Multiparametric MRI radiomics for predicting disease-free survival and high-risk histopathological features for tumor recurrence in endometrial cancer.

Background: Current preoperative imaging is insufficient to predict survival and tumor recurrence in endometrial cancer (EC), necessitating invasive procedures for risk stratification. Purpose: To establish a multiparametric MRI radiomics model for predicting disease-free survival (DFS) and high-risk histopathologic features in EC. Methods: This retrospective study included 71 patients with histopathology-proven EC and preoperative MRI over a 10-year period. Clinicopathology data were extracted from health records. Manual MRI segmentation was performed on T2-weighted (T2W), apparent diffusion coefficient (ADC) maps and dynamic contrast-enhanced T1-weighted images (DCE T1WI). Radiomic feature (RF) extraction was performed with PyRadiomics. Associations between RF and histopathologic features were assessed using logistic regression. Associations between DFS and RF or clinicopathologic features were assessed using the Cox proportional hazards model. All RF with univariate analysis p-value < 0.2 were included in elastic net analysis to build radiomic signatures. Results: The 3-year DFS rate was 68% (95% CI = 57%-80%). There were no significant clinicopathologic predictors for DFS, whilst the radiomics signature was a strong predictor of DFS (p<0.001, HR 3.62, 95% CI 1.94, 6.75). From 107 RF extracted, significant predictive elastic net radiomic signatures were established for deep myometrial invasion (p=0.0097, OR 4.81, 95% CI 1.46, 15.79), hysterectomy grade (p=0.002, OR 5.12, 95% CI 1.82, 14.45), hysterectomy histology (p=0.0061, OR 18.25, 95% CI 2.29,145.24) and lymphovascular space invasion (p<0.001, OR 5.45, 95% CI 2.07, 14.36). Conclusion: Multiparametric MRI radiomics has the potential to create a non-invasive a priori approach to predicting DFS and high-risk histopathologic features in EC.

Comparison of radiomics-based machine-learning classifiers for the pretreatment prediction of pathologic complete response to neoadjuvant therapy in breast cancer.

Background: Machine learning classifiers are increasingly used to create predictive models for pathological complete response (pCR) in breast cancer after neoadjuvant therapy (NAT). Few studies have compared the effectiveness of different ML classifiers. This study evaluated radiomics models based on pre- and post-contrast first-phase T1 weighted images (T1WI) in predicting breast cancer pCR after NAT and compared the performance of ML classifiers. Methods: This retrospective study enrolled 281 patients undergoing NAT from the Duke-Breast-Cancer-MRI dataset. Radiomic features were extracted from pre- and post-contrast first-phase T1WI images. The Synthetic Minority Oversampling Technique (SMOTE) was applied, then the dataset was randomly divided into training and validation groups (7:3). The radiomics model was built using selected optimal features. Support vector machine (SVM), random forest (RF), decision tree (DT), k-nearest neighbor (KNN), extreme gradient boosting (XGBoost), and light gradient boosting machine (LightGBM) were classifiers. Receiver operating characteristic curves were used to assess predictive performance. Results: LightGBM performed best in predicting pCR [area under the curve (AUC): 0.823, 95% confidence interval (CI) [0.743-0.902], accuracy 74.0%, sensitivity 85.0%, specificity 67.2%]. During subgroup analysis, RF was most effective in pCR prediction in luminal breast cancers (AUC: 0.914, 95% CI [0.847-0.981], accuracy 87.0%, sensitivity 85.2%, specificity 88.1%). In triple-negative breast cancers, LightGBM performed best (AUC: 0.836, 95% CI [0.708-0.965], accuracy 78.6%, sensitivity 68.2%, specificity 90.0%). Conclusion: The LightGBM-based radiomics model performed best in predicting pCR in patients with breast cancer. RF and LightGBM showed promising results for luminal and triple-negative breast cancers, respectively.

MRI Radiomics Combined with Clinicopathological Factors for Predicting 3-Year Overall Survival of Hepatocellular Carcinoma After Hepatectomy.

