Feature Importance Analysis Research Articles

Machine learning prediction of dye adsorption by hydrochar: Parameter optimization and experimental validation

In response to escalating global wastewater issues, particularly from dye contaminants, many studies have begun using hydrochar to adsorb dye from wastewater. However, the relationship between the preparation conditions of hydrochar, the properties of hydrochar, experimental conditions, types of dyes, and equilibrium adsorption capacity (Q) has not yet been fully explored. This study conducted a comprehensive assessment using twelve distinct ML models. The Gradient Boosting Regressor (GBR) model exhibited superior performance with R² (0.9629) and RMSE (0.1166) in the test dataset, marking it as the most effective among the evaluated models. Moreover, this study also proved the feasibility of the GBR model through stability testing and residual analysis. A feature importance analysis prioritized the variables as follows: experimental conditions (41.5 %), properties of hydrochar (26.0 %), preparation conditions (18.1 %), and type of dye (14.4 %). Meanwhile, experimental conditions (C0 > 30 mmol/g, pH > 8, and higher solvent temperatures) and hydrochar properties (the BET surface area > 2000 m²/g, an (O+N)/C molar ratio < 0.6, and an H/C molar ratio of approximately 0.06) show higher Q for dyes. Experimental validation of the GBR model confirmed its practical utility with a suitable predictive accuracy (R² = 0.8704). Moreover, the study developed a Python-based GUI that has integrated the best GBR models to facilitate researchers' ongoing application and improvement of this predictive model. This study not only underscores the efficacy of ML in enhancing the understanding of dye adsorption by hydrochar but also sets a precedent for future research on sustainable contaminants removal through bio-based adsorbents.

Utilizing machine learning to forecast mechanical characteristics of NaOH-Treated jute fiber reinforced composite materials

The prediction of mechanical properties in composite materials is crucial for optimizing their performance in various engineering applications. This study focuses on NaOH treated jute fiber reinforced glass epoxy composites, aiming to predict key mechanical characteristics like flexural strength, tensile strength, and the hardness. A comprehensive dataset of 974 samples was analyzed, encompassing a range of input parameters including fiber content, fiber orientation, matrix type, and processing conditions. Advanced statistical and machine learning techniques were employed to develop predictive models, with XGBoost showing itself to be the best model in terms of predicting each of the three mechanical properties. Feature importance analysis revealed that fiber content is the most significant factor influencing the predictions.

Combining visual intelligence and social-physical urban features facilitates fine-scale seasonality characterization of urban thermal environments

Traditional methods for assessing urban thermal environments (UTEs) often rely on GIS and remote sensing data, suffering from data limitations in coverage, accuracy, and availability. To address these challenges, we established a novel framework developing visual intelligence simulated indices (VISIs) for improved UTE characterization. This framework combines GIS spatial data and pre-trained large-scale and local-trained GeoAI models to generate visual intelligence, extracting GIS-like visual elements from remote sensing imagery. These GIS-like VISIs comprehensively represent land surface temperature (LST) and urban fabric for cost-effective UTE analysis. We applied our framework to Zhengzhou, China, a city with diverse landscapes. Comparative experiments among Random Forests (RF), Neural Networks (NN), and Convolutional Neural Networks (CNN) demonstrated that RF models exhibited superior performance in LST representation (0.7 test R2 and 0.9 training R2). Interestingly, CNNs performed better than RFs and NNs when less cloud cover (<6 %) was present, suggesting that data diversity is essential for CNN to learn generalizable features for LST representation. For urban fabric representation, feature importance analysis indicated that VISIs (37 %) derived from large-scale GeoAI models outperform spectral bands (20 %). Furthermore, both Pearson correlation and Shapley values showed that VISIs (e.g., building features) better characterize urban heat islands than traditional remote sensing ecological indices (RSEIs) such as NDBI, underscoring the cost-effectiveness of our visual intelligence approach. In summary, our visual intelligence method effectively consolidates UTE seasonality characterization into a predictable scenario even given the unpredictable availability of remote sensing and GIS data, revealing novel perspectives on sensing unseen details for UTEs.

