Development Of Machine Learning Models Research Articles

Precision forecasting of fertilizer components’ concentrations in mixed variable-rate fertigation through machine learning

Accurate monitoring of fertilizer concentration and variable irrigation components is crucial for achieving precision irrigation through variable-rate fertigation. However, technological and cost constraints pose challenges in effectively managing mixed variable-rate fertigation. This study addresses these challenges by integrating machine learning (ML) with easily monitored physical parameters, such as electrical conductivity (EC), pH, and temperature, to predict fertilizer solution components and concentrations in mixed variable-rate fertigation. Cubic spline interpolation (CSI) was used to enhance the training dataset. Six ML algorithms—Multivariate Linear Regression (MLR), Support Vector Machines (SVM), K-Nearest Neighbors (KNN), Extremely Randomized Trees (ETs), Multilayer Perceptron (MLP), and Extreme Gradient Boosting (XGB)—were used to develop prediction models, with their performance evaluated using the coefficient of determination (R2), root mean square error (RMSE), and Akaike information criterion (AIC). The Extended Fourier Amplitude Sensitivity Test (EFAST) assessed the sensitivity of the ML models to physical parameters. All of the ML models, except for MLR, particularly SVM, demonstrated superior performance in predicting fertilizer solution components’ concentrations, with R2 values between 0.989 and 0.997 and RMSE values between 0.089 and 0.210. The CSI significantly enhanced model performance, resulting in larger R2 values and smaller RMSE values. The AIC results and sensitivity analysis confirmed the exceptional performance of SVM, emphasizing its suitability for predicting fertilizer components using easily measured physical parameters. The developed ML models offer valuable insights for decision-makers managing irrigation and fertilization in mixed variable-rate fertigation.

Advancing hydrogen storage predictions in metal-organic frameworks: A comparative study of LightGBM and random forest models with data enhancement

The escalating consumption of fossil fuels has given rise to a substantial upsurge in greenhouse gas concentrations and global temperatures, which, in turn, has triggered severe climate-related consequences. The critical imperative to reduce CO2 emissions and combat global warming has spurred extensive investigations into clean energy alternatives, with hydrogen emerging as a compelling zero-emission energy source. As a pivotal component of clean energy strategies, hydrogen requires designing compact, lightweight, and efficient storage systems. This study focuses on the development and evaluation of machine learning models for predicting the efficiency of Metal-Organic Frameworks (MOFs) in hydrogen storage, a key aspect of advancing clean energy technologies. MOFs, a class of nanoporous materials, show remarkable potential for hydrogen storage due to their high surface area and porosity. However, selecting the most suitable MOF for this application from a vast array of possible structures is a daunting task. In this context, machine learning algorithms offer an efficient alternative for predicting MOF suitability by considering their structural and chemical properties. We used ensemble learning methods, specifically Light Gradient Boosting Machine (LightGBM) and Random Forest (RF), to predict hydrogen uptake of MOFs based on a dataset of 219 experimentally tested samples. Two modeling scenarios were considered: one using the entire dataset, and the other involving strategic data pre-processing, including outlier removal and feature engineering. The results demonstrate that the measures taken to refine the dataset significantly enhance the predictive performance of the developed models, reducing prediction errors and improving overall goodness of fit. Specifically, the Mean Absolute Error (MAE) values for both the LightGBM and random forest models were reduced from 0.48 and 0.94, respectively, to 0.16 for both models, and the coefficients of determination (R2) increased substantially from 0.84 and 0.72 to 0.95, in both cases. Moreover, feature importance analysis unveiled that pressure-related features make the most significant contributions to the formation of tree ensembles during the model training process. A parametric sensitivity analysis was conducted revealing that H2 uptake in MOFs is most sensitive to changes in adsorption enthalpy, followed by surface area and temperature, while showing lower sensitivity to variations in pressure, consistent with established literature. These results underscore the pivotal role of data enhancement methods in refining machine learning models and can be instrumental in accelerating the development and optimization of MOF materials for clean energy applications.