Background: A limited number of studies have examined the use of radiomics to predict 3-year overall survival (OS) after hepatectomy in patients with hepatocellular carcinoma (HCC). This study develops 3-year OS prediction models for HCC patients after liver resection using MRI radiomics and clinicopathological factors. Materials and Methods: A retrospective analysis of 141 patients who underwent surgical resection of HCC was performed. Patients were randomized into two set: the training set (n=98) and the validation set (n=43) including the survival groups (n=111) and non-survival groups (n=30) based on 3-year survival after hepatectomy. Furthermore, x2 or Fisher's exact test, univariate and multivariate logistic regression analyses were conducted to determine independent clinicopathological risk factors associated with 3-year OS. 1688 quantitative imaging features were extracted from preoperative T2-weighted imaging (T2WI) and contrast-enhanced magnetic resonance imaging (CE-MRI) of arterial phase (AP), portal venous phases (PVP)and delay period (DP). The features were selected using the variance threshold method, the select K best method and the least absolute shrinkage and selection operator (LASSO) algorithm. By using Bernoulli Naive Bayes (BernoulliNB) and Multinomial Naive Bayes (MultinomialNB) classifiers, we constructed models based on the independent clinicopathological factors and Rad-scores. To determine the best model, receiver operating characteristics (ROC) and Delong's test were used. Moreover, calibration curves were used to determine the calibration ability of the model, while decision curve analysis (DCA) was implemented to evaluate its clinical benefit. Results: The fusion model showed excellent prediction precision with AUC of 0.910 and 0.846 in training and validation set and revealed significant diagnostic accuracy and value in the calibration curve and DCA analysis. Conclusion: Nomograms based on MRI radiomics and clinicopathological factors have significant predictive value for 3-year OS after hepatectomy and can be used for risk classification.

Multi-parametric MRI radiomics for predicting response to neoadjuvant therapy in patients with locally advanced rectal cancer.

OBJECTIVE: This study aims to evaluate the application value of multi-parametric magnetic resonance imaging (MRI) radiomics in predicting the response of patients with locally advanced rectal cancer (LARC) to neoadjuvant chemoradiotherapy(nCRT), aiming to provide non-invasive biomarkers for clinical decision-making in personalized treatment. METHODS: A retrospective analysis was conducted on the clinical data and imaging records of patients with LARC who received nCRT and total mesorectal excision (TME) in two medical centers from 2017 to 2023. The patients were divided into a training group and a test group in a 7:3 ratio. Through radiomics analysis, radiomics features of tumor volume and mesorectal fat at baseline, before and after neoadjuvant therapy were extracted. Radiomics models based on single sequences (T2WI, DWI) and multi-sequence fusion were constructed, and the logistic regression classifier model was used to evaluate the prediction performance. RESULTS: A total of 82 patients were included, with 30 in the good response group and 52 in the poor response group. Through the selection of radiomics features, radiomics models based on baseline MRI of tumor volume, mesorectal fat, and differences before and after treatment (Delta) were constructed. The area under the receiver operating characteristic curve (AUC) of the multi-parametric radiomics fusion model in the training and test groups was 0.852 and 0.848, respectively, showing high prediction performance and good calibration. CONCLUSION: This study demonstrates that the multi-parametric MRI radiomics model can effectively predict the response of patients with locally advanced rectal cancer to neoadjuvant chemoradiotherapy. Especially, the fusion model provides high accuracy and good calibration. This result is conducive to the formulation of personalized treatment plans and optimization of treatment strategies.

Preoperative prediction of histopathological grading in patients with chondrosarcoma using MRI-based radiomics with semantic features.

BACKGROUND: Distinguishing high-grade from low-grade chondrosarcoma is extremely vital not only for guiding the development of personalized surgical treatment but also for predicting the prognosis of patients. We aimed to establish and validate a magnetic resonance imaging (MRI)-based nomogram for predicting preoperative grading in patients with chondrosarcoma. METHODS: Approximately 114 patients (60 and 54 cases with high-grade and low-grade chondrosarcoma, respectively) were recruited for this retrospective study. All patients were treated via surgery and histopathologically proven, and they were randomly divided into training (n = 80) and validation (n = 34) sets at a ratio of 7:3. Next, radiomics features were extracted from two sequences using the least absolute shrinkage and selection operator (LASSO) algorithms. The rad-scores were calculated and then subjected to logistic regression to develop a radiomics model. A nomogram combining independent predictive semantic features with radiomic by using multivariate logistic regression was established. The performance of each model was assessed by the receiver operating characteristic (ROC) curve analysis and the area under the curve, while clinical efficacy was evaluated via decision curve analysis (DCA). RESULTS: Ultimately, six optimal radiomics signatures were extracted from T1-weighted imaging (T1WI) and T2-weighted imaging with fat suppression (T2WI-FS) sequences to develop the radiomics model. Tumour cartilage abundance, which emerged as an independent predictor, was significantly related to chondrosarcoma grading (p < 0.05). The AUC values of the radiomics model were 0.85 (95% CI, 0.76 to 0.95) in the training sets, and the corresponding AUC values in the validation sets were 0.82 (95% CI, 0.65 to 0.98), which were far superior to the clinical model AUC values of 0.68 (95% CI, 0.58 to 0.79) in the training sets and 0.72 (95% CI, 0.57 to 0.87) in the validation sets. The nomogram demonstrated good performance in the preoperative distinction of chondrosarcoma. The DCA analysis revealed that the nomogram model had a markedly higher clinical usefulness in predicting chondrosarcoma grading preoperatively than either the rad-score or clinical model alone. CONCLUSION: The nomogram based on MRI radiomics combined with optimal independent factors had better performance for the preoperative differentiation between low-grade and high-grade chondrosarcoma and has potential as a noninvasive preoperative tool for personalizing clinical plans.