Predicting the compressive strength of rubberized concrete containing silica fume using stacking ensemble learning model

As the construction industry advances toward sustainability, rubberized concrete emerges as a promising material due to its potential for recycling waste rubber. While silica fume (SF) is often used to address the reduced compressive strength resulting from rubber integration, the complex interactions between these materials present significant modeling challenges. This study employs a novel machine learning approach to effectively capture these interactions and accurately predict the compressive strength of SF-enhanced rubberized concrete. Utilizing a dataset comprising 237 experimental data points curated from 25 research studies, an advanced stacking ensemble model was developed. This model features a trainable meta-structure that integrates diverse, hyper-tuned base learners, including Decision Trees (DT), Artificial Neural Networks (ANN), eXtreme Gradient Boosting (XGBoost), Support Vector Machines (SVM), and K-Nearest Neighbors (KNN) regressors. Hyperparameter tuning, performed using 10-fold cross-validation, was applied to enhance overall model performance. The findings show that XGBoost outperformed other base models, achieving an overall Coefficient of Determination (R²) of 0.9194 and a Mean Squared Error (MSE) of 10.5625. The stacking approach, with KNN as the meta-learner, further refined individual model performances, resulting in an improved R² of 0.9397 and an MSE of 7.1671 on the testing data. Compared to traditional voting ensemble techniques, the stacking models offered a more nuanced enhancement of predictive outcomes, while the averaging ensembles were noted for their simplicity and competitive accuracy. Additionally, feature importance analysis using SHapley Additive exPlanations (SHAP) revealed that superplasticizer, rubber content, and SF were the most influential inputs in the developed model.

Machine learning with multiple modalities of brain magnetic resonance imaging data to identify the presence of bipolar disorder

BackgroundBipolar disorder (BD) is a chronic psychiatric mood disorder that is solely diagnosed based on clinical symptoms. These symptoms often overlap with other psychiatric disorders. Efforts to use machine learning (ML) to create predictive models for BD based on data from brain imaging are expanding but have often been limited using only a single modality and the exclusion of the cerebellum, which may be relevant in BD. MethodsIn this study, we sought to improve ML classification of BD by combining information from structural, functional, and diffusion-weighted imaging. Participants (108 BD I, 78 control) with BD type I and matched controls were recruited into an imaging study. This dataset was randomly divided into training and testing sets. For each of the three modalities, a separate ML model was selected, trained, and then used to generate a prediction of the class of each test subject. Majority voting was used to combine results from the three models to make a final prediction of whether a subject had BD. An independent replication sample was used to evaluate the ability of the ML classification to generalize to data collected at other sites. ResultsCombining the three machine learning models through majority voting resulted in an accuracy of 89.5 % for classification of the test subjects as being in the BD or control group. Bootstrapping resulted in a 95 % confidence interval of 78.9 %–97.4 % for test accuracy. Performance was reduced when only using 2 of the 3 modalities. Analysis of feature importance revealed that the cerebellum and nodes of the emotional control network were among the most important regions for classification. The machine learning model performed at chance on the independent replication sample. ConclusionBD I could be identified with high accuracy in our relatively small sample by combining structural, functional, and diffusion-weighted imaging data within a single site but not generalize well to an independent replication sample. Future studies using harmonized imaging protocols may facilitate generalization of ML models.

Rice Yield Estimation Using Machine Learning and Feature Selection in Hilly and Mountainous Chongqing, China

To investigate effective techniques for estimating rice production in hilly and mountainous areas, in this study, we collected yield data at the field level, agro-meteorological data, and Sentinel-2/MSI remote sensing data in Chongqing, China, between 2020 and 2023. The integral values of vegetation indicators from the rice greening up to heading–filling stages were determined using the Newton–trapezoidal integration method. Using correlation analysis and importance analysis of permutation features, the effects of agro-meteorological variables and vegetation index integrals on rice yield were assessed. The chosen characteristics were then combined with three machine learning techniques—random forest (RF), support vector machine (SVM), and partial least squares regression (PLSR)—to create six rice yield estimate models. The results showed that combined vegetation indices were more effective than indices used in separate development phases. Specifically, the correlation coefficients between the integral values of eight vegetation indices from rice greening up to heading–filling stages and rice yield were all above 0.65. By introducing agro-meteorological factors as new independent variables and combining them with vegetation indices as input parameters, the predictive capability of the model was evaluated. The results showed that the performance of PLSR remained stable, while the prediction accuracies of SVM and RF improved by 13% to 21.5%. After feature selection, the inversion performance of all three machine learning models improved, with the RF model coupled with variables selected during permutation feature importance analysis achieving the optimal inversion effect, which was characterized by a coefficient of determination of 0.85, a root mean square error of 529.1 kg/hm2, and a mean relative error of 5.63%. This study provides technical support for improving the accuracy of remote sensing-based crop yield estimation in hilly and mountainous regions, facilitating precise agricultural management and informing agrarian decision making.

