Radiomics-Based Machine Learning Models Improve Acute Pancreatitis Severity Prediction

BackgroundNeoadjuvant immunotherapy has been shown to improve survival in patients with gastric cancer. This study sought to develop and validate a radiomics-based machine learning (ML) model for patients with locally advanced gastric cancer (LAGC), specifically to predict whether patients will achieve a major pathological response (MPR) following neoadjuvant immunotherapy. With its predictive capabilities, this tool shows promise for enhancing clinical decision-making processes in the future.MethodsThis study utilized a multicenter cohort design, retrospectively gathering clinical data and computed tomography (CT) images from 268 patients diagnosed with advanced gastric cancer who underwent neoadjuvant immunotherapy between January 2019 and December 2023 from two medical centers. Radiomic features were extracted from CT images, and a multi-step feature selection procedure was applied to identify the top 20 representative features. Nine ML algorithms were implemented to build prediction models, with the optimal algorithm selected for the final prediction model. The hyperparameters of the chosen model were fine-tuned using Bayesian optimization and grid search. The performance of the model was evaluated using several metrics, including the area under the curve (AUC), accuracy, and Cohen’s kappa coefficient.ResultsThree cohorts were included in this study: the development cohort (DC, n = 86), the internal validation cohort (IVC, n = 59), and the external validation cohort (EVC, n = 52). Nine ML models were developed using DC cases. Among these, an optimized Bayesian-LightGBM model, demonstrated robust predictive performance for MPR following neoadjuvant immunotherapy in LAGC patients across all cohorts. Specifically, within DC, the LightGBM model attained an AUC of 0.828, an overall accuracy of 0.791, a Cohen’s kappa coefficient of 0.552, a sensitivity of 0.742, a specificity of 0.818, a positive predictive value (PPV) of 0.586, a negative predictive value (NPV) of 0.867, a Matthews correlation coefficient (MCC) of 0.473, and a balanced accuracy of 0.780. Comparable performance metrics were validated in both the IVC and the EVC, with AUC values of 0.777 and 0.714, and overall accuracies of 0.729 and 0.654, respectively. These results suggested good fitness and generalization of the Bayesian-LightGBM model. Shapley Additive Explanations (SHAP) analysis identified significant radiomic features contributing to the model’s predictive capability. The SHAP values of the features wavelet.LLH_gldm_SmallDependenceLowGrayLevelEmphasis, wavelet.HHL_glrlm_RunVariance, and wavelet.LLH_glszm_LargeAreaHighGrayLevelEmphasis were ranked among the top three, highlighting their significant contribution to the model’s predictive performance. In contrast to existing radiomic models that exclusively focus on neoadjuvant chemotherapy, our model integrates both neoadjuvant immunotherapy and chemotherapy, thereby offering more precise predictive capabilities.ConclusionThe radiomics-based ML model demonstrated significant efficacy in predicting the pathological response to neoadjuvant immunotherapy in LAGC patients, thereby providing a foundation for personalized treatment strategies.

Medulloblastomas are a major cause of cancer-related mortality in the pediatric population. Four molecular groups have been identified, and these molecular groups drive risk stratification, prognostic modeling, and the development of novel treatment modalities. It has been demonstrated that radiomics-based machine learning (ML) models are effective at predicting the diagnosis, molecular class, and grades of CNS tumors. To assess radiomics-based ML models' diagnostic performance in predicting medulloblastoma subgroups and the methodological quality of the studies. A comprehensive literature search was performed on PubMed; the last search was conducted on 1 May 2022. Studies that predicted all four medulloblastoma subgroups in patients with histopathologically confirmed medulloblastoma and reporting area under the curve (AUC) values were included in the study. The quality assessments were conducted according to the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) and Checklist for Artificial Intelligence in Medical Imaging (CLAIM). A meta-analysis of radiomics-based ML studies' diagnostic performance for the preoperative evaluation of medulloblastoma subgrouping was performed. Five studies were included in this meta-analysis. Regarding patient selection, two studies indicated an unclear risk of bias according to the QUADAS-2. The five studies had an average CLAIM score and compliance score of 23.2 and 0.57, respectively. The meta-analysis showed pooled AUCs of 0.88, 0.82, 0.83, and 0.88 for WNT, SHH, group 3, and group 4 for classification, respectively. Radiomics-based ML studies have good classification performance in predicting medulloblastoma subgroups, with AUCs >0.80 in every subgroup. To be applied to clinical practice, they need methodological quality improvement and stability.