MRI radiomics captures early treatment response in patient-derived organoid endometrial cancer mouse models.

Background: Radiomics can capture microscale information in medical images beyond what is visible to the naked human eye. Using a clinically relevant mouse model for endometrial cancer, the objective of this study was to develop and validate a radiomic signature (RS) predicting response to standard chemotherapy. Methods: Mice orthotopically implanted with a patient-derived grade 3 endometrioid endometrial cancer organoid model (O-PDX) were allocated to chemotherapy (combined paclitaxel/carboplatin, n=11) or saline/control (n=13). During tumor progression, the mice underwent weekly T2-weighted (T2w) magnetic resonance imaging (MRI). Segmentation of primary tumor volume (vMRI) allowed extraction of radiomic features from whole-volume tumor masks. A radiomic model for predicting treatment response was derived employing least absolute shrinkage and selection operator (LASSO) statistics at endpoint images in the orthotopic O-PDX (RS_O), and subsequently applied on the earlier study timepoints (RS_O at baseline, and week 1-3). For external validation, the radiomic model was tested in a separate T2w-MRI dataset on segmented whole-volume subcutaneous tumors (RS_S) from the same O-PDX model, imaged at three timepoints (baseline, day 3 and day 10/endpoint) after start of chemotherapy (n=8 tumors) or saline/control (n=8 tumors). Results: The RS_O yielded rapidly increasing area under the receiver operating characteristic (ROC) curves (AUCs) for predicting treatment response from baseline until endpoint; AUC=0.38 (baseline); 0.80 (week 1), 0.85 (week 2), 0.96 (week 3) and 1.0 (endpoint). In comparison, vMRI yielded AUCs of 0.37 (baseline); 0.69 (w1); 0.83 (week 2); 0.92 (week 3) and 0.97 (endpoint). When tested in the external validation dataset, RS_S yielded high accuracy for predicting treatment response at day10/endpoint (AUC=0.85) and tended to yield higher AUC than vMRI (AUC=0.78, p=0.18). Neither RS_S nor vMRI predicted response at day 3 in the external validation set (AUC=0.56 for both). Conclusions: We have developed and validated a radiomic signature that was able to capture chemotherapeutic treatment response both in an O-PDX and in a subcutaneous endometrial cancer mouse model. This study supports the promising role of preclinical imaging including radiomic tumor profiling to assess early treatment response in endometrial cancer models.

Predictive value of background parenchymal enhancement on breast magnetic resonance imaging for pathological tumor response to neoadjuvant chemotherapy in breast cancers: a systematic review.

OBJECTIVES: This review aimed to assess the predictive value of background parenchymal enhancement (BPE) on breast magnetic resonance imaging (MRI) as an imaging biomarker for pathologic complete response (pCR) after neoadjuvant chemotherapy (NACT). METHODS: Two reviewers independently performed a systemic literature search using the PubMed, MEDLINE, and Embase databases for studies published up to 11 June 2022. Data from relevant articles were extracted to assess the relationship between BPE and pCR. RESULTS: This systematic review included 13 studies with extensive heterogeneity in population characteristics, MRI follow-up points, MRI protocol, NACT protocol, pCR definition, and BPE assessment. Baseline BPE levels were not associated with pCR, except in 1 study that reported higher baseline BPE of the younger participants (< 55 years) in the pCR group than the non-pCR group. A total of 5 studies qualitatively assessed BPE levels and indicated a correlation between reduced BPE after NACT and pCR; however, among the studies that quantitatively measured BPE, the same association was observed only in the subgroup analysis of 2 articles that assessed the status of hormone receptor and human epidermal growth factor receptor 2. In addition, the predictive ability of early BPE changes for pCR was reported in several articles and remains controversial. CONCLUSIONS: Changes in BPE may be a promising imaging biomarker for predicting pCR in breast cancer. Because current studies remain insufficient, particularly those that quantitatively measure BPE, prospective and multicenter large-sample studies are needed to confirm this relationship.