Use of explainable machine learning models in blast load prediction

The effects of blast waves and their consequent damage to structures have been an increasingly popular research topic in the past decade. Various methods are used in blast load prediction on structures, including experimental, semi-empirical and numerical approaches. However, there is a demand for developing time-efficient predictive methods that help various professionals, such as engineering practitioners and first responders, in their regular activities. Machine learning (ML), as a subset of artificial intelligence (AI), has gained significant attention in various engineering applications over the past decades for its ability to aid decision-making. However, a critical gap exists in the comprehensive analysis of these ML models' transparency and interpretability levels. This paper presents a novel approach to employing explainable machine learning (XML) to predict the blast loads generated by high explosives. The primary objective is to assess the validity of utilising a machine learning (ML) framework for blast load predictions, examining the scientific consistency of the model's prediction process. This research utilises XML to interpret the ML model-based blast load prediction framework, marking the first documented use of XML in published literature for blast load prediction. This research study utilised three ML models, namely a) Decision Tree, b) Random Forest, and c) Extreme Gradient Boost (XGB) models, to predict blast loads on rigid structures. Validated numerical model results were used to generate the dataset to train the ML models, and the respective dataset consists of approximately 600 independent blast events. The models were evaluated for their predictability using error metrics: adjusted coefficient of determination (adjusted R2), root mean square error (RMSE), and mean absolute error (MAE). The overall ML model performance evaluation showed that the best-performing model, XGB, could make predictions with 98 % accuracy compared to validated numerical predictions. The XML explanations for the XGB models showcased that the blast load prediction procedure of the models aligns with the blast physics principles, ensuring the credibility of the developed ML models. In summary, these XML predictive models can be considered accurate and reliable for blast load prediction on structures.

Noninvasive assessment of organ-specific and shared pathways in multi-organ fibrosis using T1 mapping.

Fibrotic diseases affect multiple organs and are associated with morbidity and mortality. To examine organ-specific and shared biologic mechanisms that underlie fibrosis in different organs, we developed machine learning models to quantify T1 time, a marker of interstitial fibrosis, in the liver, pancreas, heart and kidney among 43,881 UK Biobank participants who underwent magnetic resonance imaging. In phenome-wide association analyses, we demonstrate the association of increased organ-specific T1 time, reflecting increased interstitial fibrosis, with prevalent diseases across multiple organ systems. In genome-wide association analyses, we identified 27, 18, 11 and 10 independent genetic loci associated with liver, pancreas, myocardial and renal cortex T1 time, respectively. There was a modest genetic correlation between the examined organs. Several loci overlapped across the examined organs implicating genes involved in a myriad of biologic pathways including metal ion transport (SLC39A8, HFE and TMPRSS6), glucose metabolism (PCK2), blood group antigens (ABO and FUT2), immune function (BANK1 and PPP3CA), inflammation (NFKB1) and mitosis (CENPE). Finally, we found that an increasing number of organs with T1 time falling in the top quintile was associated with increased mortality in the population. Individuals with a high burden of fibrosis in ≥3 organs had a 3-fold increase in mortality compared to those with a low burden of fibrosis across all examined organs in multivariable-adjusted analysis (hazard ratio = 3.31, 95% confidence interval 1.77-6.19; P = 1.78 × 10-4). By leveraging machine learning to quantify T1 time across multiple organs at scale, we uncovered new organ-specific and shared biologic pathways underlying fibrosis that may provide therapeutic targets.

Machine learning model for cardiovascular disease prediction in patients with chronic kidney disease

IntroductionCardiovascular disease (CVD) is the leading cause of death in patients with chronic kidney disease (CKD). This study aimed to develop CVD risk prediction models using machine learning to support clinical decision making and improve patient prognosis.MethodsElectronic medical records from patients with CKD at a single center from 2015 to 2020 were used to develop machine learning models for the prediction of CVD. Least absolute shrinkage and selection operator (LASSO) regression was used to select important features predicting the risk of developing CVD. Seven machine learning classification algorithms were used to build models, which were evaluated by receiver operating characteristic curves, accuracy, sensitivity, specificity, and F1-score, and Shapley Additive explanations was used to interpret the model results. CVD was defined as composite cardiovascular events including coronary heart disease (coronary artery disease, myocardial infarction, angina pectoris, and coronary artery revascularization), cerebrovascular disease (hemorrhagic stroke and ischemic stroke), deaths from all causes (cardiovascular deaths, non-cardiovascular deaths, unknown cause of death), congestive heart failure, and peripheral artery disease (aortic aneurysm, aortic or other peripheral arterial revascularization). A cardiovascular event was a composite outcome of multiple cardiovascular events, as determined by reviewing medical records.ResultsThis study included 8,894 patients with CKD, with a composite CVD event incidence of 25.9%; a total of 2,304 patients reached this outcome. LASSO regression identified eight important features for predicting the risk of CKD developing into CVD: age, history of hypertension, sex, antiplatelet drugs, high-density lipoprotein, sodium ions, 24-h urinary protein, and estimated glomerular filtration rate. The model developed using Extreme Gradient Boosting in the test set had an area under the curve of 0.89, outperforming the other models, indicating that it had the best CVD predictive performance.ConclusionThis study established a CVD risk prediction model for patients with CKD, based on routine clinical diagnostic and treatment data, with good predictive accuracy. This model is expected to provide a scientific basis for the management and treatment of patients with CKD.