Digital imaging

In the era of precision therapy, early classification of breast cancer (BRCA) molecular subtypes has clinical significance for disease management and prognosis. We explored the accuracy of machine learning (ML) models for early classification of BRCA molecular subtypes through a systematic review of the literature currently available. We retrieved relevant studies published in PubMed, EMBASE, Cochrane, and Web of Science until 15 April 2022. A prediction model risk of bias assessment tool (PROBAST) was applied for the assessment of risk of bias of a genomics-based ML model, and the Radiomics Quality Score (RQS) was simultaneously used to evaluate the quality of this radiomics-based ML model. A random effects model was adopted to analyze the predictive accuracy of genomics-based ML and radiomics-based ML for Luminal A, Luminal B, Basal-like or triple-negative breast cancer (TNBC), and human epidermal growth factor receptor 2 (HER2). The PROSPERO of our study was prospectively registered (CRD42022333611). Of the 38 studies were selected for analysis, 14 ML models were based on gene-transcriptomic, with only 4 external validations; and 43 ML models were based on radiomics, with only 14 external validations. Meta-analysis results showed that c-statistic values of the ML based on radiomics for the identification of BRCA molecular subtypes Luminal A, Luminal B, Basal-like or TNBC, and HER2 were 0.76 [95% confidence interval (CI): 0.60-0.96], 0.78 (95% CI: 0.69-0.87), 0.89 (95% CI: 0.83-0.91), and 0.83 (95% CI: 0.81-0.86), respectively. The c-statistic values of ML based on the gene-transcriptomic analysis cohort for the identification of the previously described BRCA molecular subtypes were 0.96 (95% CI: 0.93-0.99), 0.96 (95% CI: 0.93-0.99), 0.98 (95% CI: 0.95-1.00), and 0.97 (95% CI: 0.96-0.98) respectively. Additionally, the sensitivity of the ML model based on radiomics for each molecular subtype ranged from 0.79 to 0.85, while the sensitivity of the ML model based on gene-transcriptomic was between 0.92 and 0.99. Both radiomics and gene transcriptomics produced ideal effects on BRCA molecular subtype prediction. Compared with radiomics, gene transcriptomics yielded better prediction results, but radiomics was simpler and more convenient from a clinical point of view.

In this study, we aimed to investigate the classification of symptomatic plaques by evaluating the models generated via two different approaches, a radiomics-based machine learning (ML) approach, and an end-to-end learning approach which utilized deep learning (DL) techniques with several representative model frameworks. We collected high-resolution magnetic resonance imaging (HRMRI) data from 104 patients with carotid artery stenosis, who were diagnosed with either symptomatic plaques (SPs) or asymptomatic plaques (ASPs), in two medical centers. 74 patients were diagnosed with SPs and 30 patients were ASPs. Sampling Perfection with Application-optimized Contrasts (SPACE) by using different flip angle Evolutions was used for MRI imaging. Repeated stratified five-fold cross-validation was used to evaluate the accuracy and receiver operating characteristic (ROC) of the trained classifier. The two proposed approaches were investigated to train the models separately. The difference in the model performance of the two proposed methods was quantitatively evaluated to find a better model to differentiate between SPs and ASPs. 3D-SE-Densenet-121 model showed the best performance among all prediction models (AUC, accuracy, precision, sensitivity, and F1-score of 0.9300, 0.9308, 0.9008, 0.8588, and 0.8614, respectively), which were 0.0689, 0.1119, 0.1043, 0.0805, and 0.1089 higher than the best radiomics-based ML model (MLP). Decision curve analysis showed that the 3D-SE-Densenet-121 model delivered more net benefit than the best radiomics-based ML model (MLP) with a wider threshold probability. The DL models were able to accurately differentiate between symptomatic and asymptomatic carotid plaques with limited data, which outperformed radiomics-based ML models in identifying symptomatic plaques.