Predicting effective thermal conductivity of HGM composite using ML

Data-driven materials research can be used as an effective tool to supplement conventional approaches in designing new materials. Hollow glass microsphere (HGM) composites are widely used for high-temperature thermal insulation applications, and their synthesis is traditionally done by varying parameters of the constituents on a trial-and-error basis. In this work, a prediction tool based on a supervised machine learning (ML) model is developed to predict the effective thermal conductivity (ETC) of an HGM composite by tailoring the composition and parameters of the constituents to reduce the time and cost involved. A comprehensive database containing the various input features is generated from previous experimental investigations conducted on various HGM composites. ML models, namely random forest regression (RF), K-nearest neighbor (KNN), support vector regression (SVR), and artificial neural network (ANN), are used to predict the ETC of HGM composites. Feature importance analysis showed that the matrix material’s thermal conductivity and the composite’s bulk density have the highest impact on the ETC of the HGM composite, followed by porosity and average microsphere size. ANN emerged as the best model to predict HGM composite ETC with the lowest root mean square error and highest R2 value for predictions. Moreover, a systematic approach for key feature selection using ANN shows that adding or omitting additional features beyond an optimal combination degrades the model’s predictive accuracy.

Statistical and machine learning models for predicting spalling in CRCP

Continuously reinforced concrete pavement (CRCP), crucial for the resilience of transportation infrastructure owing to its continuous steel reinforcement, confronts a critical challenge in the form of spalling—a distress phenomenon posing a threat to pavement durability and overall structural integrity. The detachment or breakage of concrete from the surface compromises CRCP’s functionality and raises safety concerns and escalating maintenance costs. To address this pressing issue, our study investigates the multifaceted factors influencing spalling, employing a comprehensive approach that integrates statistical and machine learning techniques for predictive modeling. Descriptive statistics meticulously profile the dataset, emphasizing age, thickness, precipitation, temperature, and traffic parameters. Regression analysis unveils key relationships, emphasizing the significance of age, annual temperature, annual precipitation, maximum humidity, and the initial International Roughness Index (IRI) as influential factors. The correlation matrix heatmap guides feature selection, elucidating intricate interdependencies. Simultaneously, feature importance analysis identifies age, Average Annual Daily Traffic (AADT), and total pavement thickness as crucial contributors to spalling. In machine learning, adopting models, including Gaussian Process Regression and ensemble tree models, is grounded in their diverse capabilities and suitability for the complex task at hand. Their varying predictive accuracies underscore the importance of judicious model selection. This research advances pavement engineering practices by offering nuanced insights into factors influencing spalling in CRCP, refining our understanding of spalling influences. Consequently, the study not only opens avenues for developing improved predictive methodologies but also enhances the durability of CRCP infrastructure, addressing broader implications for informed decision-making in transportation infrastructure management. The selection of Gaussian Process Regression and ensemble tree models stems from their adaptability to capture intricate relationships within the dataset, and their comparative performance provides valuable insights into the diverse predictive capabilities of these models in the context of CRCP spalling.

Robust prediction for characteristics of digestion products in an industrial-scale biogas project via typical non-time series and time-series machine learning algorithms

Anaerobic digestion (AD) is a well-established pathway for treating agricultural organic waste, and machine learning has emerged as a novel tool to predict its product performance. In prior research, the majority of studies concentrated on non-time series models for laboratory-scale fermentation data. Consequently, the generalization performance of these models was significantly constrained, particularly in the context of industrial-scale biogas projects. Thus, in this study, typical non-time series models (GBR and RF) and time-series models (LSTM, CNN-LSTM, and DA-LSTM) after hyperparameter optimization were chosen to accurately predict the characteristics of digestion products in a biogas project. The ideal GBR model for CH4 content was obtained, and the R2 values of the test set and training set were 0.93 (RMSE=1.11) and 0.97 (RMSE=0.69), respectively. Temperature was the most important parameter for biogas production according to feature importance and SHAP analysis of the RF model. The DA-LSTM was superior to LSTM and CNN-LSTM for the prediction of biogas production, and the R2 of DA-LSTM was 0.87 (RMSE=1048.14) with a seq_len of 10 d. This study provides direction for high-efficiency biogas production in wet biogas projects with the aid of reliable machine learning models.

Explainable Ensemble Learning and Multilayer Perceptron Modeling for Compressive Strength Prediction of Ultra-High-Performance Concrete.