Rapid quality assessment and traceability of ginger powder from Northeast India and Indian market based on near infrared spectroscopic fingerprinting.

Ginger (Zingiber officinale Rosc.) varies widely due to varying concentrations of phytochemicals and geographical origin. Rapid non-invasive quality and traceability assessment techniques ensure a sustainable value chain. The objective of this study is the development of suitable machine learning models to estimate the concentration of 6-gingerol and check traceability based on the spectral fingerprints of dried ginger samples collected from Northeast India and the Indian market using near-infrared spectrometry. Samples from the market and Northeast India underwent High Performance Liquid Chromatographic analysis for 6-gingerol content estimation. Near infrared (NIR) Spectrometer acquired spectral data. Quality prediction utilized partial least square regression (PLSR), while fingerprint-based traceability identification employed principal component analysis and t-distributed stochastic neighbor embedding (t-SNE). Model performance was assessed using RMSE and R2 values across selective wavelengths and spectral fingerprints. The standard normal variate pretreated spectral data over the wavelength region of 1,100-1,250 nm and 1,325-1,550 nm showed the optimal calibration model with root mean square error of calibration and R2 C (coefficient of determination for calibration) values of 0.87 and 0.897 respectively. A lower value (0.24) of root mean square error of prediction and a higher value (0.973) of R2 P (coefficient of determination for prediction) indicated the effectiveness of the developed model. t-SNE performed better clustering of samples based on geographical location, which was independent of gingerol content. The developed NIR spectroscopic model for Indian ginger samples predicts the 6-gingerol content and provides geographical traceability-based identification to ensure a sustainable value chain, which can promote efficiency, cost-effectiveness, consumer confidence, sustainable sourcing, traceability, and data-driven decision-making.

SuPreMo: a computational tool for streamlining in silico perturbation using sequence-based predictive models.

The increasing development of sequence-based machine learning models has raised the demand for manipulating sequences for this application. However, existing approaches to edit and evaluate genome sequences using models have limitations, such as incompatibility with structural variants, challenges in identifying responsible sequence perturbations, and the need for vcf file inputs and phased data. To address these bottlenecks, we present Sequence Mutator for Predictive Models (SuPreMo), a scalable and comprehensive tool for performing and supporting in silico mutagenesis experiments. We then demonstrate how pairs of reference and perturbed sequences can be used with machine learning models to prioritize pathogenic variants or discover new functional sequences. SuPreMo was written in Python, and can be run using only one line of code to generate both sequences and 3D genome disruption scores. The codebase, instructions for installation and use, and tutorials are on the Github page: https://github.com/ketringjoni/SuPreMo/tree/main. Supplementary data are available at Bioinformatics online.

Machine Learning to Predict Drug-Induced Liver Injury and Its Validation on Failed Drug Candidates in Development

Drug-induced liver injury (DILI) poses a significant challenge for the pharmaceutical industry and regulatory bodies. Despite extensive toxicological research aimed at mitigating DILI risk, the effectiveness of these techniques in predicting DILI in humans remains limited. Consequently, researchers have explored novel approaches and procedures to enhance the accuracy of DILI risk prediction for drug candidates under development. In this study, we leveraged a large human dataset to develop machine learning models for assessing DILI risk. The performance of these prediction models was rigorously evaluated using a 10-fold cross-validation approach and an external test set. Notably, the random forest (RF) and multilayer perceptron (MLP) models emerged as the most effective in predicting DILI. During cross-validation, RF achieved an average prediction accuracy of 0.631, while MLP achieved the highest Matthews Correlation Coefficient (MCC) of 0.245. To validate the models externally, we applied them to a set of drug candidates that had failed in clinical development due to hepatotoxicity. Both RF and MLP accurately predicted the toxic drug candidates in this external validation. Our findings suggest that in silico machine learning approaches hold promise for identifying DILI liabilities associated with drug candidates during development.