ObjectivesTo develop a computed tomography (CT) radiomics-based interpretable machine learning (ML) model to predict the pathological grade of pancreatic neuroendocrine tumors (pNETs) in a non-invasive manner.MethodsPatients with pNETs who underwent contrast-enhanced abdominal CT between 2010 and 2022 were included in this retrospective study. Radiomics features were extracted, and five radiomics-based ML models, namely logistic regression (LR), random forest (RF), support vector machine (SVM), XGBoost, and GaussianNB, were developed. The performance of these models was evaluated using a time-independent testing set, and metrics such as sensitivity, specificity, accuracy, and the area under the receiver operating characteristic curve (AUC) were calculated. The accuracy of the radiomics model was compared to that of needle biopsy. The Shapley Additive Explanation (SHAP) tool and the correlation between radiomics and biological features were employed to explore the interpretability of the model.ResultsA total of 122 patients (mean age: 50 ± 14 years; 53 male) were included in the training set, whereas 100 patients (mean age: 48 ± 13 years; 50 male) were included in the testing set. The AUCs for LR, SVM, RF, XGBoost, and GaussianNB were 0.758, 0.742, 0.779, 0.744, and 0.745, respectively, with corresponding accuracies of 73.0%, 70.0%, 77.0%, 71.9%, and 72.9%. The SHAP tool identified two features of the venous phase as the most significant, which showed significant differences among the Ki-67 index or mitotic count subgroups (p < 0.001).ConclusionsAn interpretable radiomics-based RF model can effectively differentiate between G1 and G2/3 of pNETs, demonstrating favorable interpretability.Clinical relevance statementThe radiomics-based interpretable model developed in this study has significant clinical relevance as it offers a non-invasive method for assessing the pathological grade of pancreatic neuroendocrine tumors and holds promise as an important complementary tool to traditional tissue biopsy.Key Points• A radiomics-based interpretable model was developed to predict the pathological grade of pNETs and compared with preoperative needle biopsy in terms of accuracy.• The model, based on CT radiomics, demonstrated favorable interpretability.• The radiomics model holds potential as a valuable complementary technique to preoperative needle biopsy; however, it should not be considered a replacement for biopsy.

Radiomics and machine learning have been extensively utilized in the realm of urinary stones, particularly in forecasting stone treatment outcomes. The objective of this study was to integrate clinical variables and radiomic features to develop a machine learning model for predicting the stone-free rate (SFR) following percutaneous nephrolithotomy (PCNL).A total of 212 eligible patients who underwent PCNL surgery at the Second Affiliated Hospital of Nanchang University were included in a retrospective analysis. Preoperative clinical variables and non-contrast-enhanced CT images of all patients were collected, and radiomic features were extracted after delineating the stone ROI. Univariate analysis was conducted to identify clinical variables strongly correlated with the stone-free rate after PCNL, and the least absolute shrinkage and selection operator algorithm (lasso regression) was utilized to screen radiomic features. Four supervised machine learning algorithms, including Logistic Regression, Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Gradient Boosting Decision Tree (GBDT), were employed. The clinical variables with strong correlation and screened radiomic features were integrated into the four machine learning algorithms to construct a prediction model, and the receiver operating curve was plotted. The area under the receiver operating curve (AUC), the accuracy rate, the specificity, etc., were used to evaluate the predictive performance of the four models.After analyzing postoperative statistics, the stone-free rate following the procedure was found to be 70.3% (n = 149). Among the various clinical variables examined, factors, such as stone number, stone diameter, stone CT value, stone location, and history of stone surgery, were identified as statistically significant in relation to the stone-free rate after PCNL. A total of 121 radiomic features were extracted, and through lasso regression, 7 features most closely associated with the stone-free rate post-PCNL were identified. The predictive accuracy of different models (Logistic Regression, RF, XGBoost, and GBDT) for determining the stone-free rate after PCNL was evaluated, yielding accuracies of 78.1%, 76.6%, 75.0%, and 73.4%, respectively. The corresponding area under the curve AUC (95%CI) were 0.85 (0.83-0.89), 0.81 (0.76-0.85), 0.82 (0.78-0.85), and 0.77 (0.73-0.81), positioning these models among the top performers in logistic regression prediction. In terms of predictive importance scores, the key factors identified by the logistic regression model were number of stone, zone percentage, stone diameter, and surface area. Similarly, the RF model highlighted number of stone, stone CT value, stone diameter, and surface area as the top predictors.Among the four machine learning models, the logistic regression model demonstrated the highest accuracy and discrimination ability in predicting the stone-free rate following PCNL. In comparison to XGBoost and GBDT, RF also exhibited superior accuracy and a certain level of discrimination ability. However, based on the performance of all four models, logistic regression is more likely to aid in clinical decision-making by assisting clinicians in diagnosing PCNL in patients. This enables us to effectively predict the presence of residual stones post-surgery and ultimately select patients who are suitable candidates for PCNL.