The performance of ultra-high-performance concrete (UHPC) allows for the design and creation of thinner elements with superior overall durability. The compressive strength of UHPC is a value that can be reached after a certain period of time through a series of tests and cures. However, this value can be estimated by machine-learning methods. In this study, multilayer perceptron (MLP) and Stacking Regressor, an ensemble machine-learning models, is used to predict the compressive strength of high-performance concrete. Then, the ML model's performance is explained with a feature importance analysis and Shapley additive explanations (SHAPs), and the developed models are interpreted. The effect of using different random splits for the training and test sets has been investigated. It was observed that the stacking regressor, which combined the outputs of Extreme Gradient Boosting (XGBoost), Category Boosting (CatBoost), Light Gradient Boosting Machine (LightGBM), and Extra Trees regressors using random forest as the final estimator, performed significantly better than the MLP regressor. It was shown that the compressive strength was predicted by the stacking regressor with an average R2 score of 0.971 on the test set. On the other hand, the average R2 score of the MLP model was 0.909. The results of the SHAP analysis showed that the age of concrete and the amounts of silica fume, fiber, superplasticizer, cement, aggregate, and water have the greatest impact on the model predictions.

Machine learning-based prediction of early neurological deterioration after intravenous thrombolysis for stroke: insights from a large multicenter study.

This investigation seeks to ascertain the efficacy of various machine learning models in forecasting early neurological deterioration (END) following thrombolysis in patients with acute ischemic stroke (AIS). Employing data from the Shenyang Stroke Emergency Map database, this multicenter study compiled information on 7,570 AIS patients from 29 comprehensive hospitals who received thrombolytic therapy between January 2019 and December 2021. An independent testing cohort was constituted from 2,046 patients at the First People's Hospital of Shenyang. The dataset incorporated 15 pertinent clinical and therapeutic variables. The principal outcome assessed was the occurrence of END post-thrombolysis. Model development was executed using an 80/20 split for training and internal validation, employing classifiers like logistic regression with lasso regularization (lasso regression), support vector machine (SVM), random forest (RF), gradient-boosted decision tree (GBDT), and multi-layer perceptron (MLP). The model with the highest area under the curve (AUC) was utilized to delineate feature significance. Baseline characteristics showed variability in END incidence between the training (n = 7,570; END incidence 22%) and external validation cohorts (n = 2,046; END incidence 10%; p < 0.001). Notably, all machine learning models demonstrated superior AUC values compared to the reference model, indicating their enhanced predictive capacity. The lasso regression model achieved the highest AUC at 0.829 (95% CI: 0.799-0.86; p < 0.001), closely followed by the MLP model with an AUC of 0.828 (95% CI: 0.799-0.858; p < 0.001). The SVM, RF, and GBDT models also showed commendable AUCs of 0.753, 0.797, and 0.774, respectively. Decision curve analysis revealed that the SVM and MLP models demonstrated a high net benefit. Feature importance analysis emphasized "Onset To Needle Time" and "Admission NIHSS Score" as significant predictors. Our research establishes the MLP and lasso regression as robust tools for predicting early neurological deterioration in acute ischemic stroke patients following thrombolysis. Their superior predictive accuracy, compared to traditional models, highlights the significant potential of machine learning approaches in refining prognosis and enhancing clinical decisions in stroke care management. This advancement paves the way for more tailored therapeutic strategies, ultimately aiming to improve patient outcomes in clinical practice.

Water Quality Assessment using Machine Learning: A Focus on Coliform Prediction in Water

Water quality assessment is essential for safeguarding public health and protecting water resources. This study focused on predicting water quality, specifically the presence of total coliforms, using various machine-learning techniques. The present study utilises a publicly available dataset encompassing the geographical area of India consisting of various physical water quality parameters. Various regression techniques were applied to the dataset after appropriate pre-processing including feature selection and normalisation. The findings demonstrate that gradient boosting regression outperforms other methods, achieving high accuracy with mean absolute error (MAE) of 0.0349, mean squared error (MSE) of 0.0038, and root mean squared error (RMSE) of 0.0620. Conductivity and temperature emerged as the most influential factors in total coliform prediction, as revealed by feature importance analysis. These results contribute to water quality understanding, aiding water resource management for public health protection. By accurately predicting total coliform presence, proactive measures can be taken timely to mitigate and minimise health risks associated with microbial contamination.

A Machine-Learning-Based Method for Identifying the Failure Risk State of Fissured Sandstone under Water-Rock Interaction.

