Radiomic Feature‐Based Prediction of Primary Cancer Origins in Brain Metastases Using Machine Learning

Prediction of Brain Metastasis Response to Stereotactic Radiosurgery Using MRI and Machine Learning: Effects of Primary Cancer Site and Metastasis Volume

To assess the diagnostic value of magnetic resonance imaging (MRI)-based radiomics features using machine learning (ML) models in characterizing solid renal neoplasms, in comparison/combination with qualitative radiologic evaluation. Retrospective analysis of 125 patients (mean age 59years, 67% males) with solid renal neoplasms that underwent MRI before surgery. Qualitative (signal and enhancement characteristics) and quantitative radiomics analyses (histogram and texture features) were performed on T2-weighted imaging (WI), T1-WI pre- and post-contrast, and DWI. Mann-Whitney U test and receiver-operating characteristic analysis were used in a training set (n = 88) to evaluate diagnostic performance of qualitative and radiomics features for differentiation of renal cell carcinomas (RCCs) from benign lesions, and characterization of RCC subtypes (clear cell RCC [ccRCC] and papillary RCC [pRCC]). Random forest ML models were developed for discrimination between tumor types on the training set, and validated on an independent set (n = 37). We assessed 104 RCCs (51 ccRCC, 29 pRCC, and 24 other subtypes) and 21 benign lesions in 125 patients. Significant qualitative and quantitative radiomics features (area under the curve [AUC] between 0.62 and 0.90) were included for ML analysis. Models with best diagnostic performance on validation sets showed AUC of 0.73 (confidence interval [CI] 0.5-0.96) for differentiating RCC from benign lesions (using combination of qualitative and radiomics features); AUC of 0.77 (CI 0.62-0.92) for diagnosing ccRCC (using radiomics features), and AUC of 0.74 (CI 0.53-0.95) for diagnosing pRCC (using qualitative features). ML models incorporating MRI-based radiomics features and qualitative radiologic assessment can help characterize renal masses.

Stereotactic radiosurgery (SRS) is frequently employed to treat brain metastases. However, <50% of patients treated with this method develop distant brain metastases (DBMs). As a result, these patients are followed using Magnetic Resonance Imaging (MRI) to identify DBM development. There is no current pre-treatment risk metric to identify which patients might be likely to develop DBMs. In this study, pre-treatment MRIs and radiotherapy planning data including structure sets and radiation dose maps were obtained for 81 SRS brain metastases treatment courses. Clinical variables including performance status, age, number of tumors, and primary tumor type were also collected. Pre-treatment MRIs were skull-stripped and normalized. 3D radiomic features from grey-intensity, Laws Energy, Gabor, Haralick, and CoLlAGe feature families were extracted from T1, T1 contrast-enhanced (T1w), T2, and FLAIR pre-treatment MRI sequences in brain regions receiving 0-25%, 25-50%, 50-75%, and 75-100% of prescribed radiation dose. A baseline classification model for DBM was created using clinical variables. Ablation studies were performed to determine which dose region and MRI sequence contained radiomic features most predictive for DBM development using machine learning (ML) classifiers. An ML classifier trained on 3D radiomic features from the 50-75% dose region of pre-treatment T1w MRI (AUC: 0.71, 95% CI: 0.68-0.74) outperformed the baseline model (AUC: 0.50, 95% CI: 0.47-0.53) for DBM prediction. In conclusion, we leverage radiotherapy dose regions to identify subcompartments for radiomic feature extraction from multi-parametric pre-treatment MRI data. We demonstrate that radiomic features from these dose regions can be used to predict DBM for SRS-treated brain metastases.

Objectives To differentiate the primary site of brain metastases (BMs) is of high clinical value for the successful management of patients with BM. The purpose of this study is to investigate a combined radiomics model with computer tomography (CT) and magnetic resonance imaging (MRI) images in differentiating BMs originated from lung and breast cancer. Methods Pretreatment cerebral contrast enhanced CT and T1-weighted MRI images of 78 patients with 179 BMs from primary lung and breast cancer were retrospectively analyzed. Radiomic features were extracted from contoured BM lesions and selected using the Mann–Whitney U test and the least absolute shrinkage and selection operator (LASSO) logistic regression. Binary logistic regression (BLR) and support vector machine (SVM) models were built and evaluated based on selected radiomic features from CT alone, MRI alone, and combined images to differentiate BMs originated from lung and breast cancer. Results A total of 10 and 6 optimal radiomic features were screened out of 1288 CT and 1197 MRI features, respectively. The mean area under the curves (AUCs) of the BLR and SVM models using fivefolds cross-validation were 0.703 vs. 0.751, 0.718 vs. 0.754, and 0.781 vs. 0.803 in the training dataset and 0.708 vs. 0.763, 0.715 vs. 0.717, and 0.771 vs. 0.805 in the testing dataset for models with CT alone, MRI alone, and combined CT and MRI radiomic features, respectively. Conclusions Radiomics model based on combined CT and MRI features is feasible and accurate in the differentiation of the primary site of BMs from lung and breast cancer.