Background To assess the feasibility of a machine learning (ML) approach using radiomics features of perfusion defects on rest myocardial perfusion imaging (MPI) to detect the presence of hibernating myocardium. Methodology Data of patients who underwent 99mTc-sestamibi MPI and 18F-FDG PET/CT for myocardial viability assessment were retrieved. Rest MPI data were processed on ECToolbox, and polar maps were saved using theNFile PMap tool. The reference standard for defining hibernating myocardium was the presence of mismatched perfusion-metabolism defect with impaired myocardial contractility at rest. Perfusion defects on the polar maps were delineated with regions of interest (ROIs) after spatial resampling and intensity discretization. Replicable random sampling allocated 80% (257) of the perfusion defects of the patients from January 2017 to September 2022 to the training set and the remaining 20% (64) to the validation set. An independent dataset of perfusion defects from 29 consecutive patients from October 2022 to January 2023 was used as the testing set for model evaluation. One hundred ten first and second-order texture features were extracted for each ROI. After feature normalization and imputation, 14 best-ranked features were selected using a multistep feature selection process including the Logistic Regression and Fast Correlation-Based Filter. Thirteen supervised ML algorithms were trained with stratified five-fold cross-validation on the training set and validated on the validation set. The ML algorithms with a Log Loss of <0.688 and <0.672 in the cross-validation and validation steps were evaluated on the testing set. Performance matrices of the algorithms assessed included area under the curve (AUC), classification accuracy (CA), F1 score, precision, recall, and specificity. To provide transparency and interpretability, SHapley Additive exPlanations (SHAP) values were assessed and depicted as beeswarm plots. Results Two hundred thirty-nine patients (214 males; mean age 56 ± 11 years) were enrolled in the study. There were 371 perfusion defects (321 in the training and validation sets; 50 in the testing set). Based on the reference standard, 168 perfusion defects had hibernating myocardium (139 in the training and validation sets; 29 in the testing set). On cross-validation, six ML algorithms with Log Loss <0.688 had AUC >0.800. On validation, 10 ML algorithms had a Log Loss value <0.672, among which six had AUC >0.800. On model evaluation of the selected models on the unseen testing set, nine ML models had AUC >0.800 with Gradient Boosting Random Forest (xgboost) [GB RF (xgboost)] achieving the highest AUC of 0.860 and could detect the presence of hibernating myocardium in 21/29 (72.4%) perfusion defects with a precision of 87.5% (21/24), specificity 85.7% (18/21), CA 78.0% (39/50) and F1 Score 0.792. Four models depicted a clear pattern of model interpretability based on the beeswarm SHAP plots. These were GB RF (xgboost), GB (scikit-learn), GB (xgboost), and Random Forest. Conclusion Our study demonstrates the potential of ML in detecting hibernating myocardium using radiomics features extracted from perfusion defects on rest MPI images. This proof-of-concept underscores the notion that radiomics features capture nuanced information beyond what is perceptible to the human eye, offering promising avenues for improved myocardial viability assessment.