Nanoparticle-mediated blockade of CXCL12/CXCR4 signaling enhances glioblastoma immunotherapy: Monitoring early responses with MRI radiomics.

The limited therapeutic efficacy of checkpoint blockade immunotherapy against glioblastoma is closely related to the blood-brain barrier (BBB) and tumor immunosuppressive microenvironment, where the latter is driven primarily by tumor-associated myeloid cells (TAMCs). Targeting the C-X-C motif chemokine ligand-12/C-X-C motif chemokine receptor-4 (CXCL12/CXCR4) signaling orchestrates the recruitment of TAMCs and has emerged as a promising approach for alleviating immunosuppression. Herein, we developed an iRGD ligand-modified polymeric nanoplatform for the co-delivery of CXCR4 antagonist AMD3100 and the small-molecule immune checkpoint inhibitor BMS-1. The iRGD peptide facilitated superior BBB crossing and tumor-targeting abilities both in vitro and in vivo. In mice bearing orthotopic GL261-Luc tumor, co-administration of AMD3100 and BMS-1 significantly inhibited tumor proliferation without adverse effects. A reprogramming of immunosuppression upon CXCL12/CXCR4 signaling blockade was observed, characterized by the reduction of TAMCs and regulatory T cells, and an increased proportion of CD8+T lymphocytes. The elevation of interferon-γ secreted from activated immune cells upregulated PD-L1 expression in tumor cells, highlighting the synergistic effect of BMS-1 in counteracting the PD-1/PD-L1 pathway. Finally, our research unveiled the ability of MRI radiomics to reveal early changes in the tumor immune microenvironment following immunotherapy, offering a powerful tool for monitoring treatment responses. STATEMENT OF SIGNIFICANCE: The insufficient BBB penetration and immunosuppressive tumor microenvironment greatly diminish the efficacy of immunotherapy for glioblastoma (GBM). In this study, we prepared iRGD-modified polymeric nanoparticles, loaded with a CXCR4 antagonist (AMD3100) and a small-molecule checkpoint inhibitor of PD-L1 (BMS-1) to overcome physical barriers and reprogram the immunosuppressive microenvironment in orthotopic GBM models. In this nanoplatform, AMD3100 converted the "cold" immune microenvironment into a "hot" one, while BMS-1 synergistically counteracted PD-L1 inhibition, enhancing GBM immunotherapy. Our findings underscore the potential of dual-blockade of CXCL12/CXCR4 and PD-1/PD-L1 pathways as a complementary approach to maximize therapeutic efficacy for GBM. Moreover, our study revealed that MRI radiomics provided a clinically translatable means to assess immunotherapeutic efficacy.

Feasibility of constructing a diagnostic classification model for cervical instability by magnetic resonance imaging radiomics / 中国组织工程研究

BACKGROUND:Previous studies on cervical instability failed to explain the dynamic and static interaction relationship and pathological characteristics changes in the development of cervical lesions under the traditional imaging examination.In recent years,the emerging nuclear magnetic resonance imaging(MRI)radiomics can provide a new way for in-depth research on cervical instability. OBJECTIVE:To investigate the application value of MRI radiomics in the study of cervical instability. METHODS:Through recruitment advertisements and the Second Department of Spine of Wangjing Hospital,China Academy of Chinese Medical Sciences,young cervical vertebra unstable subjects and non-unstable subjects aged 18-45 years were included in the cervical vertebra nuclear magnetic image collection.Five specific regions of interest,including the intervertebral disc region,the facet region,the prevertebral muscle region,the deep region of the posterior cervical muscle group,and the superficial region of the posterior cervical muscle group,were manually segmented to extract and screen the image features.Finally,the cervical instability diagnosis classification model was constructed,and the effectiveness of the model was evaluated using the area under the curve. RESULTS AND CONCLUSION:(1)A total of 56 subjects with cervical instability and 55 subjects with non-instability were included,and 1 688 imaging features were extracted for each region of interest.After screening,300 sets of specific image feature combinations were obtained,with 60 sets of regions of interest for each group.(2)Five regions of interest diagnostic classification models for cervical instability were initially established.Among them,the support vector machine model for the articular process region and the support vector machine model for the deep cervical muscle group had certain accuracy for the classification of instability and non-instability,and the average area under the curve of ten-fold cross-validation was 0.719 7 and 0.703 3,respectively.(3)The Logistic model in the intervertebral disc region,the LightGBM model in the prevertebral muscle region,and the Logistic model in the superficial posterior cervical muscle region were generally accurate in the classification of instability and non-instability,and the average area under the curve of ten-fold cross-validation was 0.650 4,0.620 7,and 0.644 2,respectively.(4)This study proved the feasibility of MRI radiomics in the study of cervical instability,further deepened the understanding of the pathogenesis of cervical instability,and also provided an objective basis for the accurate diagnosis of cervical instability.