#2586 Time to event prediction of new-onset diabetes after transplantation in kidney transplant recipients using machine learning models

Abstract Background and Aims New-onset diabetes after transplantation (NODAT) is a common and serious complication of kidney transplantation that contributes to long-term morbidity and mortality after the transplant procedure. Conventional methods may predict the risk of, but not time to developing NODAT. We aimed to develop a machine learning model that predicts the time to NODAT in kidney transplant recipients (KTR). Method Two large prospective cohort studies of stable KTR (inclusion >1 year after transplantation) from the University Medical Center Groningen were combined. Data collection took place between 2001 and 2003 for cohort A, and between 2008 and 2011 for cohort B, resulting in the inclusion of 750 KTR free of NODAT at baseline. To predict absolute time to NODAT, a bootstrap aggregating (bagging) approach was utilized which consisted of several extreme gradient boosting (XGBoost) models with the accelerated failure time (AFT) objective. The performance of the resulting bagging model was internally evaluated by assessing the concordance index (c-index) in 20% of the dataset not used for training. Results From the 750 KTR included, 76 (10.1%) developed NODAT during a median (interquartile range [IQR]) follow-up of 5.8 (4.7 to 9.2) years. The median (IQR) age was 51.2 (40.5 to 60.7) years and 342 (45.6%) of the KTR were female. During feature engineering, 18 features from 53 were selected. After hyperparameter tuning, model performance during internal validation was 0.84. Furthermore, SHapley Additive exPlanations (SHAP) values were used to estimate feature importance, which revealed that glucose, HbA1c, mean arterial pressure, cyclosporine concentrations and daily vegetable intake were the most important features. Conclusion The developed machine learning model shows good discriminative ability of time to NODAT prediction. Due to the methodology used, the model allows for flexible incremental learning, meaning that the bagging model can be retrained with additional data without compromising on performance.

#1833 Analysis of clinical and laboratory manifestations in different classes of lupus nephritis—a Serbian single center experience

Abstract Background and Aims Lupus nephritis represents one of the most serious manifestations of systemic lupus erythematosus. The gold standard for determining the class of lupus nephritis is percutaneous kidney biopsy and the histopathological analysis of the biopsy specimen. The aim of this study was to analyze the clinical and laboratory characteristics of different histopathological classes of lupus nephritis (LN) at the time of the systemic lupus diagnosis and onset of symptoms. Method This retrospective cross-sectional study included a total of 54 subjects who were hospitalized at the Clinic for Nephrology and Clinical Immunology of the University Clinical Center of Vojvodina from 2017 to 2022 and underwent percutaneous kidney biopsy. Available clinical and laboratory data, as well as the histopathological findings of kidney biopsies, were analyzed. Based on the histopathological classification, patients were divided into four groups (Class II, III, IV, and V), and the clinical and laboratory characteristics were analyzed for all subjects. Due to a low number of patients, no analysis of Class I and Class VI was made. Results The most prevalent class among the subjects was Class IV, in which the lowest values of the C3 complement component were recorded. Subjects in this LN class had higher blood pressure values, a higher frequency of clinical manifestations such as arthralgia, and joint swelling. Class IV LN was also associated with elevated anti-dsDNA antibody values with a significant statistical significance (p = 0.004), lower red blood cell parameters, and higher levels of nitrogenous substances compared to other classes. The lowest values of total serum proteins and albumins, as well as the highest values of total cholesterol, LDL cholesterol, and triglycerides, were observed in the group of subjects with Class V LN; C-reactive protein had the highest average values in Class III and IV, which showed a statistical significance (p = 0.007). The lowest values of total serum calcium and iron were in Class III, and 24-hour proteinuria was most pronounced in Class V (p = 0.001), whereas microscopic hematuria was most present in Class II LN (p = 0.03). Conclusion Patients with lupus nephritis exhibit different clinical and laboratory manifestations based on histopathological classification and the time of diagnosis. The results of this study suggest their potential use in diagnosis, assessment of histopathological classification, and the development of a targeted and personalized therapeutic approach. These parameters will be used to develop machine learning models which might help in further evaluation and treatment.