ObjectiveThis study aims to develop and evaluate a radiomics-based machine learning model using T1-enhanced magnetic resonance imaging (MRI) features to differentiate between lung squamous cell carcinoma (SCC) and adenocarcinoma (AC) in patients with brain metastases (BMs). While prior studies have largely focused on primary lung tumors, our work uniquely targets metastatic brain lesions, which pose distinct diagnostic and therapeutic challenges.MethodsIn this retrospective study, 173 patients with BMs from lung cancer were included, consisting of 88 with AC and 85 with SCC. MRI images were acquired using a standardized protocol, and 833 radiomic features were identified from the segmented lesions utilizing the PyRadiomics package. Feature selection was performed using a combination of univariate analysis, correlation analysis, and the least absolute shrinkage and selection operator (LASSO) regression. Ten machine learning classifiers were trained and validated utilizing the selected features. The performance of the classifier models was assessed through receiver operating characteristic (ROC) curves, and the area under the curve (AUC) was examined for analysis.ResultsTen classifier models were built on the basis of features derived from MRI. Among the ten classifier models, the LightGBM model performed the best. In the training dataset, the LightGBM classifier achieved an accuracy of 0.814, with a sensitivity of 0.726 and specificity of 0.896. The classifier’s efficiency was validated on an independent testing dataset, where it maintained an accuracy of 0.779, with a sensitivity of 0.725 and specificity of 0.857. The AUC was 0.858 for the training dataset and 0.857 for the testing dataset. The model effectively distinguished between SCC and AC based on radiomic features, highlighting its potential for noninvasive non-small cell lung cancer (NSCLC) subtype classification.ConclusionThis research demonstrates the efficacy of a radiomics-based machine learning model in accurately classifying NSCLC subtypes from BMs, providing a valuable noninvasive tool for guiding personalized treatment strategies. Further validation on larger, multi-center datasets is crucial to verify these findings.

Gamma knife treatment is a clinically effective, safe, and cost-effective adjunctive therapy for primary malignant brain tumors. For most brain metastases, radiosurgery is the treatment of choice and will result in effective tumor control in more than 90% of treated tumors.

Renal Cell Carcinoma: Diagnosis Based on Metastatic Manifestations

Reporting Standards for Percutaneous Thermal Ablation of Renal Cell Carcinoma

Solid renal masses are often indeterminate for benignity versus malignancy on magnetic resonance imaging. Such masses are typically evaluated with either percutaneous biopsy or surgical resection. Percutaneous biopsy can be non-diagnostic and some surgically resected lesions are inadvertently benign. To assess the performance of ten machine learning (ML) algorithms trained with MRI-based radiomics features in distinguishing benign from malignant solid renal masses. Patients with solid renal masses identified on pre-intervention MRI were curated from our institutional database. Masses with a definitive diagnosis via imaging (for angiomyolipomas) or via biopsy or surgical resection (for oncocytomas or renal cell carcinomas) were selected. Each mass was segmented for both T2- and post-contrast T1-weighted images. Radiomics features were derived from the segmented masses for each imaging sequence. Ten ML algorithms were trained with the radiomics features gleaned from each MR sequence, as well as the combination of MR sequences. In total, 182 renal masses in 160 patients were included in the study. The support vector machine algorithm trained on radiomics features from T2-weighted images performed superiorly, with an accuracy of 0.80 and an area under the curve (AUC) of 0.79. Linear discriminant analysis (accuracy = 0.84 and AUC = 0.77) and logistic regression (accuracy = 0.78 and AUC = 0.78) algorithms trained on T2-based radiomics features performed similarly. ML algorithms trained on radiomics features from post-contrast T1-weighted images or the combination of radiomics features from T2- and post-contrast T1-weighted images yielded lower performance. Machine learning models trained with radiomics features derived from T2-weighted images can provide high accuracy for distinguishing benign from malignant solid renal masses. Machine learning models derived from MRI-based radiomics features may improve the clinical management of solid renal masses and have the potential to reduce the frequency with which benign solid renal masses are biopsied or surgically resected.