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.

Could normalization improve robustness of abdominal MRI radiomic features?

Radiomics-based systems could improve the management of oncological patients by supporting cancer diagnosis, treatment planning, and response assessment. However, one of the main limitations of these systems is the generalizability and reproducibility of results when they are applied to images acquired in different hospitals by different scanners. Normalization has been introduced to mitigate this issue, and two main approaches have been proposed: one rescales the image intensities (image normalization), the other the feature distributions for each center (feature normalization). The aim of this study is to evaluate how different image and feature normalization methods impact the robustness of 93 radiomics features acquired using a multicenter and multi-scanner abdominal Magnetic Resonance Imaging (MRI) dataset. To this scope, 88 rectal MRIs were retrospectively collected from 3 different institutions (4 scanners), and for each patient, six 3D regions of interest on the obturator muscle were considered. The methods applied were min-max, 1st-99th percentiles and 3-Sigma normalization, z-score standardization, mean centering, histogram normalization, Nyul-Udupa and ComBat harmonization. The Mann-Whitney U-test was applied to assess features repeatability between scanners, by comparing the feature values obtained for each normalization method, including the case in which no normalization was applied. Most image normalization methods allowed to reduce the overall variability in terms of intensity distributions, while worsening or showing unpredictable results in terms of feature robustness, except for thez-score, which provided a slight improvement by increasing the number of statistically similar features from 9/93 to 10/93. Conversely, feature normalization methods positively reduced the overall variability across the scanners, in particular, 3sigma,z_scoreandComBatthat increased the number of similar features (79/93). According to our results, it emerged that none of the image normalization methods was able to strongly increase the number of statistically similar features.

Radiomics for the Prediction of Pathological Complete Response to Neoadjuvant Chemoradiation in Locally Advanced Rectal Cancer: A Prospective Observational Trial.

(1) Background: An increasing amount of research has supported the role of radiomics for predicting pathological complete response (pCR) to neoadjuvant chemoradiation treatment (nCRT) in order to provide better management of locally advanced rectal cancer (LARC) patients. However, the lack of validation from prospective trials has hindered the clinical adoption of such studies. The purpose of this study is to validate a radiomics model for pCR assessment in a prospective trial to provide informative insight into radiomics validation. (2) Methods: This study involved a retrospective cohort of 147 consecutive patients for the development/validation of a radiomics model, and a prospective cohort of 77 patients from two institutions to test its generalization. The model was constructed using T2-weighted, diffusion-weighted, and dynamic contrast-enhanced MRI to understand the associations with pCR. The consistency of physicians' evaluations and agreement on pathological complete response prediction were also evaluated, with and without the aid of the radiomics model. (3) Results: The radiomics model outperformed both physicians' visual assessments in the prospective test cohort, with an area under the curve (AUC) of 0.84 (95% confidence interval of 0.70-0.94). With the aid of the radiomics model, a junior physician could achieve comparable performance as a senior oncologist. (4) Conclusion: We have built and validated a radiomics model with pretreatment MRI for pCR prediction of LARC patients undergoing nCRT.

Predicting survival after radiosurgery in patients with lung cancer brain metastases using deep learning of radiomics and EGFR status.