Federated Learning on Distributed and Encrypted Data for Smart Manufacturing

Abstract Industry 4.0 drives exponential growth in the amount of operational data collected in factories. These data are commonly distributed and stored in different business units or cooperative companies. Such data-rich environments increase the likelihood of cyber attacks, privacy breaches, and security violations. Also, this poses significant challenges on developing machine learning models on sensitive data that are distributed among different business units. To fill this gap, this paper presents a novel privacy-preserving framework to enable federated learning on siloed and encrypted data for smart manufacturing. Specifically, we leverage fully homomorphic encryption (FHE) to allow for computation on ciphertexts and generate encrypted results which, when decrypted, match the results of mathematical operations performed on the plaintexts. Multi-layer encryption and privacy protection reduce the likelihood of data breaches while maintaining the prediction performance of machine learning models. Experimental results in real-world case studies show that the proposed framework yields superior performance to reduce the risk of cyber attacks and harness siloed data for smart manufacturing.

CLASSIFICATION OF KAZAKH MUSIC GENRES USING MACHINE LEARNING TECHNIQUES

This article analysis a Kazakh Music dataset, which consists of 800 audio tracks equally distributed across 5 different genres. The purpose of this research is to classify music genres by using machine learning algorithms Decision Tree Classifier and Logistic regression. Before the classification, the given data was pre-processed, missing or irrelevant data was removed. The given dataset was analyzed using a correlation matrix and data visualization to identify patterns. To reduce the dimension of the original dataset, the PCA method was used while maintaining variance. Several key studies aimed at analyzing and developing machine learning models applied to the classification of musical genres are reviewed. Cumulative explained variance was also plotted, which showed the maximum proportion (90%) of discrete values ​​generated from multiple individual samples taken along the Gaussian curve. A comparison of the decision tree model to a logistic regression showed that for f1 Score Logistic regression produced the best result for classical music - 82%, Decision tree classification - 75%. For other genres, the harmonic mean between precision and recall for the logistic regression model is equal to zero, which means that this model completely fails to classify the genres Zazz, Kazakh Rock, Kazakh hip hop, Kazakh pop music. Using the Decision tree classifier algorithm, the Zazz and Kazakh pop music genres were not recognized, but Kazakh Rock with an accuracy and completeness of 33%. Overall, the proposed model achieves an accuracy of 60% for the Decision Tree Classifier and 70% for the Logistic regression model on the training and validation sets. For uniform classification, the data were balanced and assessed using the cross-validation method. The approach used in this study may be useful in classifying different music genres based on audio data without relying on human listening.

Identification of patients' smoking status using an explainable AI approach: a Danish electronic health records case study.

Smoking is a critical risk factor responsible for over eight million annual deaths worldwide. It is essential to obtain information on smoking habits to advance research and implement preventive measures such as screening of high-risk individuals. In most countries, including Denmark, smoking habits are not systematically recorded and at best documented within unstructured free-text segments of electronic health records (EHRs). This would require researchers and clinicians to manually navigate through extensive amounts of unstructured data, which is one of the main reasons that smoking habits are rarely integrated into larger studies. Our aim is to develop machine learning models to classify patients' smoking status from their EHRs. This study proposes an efficient natural language processing (NLP) pipeline capable of classifying patients' smoking status and providing explanations for the decisions. The proposed NLP pipeline comprises four distinct components, which are; (1) considering preprocessing techniques to address abbreviations, punctuation, and other textual irregularities, (2) four cutting-edge feature extraction techniques, i.e. Embedding, BERT, Word2Vec, and Count Vectorizer, employed to extract the optimal features, (3) utilization of a Stacking-based Ensemble (SE) model and a Convolutional Long Short-Term Memory Neural Network (CNN-LSTM) for the identification of smoking status, and (4) application of a local interpretable model-agnostic explanation to explain the decisions rendered by the detection models. The EHRs of 23,132 patients with suspected lung cancer were collected from the Region of Southern Denmark during the period 1/1/2009-31/12/2018. A medical professional annotated the data into 'Smoker' and 'Non-Smoker' with further classifications as 'Active-Smoker', 'Former-Smoker', and 'Never-Smoker'. Subsequently, the annotated dataset was used for the development of binary and multiclass classification models. An extensive comparison was conducted of the detection performance across various model architectures. The results of experimental validation confirm the consistency among the models. However, for binary classification, BERT method with CNN-LSTM architecture outperformed other models by achieving precision, recall, and F1-scores between 97% and 99% for both Never-Smokers and Active-Smokers. In multiclass classification, the Embedding technique with CNN-LSTM architecture yielded the most favorable results in class-specific evaluations, with equal performance measures of 97% for Never-Smoker and measures in the range of 86 to 89% for Active-Smoker and 91-92% for Never-Smoker. Our proposed NLP pipeline achieved a high level of classification performance. In addition, we presented the explanation of the decision made by the best performing detection model. Future work will expand the model's capabilities to analyze longer notes and a broader range of categories to maximize its utility in further research and screening applications.

