomic features were then extracted using an in-house software based on pyradiomics applying Wavelet and Gaussian filters. LASSO algorithm was employed to select correlated features with outcome and summarize them in a radiomic signature. Using radiomic signature alone and then merged with meaningful clinical data we trained ML-algorithms using 70% of cases for training/validation, with a 10-fold internal cross-validation, and 30% for model testing. Model performance was assessed using AUC with a 95% confidence interval (CI). Results Seventy coupled tests (63 patients) were included and randomly subdivided into a training/validation cohort (n = 50) and a test cohort (n = 20). Median follow-up was 24.73 months (interquartile range 13.64-46.57). Radiomic model had an AUC of 0.88 (95%_CI 0.70-0.97) in training and 0.57 (95%_CI 0.23-0.86) in test cohort. Radiomic-clinical mixed model had an AUC of 0.90 (95%_CI 0.73-0.98) in training and 0.66 (95%_CI 0.33-0.88) in test cohort. Conclusion We produced an MRI radiomic-based ML algorithm with good prediction ability for future graft failure in patients with kidney transplant. Given the limited number of enrolled patients, validation in larger (and ideally prospective) cohorts is required to confirm our findings. Comparison of radiomic-based ML with histological parameters (i.e., IFTA) for the prediction of graft failure in our cohort is currently under assessment.

Development and validation of a radiomics model of magnetic resonance for predicting liver metastasis in resectable pancreatic ductal adenocarcinoma patients

BackgroundNearly one fourth of patients with pancreatic ductal adenocarcinoma (PDAC) occur to liver metastasis after surgery, and liver metastasis is a risk factor for prognosis for those patients with surgery therapy. However, there is no effective way to predict liver metastasis post-operation.MethodClinical data and preoperative magnetic resonance imaging (MRI) of PDAC patients diagnosed between July 2010 and July 2020 were retrospectively collected from three hospital centers in China. The significant MRI radiomics features or clinicopathological characteristics were used to establish a model to predict liver metastasis in the development and validation cohort.ResultsA total of 204 PDAC patients from three hospital centers were divided randomly (7:3) into development and validation cohort. Due to poor predictive value of clinical features, MRI radiomics model had similar receiver operating characteristics curve (ROC) value to clinical-radiomics combing model in development cohort (0.878 vs. 0.880, p = 0.897) but better ROC in validation dataset (0.815 vs. 0.732, p = 0.022). Radiomics model got a sensitivity of 0.872/0.750 and a specificity of 0.760/0.822 to predict liver metastasis in development and validation cohort, respectively. Among 54 patients randomly selected with post-operation specimens, fibrosis markers (α-smooth muscle actin) staining was shown to promote radiomics model with ROC value from 0.772 to 0.923 (p = 0.049) to predict liver metastasis.ConclusionThis study developed and validated an MRI-based radiomics model and showed a good performance in predicting liver metastasis in resectable PDAC patients.