The early prediction of overall survival (OS) in patients with lung cancer brain metastases (BMs) after Gamma Knife radiosurgery (GKRS) can facilitate patient management and outcome improvement. However, the disease progression is influenced by multiple factors, such as patient characteristics and treatment strategies, and hence satisfactory performance of OS prediction remains challenging. Accordingly, we proposed a deep learning approach based on comprehensive predictors, including clinical, imaging, and genetic information, to accomplish reliable and personalized OS prediction in patients with BMs after receiving GKRS. Overall 1793 radiomic features extracted from pre-GKRS magnetic resonance images (MRI), clinical information, and epidermal growth factor receptor (EGFR) mutation status were retrospectively collected from 237 BM patients who underwent GKRS. DeepSurv, a multi-layer perceptron model, with 4 different aggregation methods of radiomics was applied to predict personalized survival curves and survival status at 3, 6, 12, and 24 months. The model combining clinical features, EGFR status, and radiomics from the largest BM showed the best prediction performance with concordance index of 0.75 and achieved areas under the curve of 0.82, 0.80, 0.84, and 0.92 for predicting survival status at 3, 6, 12, and 24 months, respectively. The DeepSurv model showed a significant improvement (p < 0.001) in concordance index compared to the validated lung cancer BM prognostic molecular markers. Furthermore, the model provided a novel estimate of the risk-of-death period for patients. The personalized survival curves generated by the DeepSurv model effectively predicted the risk-of-death period which could facilitate personalized management of patients with lung cancer BMs.

Automated Prediction of Neoadjuvant Chemoradiotherapy Response in Locally Advanced Cervical Cancer Using Hybrid Model-Based MRI Radiomics.

BACKGROUND: This study aimed to develop a model that automatically predicts the neoadjuvant chemoradiotherapy (nCRT) response for patients with locally advanced cervical cancer (LACC) based on T2-weighted MR images and clinical parameters. METHODS: A total of 138 patients were enrolled, and T2-weighted MR images and clinical information of the patients before treatment were collected. Clinical information included age, stage, pathological type, squamous cell carcinoma (SCC) level, and lymph node status. A hybrid model extracted the domain-specific features from the computational radiomics system, the abstract features from the deep learning network, and the clinical parameters. Then, it employed an ensemble learning classifier weighted by logistic regression (LR) classifier, support vector machine (SVM) classifier, K-Nearest Neighbor (KNN) classifier, and Bayesian classifier to predict the pathologic complete response (pCR). The area under the receiver operating characteristics curve (AUC), accuracy (ACC), true positive rate (TPR), true negative rate (TNR), and precision were used as evaluation metrics. RESULTS: Among the 138 LACC patients, 74 were in the pCR group, and 64 were in the non-pCR group. There was no significant difference between the two cohorts in terms of tumor diameter (p = 0.787), lymph node (p = 0.068), and stage before radiotherapy (p = 0.846), respectively. The 109-dimension domain features and 1472-dimension abstract features from MRI images were used to form a hybrid model. The average AUC, ACC, TPR, TNR, and precision of the proposed hybrid model were about 0.80, 0.71, 0.75, 0.66, and 0.71, while the AUC values of using clinical parameters, domain-specific features, and abstract features alone were 0.61, 0.67 and 0.76, respectively. The AUC value of the model without an ensemble learning classifier was 0.76. CONCLUSIONS: The proposed hybrid model can predict the radiotherapy response of patients with LACC, which might help radiation oncologists create personalized treatment plans for patients.

Precision of MRI radiomics features in the liver and hepatocellular carcinoma.

OBJECTIVES: To assess the precision of MRI radiomics features in hepatocellular carcinoma (HCC) tumors and liver parenchyma. METHODS: The study population consisted of 55 patients, including 16 with untreated HCCs, who underwent two repeat contrast-enhanced abdominal MRI exams within 1 month to evaluate: (1) test-retest repeatability using the same MRI system (n = 28, 10 HCCs); (2) inter-platform reproducibility between different MRI systems (n = 27, 6 HCCs); (3) inter-observer reproducibility (n = 16, 16 HCCs). Shape and 1st- and 2nd-order radiomics features were quantified on pre-contrast T1-weighted imaging (WI), T1WI portal venous phase (pvp), T2WI, and ADC (apparent diffusion coefficient), on liver regions of interest (ROIs) and HCC volumes of interest (VOIs). Precision was assessed by calculating intraclass correlation coefficient (ICC), concordance correlation coefficient (CCC), and coefficient of variation (CV). RESULTS: There was moderate to excellent test-retest repeatability of shape and 1st- and 2nd-order features for all sequences in HCCs (ICC: 0.53-0.99; CV: 3-29%), and moderate to good test-retest repeatability of 1st- and 2nd-order features for T1WI sequences, and 2nd-order features for T2WI in the liver (ICC: 0.53-0.73; CV: 12-19%). There was poor inter-platform reproducibility for all features and sequences, except for shape and 1st-order features on T1WI in HCCs (CCC: 0.58-0.99; CV: 3-15%). Good to excellent inter-observer reproducibility was found for all features and sequences in HCCs (CCC: 0.80-0.99; CV: 4-15%) and moderate to good for liver (CCC: 0.45-0.86; CV: 6-25%). CONCLUSIONS: MRI radiomics features have acceptable repeatability in the liver and HCC when using the same MRI system and across readers but have low reproducibility across MR systems, except for shape and 1st-order features on T1WI. Data must be interpreted with caution when performing multiplatform radiomics studies. KEY POINTS: • MRI radiomics features have acceptable repeatability when using the same MRI system but less reproducible when using different MRI platforms. • MRI radiomics features extracted from T1 weighted-imaging show greater stability across exams than T2 weighted-imaging and ADC. • Inter-observer reproducibility of MRI radiomics features was found to be good in HCC tumors and acceptable in liver parenchyma.