AntiMicro.ai: An Artificial Intelligence powered web application for predicting antibacterial/antifungal susceptibility and constructing personalized machine learning models

Background To train and develop machine learning models on the Pfizer antibacterial and antifungal datasets with subsequent deployment to an interactive Web Application. Methods We utilized R version 4.3.1 to perform descriptive analysis to obtain features/predictors. Python 3.10 libraries NumPy, Pandas, Scikit learn, Pycaret were used to train machine learning models. All these models were scored using area under the curve, recall, precision, F1, Kappa and the Mathews correlation coefficient. The best performing model was then deployed into a web application built on Streamlit after which deployment was done using GitHub and Streamlit cloud. A prototype android app was also developed using GoNative. Results The exploratory data analysis, S Aureus (17.2%) was the most common species however in the sub group analysis of the isolates with genotypic values, K Pneumoniae(48.2%) and E Coli (36.9%) were dominant. Amongst the fungi, Candida albicans (38.3%) and Candida glabrata (15.5%) were dominant. Feature selection was done using Shapley additive explanation plots. Using Extreme Gradient Boosting (XGBoost), we were able to achieve 99% and 97.8% AUC in the prediction of antibacterial and antifungal susceptibility respectively with minimal overfitting. Conclusions Decision tree based methods are a viable options of predicting antibacterial and antifungal drug resistance. When presented in visually appealing modes like web applications and android apps, it can be a useful guide to clinicians on initial treatment while awaiting definitive phenotypic testing. It can also be a surveillance tool that can craft antimicrobial resistance strategies.

Development of interpretable machine learning models to predict in-hospital prognosis of acute heart failure patients.

In recent years, there has been remarkable development in machine learning (ML) models, showing a trend towards high prediction performance. ML models with high prediction performance often become structurally complex and are frequently perceived as black boxes, hindering intuitive interpretation of the prediction results. We aimed to develop ML models with high prediction performance, interpretability, and superior risk stratification to predict in-hospital mortality and worsening heart failure (WHF) in patients with acute heart failure (AHF). Based on the Kyoto Congestive Heart Failure registry, which enrolled 4056 patients with AHF, we developed prediction models for in-hospital mortality and WHF using information obtained on the first day of admission (demographics, physical examination, blood test results, etc.). After excluding 16 patients who died on the first or second day of admission, the original dataset (n=4040) was split 4:1 into training (n=3232) and test datasets (n=808). Based on the training dataset, we developed three types of prediction models: (i) the classification and regression trees (CART) model; (ii) the random forest (RF) model; and (iii) the extreme gradient boosting (XGBoost) model. The performance of each model was evaluated using the test dataset, based on metrics including sensitivity, specificity, area under the receiver operating characteristic curve (AUC), Brier score, and calibration slope. For the complex structure of the XGBoost model, we performed SHapley Additive exPlanations (SHAP) analysis, classifying patients into interpretable clusters. In the original dataset, the proportion of females was 44.8% (1809/4040), and the average age was 77.9±12.0. The in-hospital mortality rate was 6.3% (255/4040) and the WHF rate was 22.3% (900/4040) in the total study population. In the in-hospital mortality prediction, the AUC for the XGBoost model was 0.816 [95% confidence interval (CI): 0.815-0.818], surpassing the AUC values for the CART model (0.683, 95% CI: 0.680-0.685) and the RF model (0.755, 95% CI: 0.753-0.757). Similarly, in the WHF prediction, the AUC for the XGBoost model was 0.766 (95% CI: 0.765-0.768), outperforming the AUC values for the CART model (0.688, 95% CI: 0.686-0.689) and the RF model (0.713, 95% CI: 0.711-0.714). In the XGBoost model, interpretable clusters were formed, and the rates of in-hospital mortality and WHF were similar among each cluster in both the training and test datasets. The XGBoost models with SHAP analysis provide high prediction performance, interpretability, and reproducible risk stratification for in-hospital mortality and WHF for patients with AHF.