With the widespread application of machine learning (ML) in the diagnosis and treatment of colorectal cancer (CRC), some studies have investigated the use of ML techniques for the diagnosis of KRAS (Kirsten rat sarcoma) mutation. Nevertheless, there is scarce evidence from evidence-based medicine to substantiate its efficacy. Our study was carried out to systematically review the performance of ML models developed using different modeling approaches, in diagnosing KRAS mutations in CRC. We aim to offer evidence-based foundations for the development and enhancement of future intelligent diagnostic tools. PubMed, Cochrane Library, Embase, and Web of Science were systematically retrieved, with the search cutoff date set to December 22, 2024. The encompassed studies are publicly published research papers that use ML to diagnose KRAS gene mutations in CRC. The risk of bias in the encompassed models was evaluated via the PROBAST (Prediction Model Risk of Bias Assessment Tool). A meta-analysis of the model's concordance index (c-index) was performed, and a bivariate mixed-effects model was used to summarize sensitivity and specificity based on diagnostic contingency tables. A total of 43 studies involving 10,888 patients were included. The modeling variables were derived from clinical characteristics, computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography/computed tomography, and pathological histology. In the validation cohort, for the ML model developed based on CT radiomic features, the c-index, sensitivity, and specificity were 0.87 (95% CI 0.84-0.90), 0.85 (95% CI 0.80-0.89), and 0.83 (95% CI 0.73-0.89), respectively. For the model developed using MRI radiomic features, the c-index, sensitivity, and specificity were 0.77 (95% CI 0.71-0.83), 0.78 (95% CI 0.72-0.83), and 0.73 (95% CI 0.63-0.81), respectively. For the ML model developed based on positron emission tomography/computed tomography radiomic features, the c-index, sensitivity, and specificity were 0.84 (95% CI 0.77-0.90), 0.73, and 0.83, respectively. Notably, the deep learning (DL) model based on pathological images demonstrated a c-index, sensitivity, and specificity of 0.96 (95% CI 0.94-0.98), 0.83 (95% CI 0.72-0.91), and 0.87 (95% CI 0.77-0.92), respectively. The DL model MRI-based model showed a c-index of 0.93 (95% CI 0.90-0.96), sensitivity of 0.85 (95% CI 0.75-0.91), and specificity of 0.83 (95% CI 0.77-0.88). ML is highly accurate in diagnosing KRAS mutations in CRC, and DL models based on MRI and pathological images exhibit particularly strong diagnosis accuracy. More broadly applicable DL-based diagnostic tools may be developed in the future. However, the clinical application of DL models remains relatively limited at present. Therefore, future research should focus on increasing sample sizes, improving model architectures, and developing more advanced DL models to facilitate the creation of highly efficient intelligent diagnostic tools for KRAS mutation diagnosis in CRC.

To investigate the optimal treatment and prognostic factors of renal cell carcinoma with brain metastases (RCCBM) by the analysis of clinical features. The clinical data of 10 patients with RCCBM in our hospital from Jul. 2003 to Aug. 2011 were analyzed retrospectively. The age range of the patients was 48-80 years, and the mean age was 64 years old. Nine patients were male and one patient was female. Six patients had neurological symptoms. Brain metastasis and kidney cancer were found in the same period in 7 patients. Two patients had brain metastases alone, and 8 patients had metastasis in brain and other parts. There were 4 patients had more than 3 brain metastatic foci. Brain metastatic foci located in: frontal lobe in 6 patients, parietal lobe in 2 patients, occipital lobe in 2 patients, cerebellum in 2 patients, temporal lobe in 1 patient, thalamus and basal ganglia in 1 patient. The overall survival (OS) of 10 patients with RCCBM was 0.8 to 62.6 months, with an average OS of 12.7 months. Two patients refused therapy and those average OS is only 1.4 months. Three patients were performed stereotactic radiotherapy (SRS). The tumor volume of 1 patient reduced >25% and 2 patients showed no progression in magnetic resonance images at 2-months after SRS. Two patients accepted sorafenib treatment and the average OS was 16.7 months. Patients with 3 or more brain metastases had an average OS of 1.9 months, and the average OS of those with less than 3 brain metastases was 20 months. Patients with synchronous or metachronous brain metastases of renal cell carcinoma had an average OS of 6.8 or 26.5 months, respectively. Patients with or without neurological symptoms had an average OS of 17.9 or 4.9 months, respectively. For RCCBM patients, SRS could effectively control the local brain metastases while sorafenib could extend the overall survival. With 3 or more metastases, synchronous brain metastases and no neurological symptoms were the adverse prognostic factors of RCCBM.