Impact of Preprocessing and Harmonization Methods on the Removal of Scanner Effects in Brain MRI Radiomic Features.

In brain MRI radiomics studies, the non-biological variations introduced by different image acquisition settings, namely scanner effects, affect the reliability and reproducibility of the radiomics results. This paper assesses how the preprocessing methods (including N4 bias field correction and image resampling) and the harmonization methods (either the six intensity normalization methods working on brain MRI images or the ComBat method working on radiomic features) help to remove the scanner effects and improve the radiomic feature reproducibility in brain MRI radiomics. The analyses were based on in vitro datasets (homogeneous and heterogeneous phantom data) and in vivo datasets (brain MRI images collected from healthy volunteers and clinical patients with brain tumors). The results show that the ComBat method is essential and vital to remove scanner effects in brain MRI radiomic studies. Moreover, the intensity normalization methods, while not able to remove scanner effects at the radiomic feature level, still yield more comparable MRI images and improve the robustness of the harmonized features to the choice among ComBat implementations.

Multiparametric MRI Features Predict the SYP Gene Expression in Low-Grade Glioma Patients: A Machine Learning-Based Radiomics Analysis.

PURPOSE: Synaptophysin (SYP) gene expression levels correlate with the survival rate of glioma patients. This study aimed to explore the feasibility of applying a multiparametric magnetic resonance imaging (MRI) radiomics model composed of a convolutional neural network to predict the SYP gene expression in patients with glioma. METHOD: Using the TCGA database, we examined 614 patients diagnosed with glioma. First, the relationship between the SYP gene expression level and outcome of survival rate was investigated using partial correlation analysis. Then, 7266 patches were extracted from each of the 108 low-grade glioma patients who had available multiparametric MRI scans, which included preoperative T1-weighted images (T1WI), T2-weighted images (T2WI), and contrast-enhanced T1WI images in the TCIA database. Finally, a radiomics features-based model was built using a convolutional neural network (ConvNet), which can perform autonomous learning classification using a ROC curve, accuracy, recall rate, sensitivity, and specificity as evaluation indicators. RESULTS: The expression level of SYP decreased with the increase in the tumor grade. With regard to grade II, grade III, and general patients, those with higher SYP expression levels had better survival rates. However, the SYP expression level did not show any significant association with the outcome in Level IV patients. CONCLUSION: Our multiparametric MRI radiomics model constructed using ConvNet showed good performance in predicting the SYP gene expression level and prognosis in low-grade glioma patients.

Superficial fibromatosis: MRI radiomics and T2 mapping correlate with treatment response.