Identifying potential (re)hemorrhage among sporadic cerebral cavernous malformations using machine learning

The (re)hemorrhage in patients with sporadic cerebral cavernous malformations (CCM) was the primary aim for CCM management. However, accurately identifying the potential (re)hemorrhage among sporadic CCM patients in advance remains a challenge. This study aims to develop machine learning models to detect potential (re)hemorrhage in sporadic CCM patients. This study was based on a dataset of 731 sporadic CCM patients in open data platform Dryad. Sporadic CCM patients were followed up 5 years from January 2003 to December 2018. Support vector machine (SVM), stacked generalization, and extreme gradient boosting (XGBoost) were used to construct models. The performance of models was evaluated by area under receiver operating characteristic curves (AUROC), area under the precision-recall curve (PR-AUC) and other metrics. A total of 517 patients with sporadic CCM were included (330 female [63.8%], mean [SD] age at diagnosis, 42.1 [15.5] years). 76 (re)hemorrhage (14.7%) occurred during follow-up. Among 3 machine learning models, XGBoost model yielded the highest mean (SD) AUROC (0.87 [0.06]) in cross-validation. The top 4 features of XGBoost model were ranked with SHAP (SHapley Additive exPlanations). All-Elements XGBoost model achieved an AUROCs of 0.84 and PR-AUC of 0.49 in testing set, with a sensitivity of 0.86 and a specificity of 0.76. Importantly, 4-Elements XGBoost model developed using top 4 features got a AUROCs of 0.83 and PR-AUC of 0.40, a sensitivity of 0.79, and a specificity of 0.72 in testing set. Two machine learning-based models achieved accurate performance in identifying potential (re)hemorrhages within 5 years in sporadic CCM patients. These models may provide insights for clinical decision-making.

The influence of manual segmentation strategies and different phases selection on machine learning-based computed tomography in renal tumors: a systematic review and meta-analysis.

Delineating the region/volume of interest (ROI/VOI) and selecting the phases are of importance in developing machine learning (ML). The results will change when choosing different methods of drawing the ROI/VOI and selecting different phases. However, there is no related standard for delineating the ROI/VOI and selecting the phases in renal tumors to develop ML based on computed tomography (CT). The PubMed and Web of Science were searched for related studies published until March 1, 2023. Inclusion criteria were studies that developed ML models in renal tumors from CT images. And the binary diagnostic accuracy data were extracted to obtain the outcomes, such as sensitivity (SE), specificity (SP), accuracy (ACC), and area under the curve (AUC). Twenty-three papers were included in the meta-analysis with a pooled SE of 87% (95% CI 85-88%), SP of 82% (95% CI 79-85%), and AUC of 91% (95% CI 89-93%) in phases; a pooled SE of 82% (95% CI 80-84%), SP of 85% (95% CI 83-86%), and AUC of 90% (95% CI 88-93%) in phases combined with delineating strategies, respectively. In all different combinations, the contour-focused and single phase produce the highest AUC of 93% (95% CI 90-95%). In subgroup analyses (sample size, year of publication, and geographical distribution), the performance was acceptable on phases and phases combined strategies. To explore the effect of manual segmentation strategies and different phases selection on ML-based CT, we find that the method of single phase (CMP or NP) combined with contour-focused was considered a better strategy compared to the other strategies.

Developing machine learning models to predict multi-class functional outcomes and death three months after stroke in Sweden.

Globally, stroke is the third-leading cause of mortality and disability combined, and one of the costliest diseases in society. More accurate predictions of stroke outcomes can guide healthcare organizations in allocating appropriate resources to improve care and reduce both the economic and social burden of the disease. We aim to develop and evaluate the performance and explainability of three supervised machine learning models and the traditional multinomial logistic regression (mLR) in predicting functional dependence and death three months after stroke, using routinely-collected data. This prognostic study included adult patients, registered in the Swedish Stroke Registry (Riksstroke) from 2015 to 2020. Riksstroke contains information on stroke care and outcomes among patients treated in hospitals in Sweden. Prognostic factors (features) included demographic characteristics, pre-stroke functional status, cardiovascular risk factors, medications, acute care, stroke type, and severity. The outcome was measured using the modified Rankin Scale at three months after stroke (a scale of 0-2 indicates independent, 3-5 dependent, and 6 dead). Outcome prediction models included support vector machines, artificial neural networks (ANN), eXtreme Gradient Boosting (XGBoost), and mLR. The models were trained and evaluated on 75% and 25% of the dataset, respectively. Model predictions were explained using SHAP values. The study included 102,135 patients (85.8% ischemic stroke, 53.3% male, mean age 75.8 years, and median NIHSS of 3). All models demonstrated similar overall accuracy (69%-70%). The ANN and XGBoost models performed significantly better than the mLR in classifying dependence with F1-scores of 0.603 (95% CI; 0.594-0.611) and 0.577 (95% CI; 0.568-0.586), versus 0.544 (95% CI; 0.545-0.563) for the mLR model. The factors that contributed most to the predictions were expectedly similar in the models, based on clinical knowledge. Our ANN and XGBoost models showed a modest improvement in prediction performance and explainability compared to mLR using routinely-collected data. Their improved ability to predict functional dependence may be of particular importance for the planning and organization of acute stroke care and rehabilitation.