Background and Aims Interstitial fibrosis / tubular atrophy (IFTA) is a common, irreversible and progressive form of chronic allograft injury, and it is considered a critical predictor of kidney allograft outcomes. Inflammation, both microvascular and interstitial, is on the contrary regarded as a reversible form of graft injury. Since treatments for rejection and other causes of graft dysfunction bear substantial toxicity and could have limited efficacy, the extent of irreversible graft scarring is a crucial information for the clinician, to evaluate risks and benefits of specific therapies. The diagnosis of kidney graft pathology is acquired through graft biopsy, which is an invasive procedure and can be subjected to sampling bias. Magnetic resonance imaging (MRI), especially with functional techniques, has emerged as a possibility for non-invasive estimation of tissue fibrosis; nevertheless, functional MRI is not widely available. Texture analysis MRI (TA-MRI) is a radiomic technique that provides a quantitative assessment of tissue heterogeneity from standard MRI images, generating features that can be fitted into a machine-learning model to assess their ability to predict clinical or histological parameters. Method Single-center cross-sectional observational cohort study enrolling 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. The study was approved by the local Ethical Committee (AOU0010167/20). The primary outcome was to identify the best TA-MRI features subset for estimation of IFTA > 50% in graft biopsy. Secondary outcomes were estimation of: IFTA > 25%, presence of total inflammation (ti) and microvascular inflammation (glomerulitis + peritubular capillaritis [g+ptc]). Graft biopsy was reported according to Banff 2017 system. Radiomic analysis was performed on axial T2 pre-contrast and T1 fat-suppressed post-contrast sequences. The whole renal parenchyma (PAR) was segmented and labelled on T2 and T1, renal cortex (COR) only on T2. After imaging preprocessing, PyRadiomics was used to extract radiomic features. After removal of shape features, 93 features were included and reduced using LASSO regression to produce radiomic signatures. These were introduced in Machine Learning (ML) models to test the association with outcomes. Results are reported as AUC and a value of sensitivity and specificity. Results Sixty patients were included in the study, and 67 graft biopsy – graft MRI pairs were available for analysis. Demographic and clinical characteristics of enrolled patients are depicted in table 1; histological diagnosis and main Banff histological parameters from graft biopsies in table 2. Among ML models, three showed an acceptable performance. T2 COR “firstorder_minimum/firstorder_range/glrlm_run_entropy” for IFTA>50% (AUC=0.77, sensitivity=73%, specificity=71%), T1 PAR “firstorder_energy” for IFTA>25% (AUC=0.71, sensitivity=74%, specificity=51%), T1 PAR “firstorder_energy/gldm_small_dependence_low_gray_level_emphasis” for g+ptc >0 (AUC=0.74, sensitivity= 78%, specificity=68%); see figures 1–3. No acceptable prediction was detected for ti >0. Conclusion Our study shows that TA-MRI feature signatures can predict the degree of IFTA in graft biopsies, with an acceptable diagnostic performance. These results suggest to further investigating TA-MRI from standard MRI sequences as potential tool to assess graft chronic parenchymal injury. Moreover, since graft biopsy results can be jeopardized by limited sample size, we hypothesize that evaluation of IFTA through TA-MRI could provide more comprehensive information regarding the whole parenchyma. To test this hypothesis, we are currently evaluating the association of TA-MRI radiomic features and baseline eGFR and eGFR variation over time.