BackgroundThe objective of this study was to develop and validate a radiomics-based machine learning (ML) model to differentiate between renal medullary carcinoma (RMC) and clear cell renal carcinoma (ccRCC).MethodsThis retrospective Institutional Review Board -approved study analyzed CT images and clinical data from patients with RMC (n = 87) and ccRCC (n = 93). Patients without contrast-enhanced CT scans obtained before nephrectomy were excluded. A standard volumetric software package (MIM 7.1.4, MIM Software Inc.) was used for contouring, after which 949 radiomics features were extracted with PyRadiomics 3.1.0. Radiomics analysis was then performed with RadAR for differential radiomics analysis. ML was then performed with extreme gradient boosting (XGBoost 2.0.3) to differentiate between RMC and ccRCC. Three separate ML models were created to differentiate between ccRCC and RMC. These models were based on clinical demographics, radiomics, and radiomics incorporating hemoglobin electrophoresis for sickle cell trait, respectively.ResultsPerformance metrics for the 3 developed ML models were as follows: demographic factors only (AUC = 0.777), calibrated radiomics (AUC = 0.915), and calibrated radiomics with sickle cell trait incorporated (AUC = 1.0). The top 4 ranked features from differential radiomic analysis, ranked by their importance, were run entropy (preprocessing filter = original, AUC = 0.67), dependence entropy (preprocessing filter = wavelet, AUC = 0.67), zone entropy (preprocessing filter = original, AUC = 0.67), and dependence entropy (preprocessing filter = original, AUC = 0.66).ConclusionA radiomics-based machine learning model effectively differentiates between ccRCC and RMC. This tool can facilitate the radiologist’s ability to suspicion and decrease the misdiagnosis rate of RMC.

Aim: Machine learning (ML) and deep learning (DL) predictive models have been employed widely in clinical settings. Their potential support and aid to the clinician of providing an objective measure that can be shared among different centers enables the possibility of building more robust multicentric studies. This study aimed to propose a user-friendly and low-cost tool for COVID-19 mortality prediction using both an ML and a DL approach. Method: We enrolled 2348 patients from several hospitals in the Province of Reggio Emilia. Overall, 19 clinical features were provided by the Radiology Units of Azienda USL-IRCCS of Reggio Emilia, and 5892 radiomic features were extracted from each COVID-19 patient’s high-resolution computed tomography. We built and trained two classifiers to predict COVID-19 mortality: a machine learning algorithm, or support vector machine (SVM), and a deep learning model, or feedforward neural network (FNN). In order to evaluate the impact of the different feature sets on the final performance of the classifiers, we repeated the training session three times, first using only clinical features, then employing only radiomic features, and finally combining both information. Results: We obtained similar performances for both the machine learning and deep learning algorithms, with the best area under the receiver operating characteristic (ROC) curve, or AUC, obtained exploiting both clinical and radiomic information: 0.803 for the machine learning model and 0.864 for the deep learning model. Conclusions: Our work, performed on large and heterogeneous datasets (i.e., data from different CT scanners), confirms the results obtained in the recent literature. Such algorithms have the potential to be included in a clinical practice framework since they can not only be applied to COVID-19 mortality prediction but also to other classification problems such as diabetic prediction, asthma prediction, and cancer metastases prediction. Our study proves that the lesion’s inhomogeneity depicted by radiomic features combined with clinical information is relevant for COVID-19 mortality prediction.

Differentiating benign gallbladder diseases from gallbladder cancer (GBC) remains a radiological challenge because they can appear very similar on imaging. This study aimed at investigating whether CT-based radiomic features of suspicious gallbladder lesions analyzed by machine learning algorithms could adequately discriminate benign gallbladder disease from GBC. In addition, the added value of machine learning models to radiological visual CT-scan interpretation was assessed. Patients were retrospectively selected based on confirmed histopathological diagnosis and available contrast-enhanced portal venous phase CT-scan. The radiomic features were extracted from the entire gallbladder, then further analyzed by machine learning classifiers based on Lasso regression, Ridge regression, and XG Boosting. The results of the best-performing classifier were combined with radiological visual CT diagnosis and then compared with radiological visual CT assessment alone. In total, 127 patients were included: 83 patients with benign gallbladder lesions and 44 patients with GBC. Among all machine learning classifiers, XG boosting achieved the best AUC of 0.81 (95% CI 0.72-0.91) and the highest accuracy rate of 73% (95% CI 65-80%). When combining radiological visual interpretation and predictions of the XG boosting classifier, the highest diagnostic performance was achieved with an AUC of 0.98 (95% CI 0.96-1.00), a sensitivity of 91% (95% CI 86-100%), a specificity of 93% (95% CI 90-100%), and an accuracy of 92% (95% CI 90-100%). Machine learning analysis of CT-based radiomic features shows promising results in discriminating benign from malignant gallbladder disease. Combining CT-based radiomic analysis and radiological visual interpretation provided the most optimal strategy for GBC and benign gallbladder disease differentiation. Radiomic-based machine learning algorithms are able to differentiate benign gallbladder disease from gallbladder cancer. Combining machine learning algorithms with a radiological visual interpretation of gallbladder lesions at CT increases the specificity, compared to visual interpretation alone, from 73 to 93% and the accuracy from 85 to 92%. Combined use of machine learning algorithms and radiological visual assessment seems the most optimal strategy for GBC and benign gallbladder disease differentiation.