Using spectral vegetation indices and machine learning models for predicting the yield of sugar beet (Beta vulgaris L.) under different irrigation treatments

Yield prediction is essential for production planning, resource management, and competitive advantages. The use of vegetation indices and machine learning for yield prediction is rapid, practical, and accurate and contributes to the development of digital agriculture. Although many studies have been carried out on yield estimation, the number of studies on sugar beet and irrigation applications is limited. To solve this problem, in this study yield prediction of sugar beet under different irrigation treatments was carried out using spectral reflection-based vegetation indices obtained by proximal measurements and developed machine learning models. The multilayer perceptron (MLP), random forest (RF), k-nearest (kNN), bagging (BAG), support vector regression (SVR), gaussian processes (GP), and CustomNet (CN) predictors were used in yield prediction. The irrigation treatments included I100 (application of irrigation to increase the available moisture to field capacity by around 45–50% of the water holding capacity at the root depth), I75 (75% of the water applied in the control group), I50 (50% of the water applied in the control group), and I125 (25% of the water applied in the control group). The combinations of vegetation indices were determined using Pearson correlation and principal component analysis. OSAVI, SAVI, and NDVI had the most significant influence on the yield prediction of sugar beet. Among the combinations used, the kNN1 model (including all attributes as input) showed high R2 values of 0.99 for training and 0.64 for testing. In addition, the kNN3 model (including Treatment, NDVI, NWI, and OSAVI as inputs) had R2 values of 0.97 for training and 0.65 for testing, indicating successful outcomes. However, the MLP1 and SVR1 models had lower results with test R2 values of 0.53 and 0.56, respectively. The results demonstrate the successful application of machine learning and vegetation indices for yield prediction. The proposed approaches could encourage agricultural experts and researchers in future work for yield mapping and identification.

Predicting blood transfusion following traumatic injury using machine learning models: A systematic review and narrative synthesis.

Haemorrhage is a leading cause of preventable death in trauma. Accurately predicting a patient's blood transfusion requirement is essential but can be difficult. Machine learning (ML) is a field of artificial intelligence that is emerging within medicine for accurate prediction modelling. This systematic review aimed to identify and evaluate all ML models that predict blood transfusion in trauma. This systematic review was registered on The International Prospective register of Systematic Reviews (CRD4202237110). MEDLINE, Embase and the Cochrane Central Register of Controlled Trials were systematically searched. Publications reporting a ML model that predicted blood transfusion in injured adult patients were included. Data extraction and risk of bias assessment was performed using validated frameworks. Data was synthesised narratively due to significant heterogeneity. Twenty-five ML models for blood transfusion prediction in trauma were identified. Models incorporated diverse predictors and varied ML methodologies. Predictive performance was variable but eight models achieved excellent discrimination (AUROC >0.9) and nine models achieved good discrimination (AUROC >0.8) in internal validation. Only two models reported measures of calibration. Four models have been externally validated in prospective cohorts: the Bleeding Risk Index, Compensatory Reserve Index, the Marsden model and the Mina model. All studies were considered at high risk of bias often due to retrospective datasets, small sample size and lack of external validation. This review identified twenty-five ML models developed to predict blood transfusion requirement after injury. Seventeen ML models demonstrated good to excellent performance in-silico but only four models were externally validated. To date ML models demonstrate the potential for early and individualised blood transfusion prediction but further research is critically required to narrow the gap between ML model development and clinical application. Systematic Review Without Meta-Analysis, Level IV.