BACKGROUND: Superficial fibromatosis exhibits variable MR signal intensity due to collagenous and fibroproliferative components. Quantifying this signal heterogeneity using image texture analysis and T2-mapping could have prognostic and therapeutic implications. METHODS: This IRB-approved retrospective study included 13 patients with superficial fibromatosis, managed by observation, electron beam radiotherapy (EBT), or pentoxifylline/vitamin E. Two-dimensional regions of interest (ROIs) were drawn on proton-density or T2-weighted MRI for radiomics feature analysis, and corresponding T2-maps. Comparisons were made between baseline and follow-up T2 relaxation times and radiomics features: Shannon's entropy, kurtosis, skewness, mean of positive pixels (MPP), and uniformity of distribution of positive gray-level pixel values (UPP). RESULTS: There were 19 nodules in 13 subjects. Mean patient age was 60 years; 62% (8/13) were female; mean follow-up was 9.7 months. Nodule diameter at baseline averaged 18.2 mm (std dev 16.2 mm) and decreased almost 10% to 16.6 mm (p = 0.1, paired t-test). Normalized T2 signal intensity decreased 23% from 0.71 to 0.55 (p = 0.03, paired t-test). T2 relaxation time decreased 16% from 46.5 to 39.1 ms (p < 0.001, paired t-test). Among radiomics features, skewness increased to 0.71 from 0.41 (p = 0.03, paired t-test), and entropy decreased from 8.37 to 8.03 (p = 0.05, paired t-test); differences in other radiomics features were not significant. CONCLUSIONS: Radiomics analysis and T2-mapping of superficial fibromatosis is feasible; robust decreases in absolute T2 relaxation time, and changes in image textural features (increased skewness and decreased entropy) offer novel imaging biomarkers of nodule collagenization and maturation.

Detection of Immunotherapeutic Response in a Transgenic Mouse Model of Pancreatic Ductal Adenocarcinoma Using Multiparametric MRI Radiomics: A Preliminary Investigation.

RATIONALE AND OBJECTIVES: To develop classification and regression models interpreting tumor characteristics obtained from structural (T1w and T2w) magnetic resonance imaging (MRI) data for early detection of dendritic cell (DC) vaccine treatment effects and prediction of long-term outcomes for LSL-KrasG12D; LSL-Trp53R172H; Pdx-1-Cre (KPC) transgenic mice model of pancreatic ductal adenocarcinoma. MATERIALS AND METHODS: Eight mice were treated with DC vaccine for 3 weeks while eight KPC mice were used as untreated control subjects. The reproducibility of the computed 264 features was evaluated using the intraclass correlation coefficient. Key variables were determined using a three-step feature selection approach. Support vector machines classifiers were generated to differentiate treatment-related changes on tumor tissue following first- and third weeks of the DC vaccine therapy. The multivariable regression models were generated to predict overall survival (OS) and histological tumor markers of KPC mice using quantitative features. RESULTS: The quantitative features computed from T1w MRI data have better reproducibility than T2w MRI features. The KPC mice in treatment and control groups were differentiated with a longitudinally increasing accuracy (first- and third weeks: 87.5% and 93.75%). The linear regression model generated with five features of T1w MRI data predicted OS with a root-mean-squared error (RMSE) <6 days. The proposed multivariate regression models predicted histological tumor markers with relative error <2.5% for fibrosis percentage (RMSE: 0.414), CK19+ area (RMSE: 0.027), and Ki67+ cells (RMSE: 0.190). CONCLUSION: Our results demonstrated that proposed models generated with quantitative MRI features can be used to detect early treatment-related changes in tumor tissue and predict OS of KPC mice following DC vaccination.

Machine learning prediction of axillary lymph node metastasis in breast cancer: 2D versus 3D radiomic features.

PURPOSE: The purpose of this study was to distinguish axillary lymph node (ALN) status using preoperative breast DCE-MRI radiomics and compare the effects of two-dimensional (2D) and three-dimensional (3D) analysis. METHODS: A retrospective study including 154 breast cancer patients all confirmed by pathology; 80 with ALN metastasis and 74 without. All MRI scans were achieved at a 3.0 Tesla scanner with 7 post-contrast MR phases sequentially acquired with a temporal resolution of 60 s. MRI radiomic features were extracted separately from a 2D single slice (i.e., the representative slice) and the 3D tumor volume. Several machine learning classifiers were built and compared using 2D or 3D analysis to distinguish positive vs negative ALN status. We performed independent test and 10-fold cross validation with multiple repetitions, and used bootstrap test, least absolute shrinkage selection operator, and receiver operating characteristic (ROC) curve analysis as statistical tests. RESULTS: The highest area under the ROC curve (AUC) was 0.81 (95% confidence intervals [CI]: 0.80-0.83) and 0.82 (95% CI: 0.81-0.82) for 2D and 3D analysis, respectively; the corresponding accuracy was 79% and 80%. The linear discriminant analysis (LDA) classifier achieved the highest classification performance. None of the AUC differences between 2D and 3D analysis was statistically significant for the several tested machine learning classifiers (all P> 0.05). CONCLUSIONS: Radiomic features from segmented tumor region in breast MRI were associated with ALN status. The separate radiomic analysis on 3D tumor volume showed a similar effect to the 2D analysis on the single representative slice in the tested machine learning classifiers.