ObjectiveThe indocyanine green retention rate at 15 min (ICG-R15) is a useful tool to evaluate the functional liver reserve before hepatectomy for liver cancer. Taking ICG-R15 as criteria, we investigated the ability of a machine learning (ML)-based radiomics model produced by Gd-EOB-DTPA-enhanced hepatic magnetic resonance imaging (MRI) or contrast-enhanced computed tomography (CT) image in evaluating functional liver reserve of hepatocellular carcinoma (HCC) patients.MethodsA total of 190 HCC patients with CT, among whom 112 also with MR, were retrospectively enrolled and randomly classified into a training dataset (CT: n = 133, MR: n = 78) and a test dataset (CT: n = 57, MR: n = 34). Then, radiomics features from Gd-EOB-DTPA MRI and CT images were extracted. The features associated with the ICG-R15 classification were selected. Five ML classifiers were used for the ML-model investigation. The accuracy (ACC) and the area under curve (AUC) of receiver operating characteristic (ROC) with 95% confidence intervals (CI) were utilized for ML-model performance evaluation.ResultsA total of 107 different radiomics features were extracted from MRI and CT, respectively. The features related to ICG-R15 which was classified into 10%, 20% and 30% were selected. In MRI groups, classifier XGBoost performed best with its AUC = 0.917 and ACC = 0.882 when the threshold was set as ICG-R15 = 10%. When ICG-R15 = 20%, classifier Random Forest performed best with AUC = 0.979 and ACC = 0.882. When ICG-R15 = 30%, classifier XGBoost performed best with AUC = 0.961 and ACC = 0.941. For CT groups, the classifier XGBoost performed best when ICG-R15 = 10% with AUC = 0.822 and ACC = 0.842. When ICG-R15 = 20%, classifier SVM performed best with AUC = 0.860 and ACC = 0.842. When ICG-R15 = 30%, classifier XGBoost performed best with AUC = 0.938 and ACC = 0.965.ConclusionsBoth the MRI- and CT-based machine learning models are proved to be valuable noninvasive methods for functional liver reserve evaluation.

Despite the promise for performance and accuracy improvements of quantum inspired (QI) algorithms over classical machine learning (ML) algorithms, such gains have not been realized in practice. The quantum inspired algorithms can theoretically achieve significant speed up based on sampling assumptions and have thus far failed to outperform the existing classical ML models in practical applications. The speedup of quantum machine learning (QML) algorithms assume the access to data in quantum random access memory (QRAM) which is a strong assumption with current quantum architectures. QI algorithms assume sample and query (SQ) access to input vector and norms of matrices using a dynamic data structure. We explore the components of these models and the assumptions in this paper by surveying the recent works in QML and QI Machine learning (QIML) algorithms. We limit our study to QML and QIML models on achieving a speed up over classical ML techniques rather than individual proofs of these algorithms. This study highlights the assumptions being made that are currently not practical for QML and QIML algorithms in achieving performance advantage over classical ML algorithms.