To evaluate radiomics features' reproducibility using inter-package/inter-observer measurement analysis in renal masses (RMs) based on MRI and to employ machine learning (ML) models for RM characterization. 32 Patients (23M/9F; age 61.8 ± 10.6years) with RMs (25 renal cell carcinomas (RCC)/7 benign masses; mean size, 3.43 ± 1.73cm) undergoing resection were prospectively recruited. All patients underwent 1.5T MRI with T2-weighted (T2-WI), diffusion-weighted (DWI)/apparent diffusion coefficient (ADC), and pre-/post-contrast-enhanced T1-weighted imaging (T1-WI). RMs were manually segmented using volume of interest (VOI) on T2-WI, DWI/ADC, and T1-WI pre-/post-contrast imaging (1-min, 3-min post-injection) by two independent observers using two radiomics software packages for inter-package and inter-observer assessments of shape/histogram/texture features common to both packages (104 features; n = 26 patients). Intra-class correlation coefficients (ICCs) were calculated to assess inter-observer and inter-package reproducibility of radiomics measurements [good (ICC ≥ 0.8)/moderate (ICC = 0.5-0.8)/poor (ICC < 0.5)]. ML models were employed using reproducible features (between observers and packages, ICC > 0.8) to distinguish RCC from benign RM. Inter-package comparisons demonstrated that radiomics features from T1-WI-post-contrast had the highest proportion of good/moderate ICCs (54.8-58.6% for T1-WI-1min), while most features extracted from T2-WI, T1-WI-pre-contrast, and ADC exhibited poor ICCs. Inter-observer comparisons found that radiomics measurements from T1-WI pre/post-contrast and T2-WI had the greatest proportion of features with good/moderate ICCs (95.3-99.1% T1-WI-post-contrast 1-min), while ADC measurements yielded mostly poor ICCs. ML models generated an AUC of 0.71 [95% confidence interval = 0.67-0.75] for diagnosis of RCC vs. benign RM. Radiomics features extracted from T1-WI-post-contrast demonstrated greater inter-package and inter-observer reproducibility compared to ADC, with fair accuracy for distinguishing RCC from benign RM. Knowledge of reproducibility of MRI radiomics features obtained on renal masses will aid in future study design and may enhance the diagnostic utility of radiomics models for renal mass characterization.

BACKGROUND Recently, non-invasive characterization of brain tumors on MRI has emerged as a promising field of research. The identification of quantitative imaging biomarkers, also known as radiomics, may complement molecular characterization and thereby improve clinical management of neuro-oncologic patients. In this study we aimed to identify imaging predictors with improved performance over clinical parameters in order to stratify patients with brain metastases into high and low risk groups for overall survival (OS). MATERIAL AND METHODS 422 patients (allocated in a 3:1 ratio to a discovery [n=317] and test [n= 105] set) with first diagnosis of brain metastases from different primary tumors from two neuro-oncologic centers were included. In each patient, eight clinical features (age, gender, KPS, systemic disease status, presence of extracranial metastases, number of cerebral metastases, primary tumor and available individual prognostic molecular status eg HER2, BRAF,⋯) were gathered and a total of 321 radiomic MRI features (including shape, first-order and higher-order features) from cerebral MRI (contrast T1-weighted and apparent diffusion coefficient maps) were extracted. Radiomic and clinical features of patients in the discovery set were subjected to different machine learning models in order to classify patients into low- and high-risk groups for OS. By performing a comprehensive grid search the best machine learning model according to the macro-evaluated score and accuracy was identified and evaluated on the testing set. In addition, a subgroup analysis, based on the primary tumor entity was done and confusion matrices were calculated in order to evaluate final predictions. RESULTS With an extra trees classifier including all clinical and 30 radiomic features, we were able to stratify patients into a high- and low-risk group for OS (test set: macro-evaluated = 0.60, accuracy = 0.67). The best performing model was a gradient boosting model only including clinical features (test set: macro-evaluated = 0.62, accuracy = 0.72), while the radiomic features alone led to the poorest results (test set: macro-evaluated = 0.52, accuracy = 0.63). Interestingly, in the subgroup of melanoma patients, the predictive power of radiomic features outperformed the clinical and the combined (radiomics and clinical) features. With a prediction solely based on radiomic features, 80% and 67% of patients were correctly classified into the low- and the high-risk group for OS, respectively. CONCLUSION In conclusion, we found that in the entire study population of patients with brain metastases radiomic features did not allow for a better prediction of the clinical outcome compared to the clinical parameters alone. However, in the subgroup of melanoma patients the predictive power of radiomic features alone was superior compared to clinical features or all features combined.

Leveraging Serial MRI Radiomics and Machine Learning to Predict Risk of Radiation Necrosis in Patients with Brain Metastases Managed with Stereotactic Radiation and Immunotherapy