Summary A meticulous interpretation of steady-state or unsteady-state relative permeability (Kr) experimental data is required to determine a complete set of Kr curves. In this work, different machine learning (ML) models were developed to assist in a faster estimation of these curves from steady-state drainage coreflooding experimental runs. These ML algorithms include gradient boosting (GB), random forest (RF), extreme gradient boosting (XGB), and deep neural network (DNN) with a main focus on and comparison of the two latter algorithms (XGB and DNN). Based on existing mathematical models, a leading-edge framework was developed where a large database of Kr and capillary pressure (Pc) curves were generated. This database was used to perform thousands of coreflood simulation runs representing oil-water drainage steady-state experiments. The results obtained from these simulation runs, mainly pressure drop along with other conventional core analysis data, were used to estimate analytical Kr curves based on Darcy’s law. These analytically estimated Kr curves along with the previously generated Pc curves were fed as features into the ML model. The entire data set was split into 80% for training and 20% for testing. The k-fold cross-validation technique was applied to increase the model’s accuracy by splitting 80% of the training data into 10 folds. In this manner, for each of the 10 experiments, nine folds were used for training and the remaining fold was used for model validation. Once the model was trained and validated, it was subjected to blind testing on the remaining 20% of the data set. The ML model learns to capture fluid flow behavior inside the core from the training data set. In terms of applicability of these ML models, two sets of experimental data were needed as input; the first was the analytically estimated Kr curves from the steady-state drainage coreflooding experiments, while the other was the Pc curves estimated from centrifuge or mercury injection capillary pressure (MICP) measurements. The trained/tested model was then able to estimate Kr curves based on the experimental results fed as input. Furthermore, to test the performance of the ML model when only one set of experimental data is available to an end user, a recurrent neural network (RNN) algorithm was trained/tested to predict Kr curves in the absence of Pc curves as an input. The performance of the three developed models (XGB, DNN, and RNN) was assessed using the values of the coefficient of determination (R2) along with the loss calculated during training/validation of the model. The respective crossplots along with comparisons of ground truth vs. artificial intelligence (AI)-predicted curves indicated that the model is capable of making accurate predictions with an error percentage between 0.2% and 0.6% on history-matching experimental data for all three tested ML techniques. This implies that the AI-based model exhibits better efficiency and reliability in determining Kr curves when compared to conventional methods. The developed ML models by no means replace the need to conduct drainage coreflooding or centrifuge experiments but act as an alternative to existing commercial platforms that are used to interpret experimental data to predict Kr curves. The two main advantages of the developed ML models are their capability of predicting Kr curves within a matter of a few minutes as well as with limited intervention from the end user. The results also include a comparison between classical ML approaches, shallow neural networks, and DNNs in terms of accuracy in predicting the final Kr curves. The research presented here is an extension of the state-of-the-art framework proposed by Mathew et al. (2021). However, the two main aspects of the current study are the application of deep learning for the prediction of Kr curves and the application of feature engineering. The latter not only reduces the training/testing time for the ML models but also enables the end user to obtain the final predictions with the least set of experimental data. The various models discussed in this research work currently focus on the prediction of Kr curves for drainage steady-state experiments; however, the work can be extended to capture the imbibition cycle as well.

Aim: Radiomics refers to automatic extraction of numerous quantitative features from medical images to supplement visual assessment. Machine-learning algorithms provide a suitable statistical methodology for devising predictive classifiers based on large radiomics datasets. We aimed to predict intracerebral hemorrhage (ICH) outcome by applying machine-learning classifiers to both clinical data and hematoma radiomics features. Methods: Patients enrolled in the Yale Longitudinal Study of ICH were included if they had (1) spontaneous supratentorial ICH, (2) baseline CT scan, (3) known admission Glasgow Coma Scale (GCS), and (4) 3-month modified Rankin Scale (mRS). A total of 1134 radiomics features related to the intensity, shape, texture, and waveform were extracted from manually segmented ICH lesions on baseline CT. Clinical variables were patients’ age, gender, GCS, presence of intraventricular hemorrhage, and thalamic ICH. We calculated the averaged receiver operating characteristics (ROC) area under curve (AUC) in outcome prediction among 100 repeats of 5-fold cross-validation (x500 iterations) for different combinations of feature selection and machine-learning algorithms. Results: A total of 119 ICH patients were included, of whom 60 had poor outcome (mRS ≥4). Among different combinations, lasso regression feature selection and partial least square (PLS) classification model yielded the highest accuracy in outcome prediction (Figure), with an averaged (95% confidence interval) ROC AUC of 0.86 (0.83 - 0.89) using clinical variables “only”, versus 0.92 (0.89 - 0.95) using combination of clinical variables and 54 radiomics features selected by lasso regression. Among radiomics features selected by lasso regression, ICH lesion flatness had the highest variable importance and was the only shape feature selected. Conclusion: Addition of ICH lesion radiomics to clinical variables using machine-learning models can improve outcome prediction.