The mechanical properties of fissured sandstone will deteriorate under water-rock interaction. It is crucial to extract the precursor information of fissured sandstone instability under water-rock interaction. The potential of each acoustic emission (AE) parameter as a precursor for instability in the failure process of fissured sandstone was investigated in this study. An experimental dataset comprising 586 acoustic emission experiments was established, and subsequent classification training and testing were conducted using three machine learning (ML) models: AdaBoost, MLP, and Random Forest (RF). The primary parameters for identifying the instability risk state of fissured sandstone include acoustic emission ringing count, energy (mV·ms), centroid frequency, peak frequency, Rise Angle (RA), Average Frequency (AF), b value, and the natural/saturated state of fissured sandstone: state. To enhance data utilization, a 10-fold cross-validation method was employed during the model training process. The machine learning models were developed and designed to identify the instability risk of fissured sandstone under the natural and saturated states. The results demonstrated that the established RF model was capable of identifying fissured sandstone instability risks with an accuracy of 97.87%. Feature importance analysis revealed that state and b value exerted the most significant influence on identification results. The Spearman correlation coefficient was utilized to assess the correlation between input features. This study can provide technical support to identify the risk of instability of fissured sandstones under both natural and saturated water conditions. Based on the models developed in this study, it is possible to implement an early warning method for instability in fissured sandstone that meets realistic working conditions. Compared with the traditional empirical and formulaic methods, the machine learning method can more quickly process huge amounts of AE data and accurately identify the damage state of fissured sandstone.

Analysis of bond strength of CFRP cables with concrete using random forest model

The use of carbon fiber-reinforced polymer (CFRP) as an alternative to steel tendons in prestressed concrete is increasingly favored due to its non-corrosive properties. A critical aspect of this substitution is the bond strength required to effectively transfer pre-stressing force. This study investigates the bond strengths of CFRP rods, CFRP strands, and steel strands following ISO 10406 standards. The research comprehensively explores the influence of cable type, cable surface roughness, and concrete strength on bond-slip behavior. Experimental results were augmented using a data augmentation method to enrich the dataset, which facilitated the development of a robust random forest (RF)-based prediction model for bond strength. The RF model achieved strong performance metrics, including a mean absolute percentage error (MAE) of 17.9 % and an R-value of 0.97, underscoring its efficacy in predicting bond strength. Notably, the model highlighted the significant impact of cable surface roughness on bond strength relative to other parameters studied based on the feature importance analysis. This study contributes to advancing the understanding of CFRP's applicability as a steel tendon replacement in prestressed concrete structures. The findings emphasize the importance of surface characteristics in optimizing bond strength, providing valuable insights for structural engineers and material scientists.

An Empirical Investigation into an Economic Analysis of State Road Transport Undertakings in India

State Road Transport Undertakings have been the backbone of the transportation network in India since their inception. These corporations were set up to provide a sustainable and cheap mode of transport for citizens of India, along with providing connectivity to rural parts of the nation. Over the years, the financial and physical performance of most State Road Transport Undertakings has dropped significantly. The social goal, political influence and poor management have been some of the major factors contributing to the losses made by most of the State Road Transport Undertakings. We have conducted this study by considering eight State Road Transport Undertakings, which are loss and profit-making, to understand the factors responsible for improving the physical factors of these State Road Transport Undertakings. To undertake this objective, we used correlation analysis and feature importance analysis using random forest regressor. We have proposed some policy-based measures for each State Road Transport Undertaking analysed based on the results and analysis. JEL Codes: C12, I38, L32, L91, N35, O53, R41, R42

Building Information Modeling and AI Algorithms for Optimizing Energy Performance in Hot Climates: A Comparative Study of Riyadh and Dubai

In response to increasing global temperatures and energy demands, optimizing buildings’ energy efficiency, particularly in hot climates, is an urgent challenge. While current research often relies on conventional energy estimation methods, there has been a decrease in the efforts dedicated to leveraging AI-based methodologies as technology advances. This implies a dearth of multiparameter examinations in AI-driven extreme case studies. For this reason, this study aimed to enhance the energy performance of residential buildings in the hot climates of Dubai and Riyadh by integrating Building Information Modeling (BIM) and Machine Learning (ML). Detailed BIM models of a typical residential villa in these regions were created using Revit, incorporating conventional, modern, and green building envelopes (BEs). These models served as the basis for energy simulations conducted with Green Building Studio (GBS) and Insight, focusing on crucial building features such as floor area, external and internal walls, windows, flooring, roofing, building orientation, infiltration, daylighting, and more. To predict Energy Use Intensity (EUI), four ML algorithms, namely, Gradient Boosting Machine (GBM), Random Forest (RF), Support Vector Machine (SVM), and Lasso Regression (LR), were employed. GBM consistently outperformed the others, demonstrating superior prediction accuracy with an R2 of 0.989. This indicates that the model explains 99% of the variance in EUI, highlighting its effectiveness in capturing the relationships between building features and energy consumption. Feature importance analysis (FIA) revealed that roofs (29% in Dubai scenarios (DS) and 40% in Riyadh scenarios (RS)), external walls (19% in DS and 29% in RS), and windows (15% in DS and 9% in RS) have the most impact on energy consumption. Additionally, the study explored the potential for energy optimization, such as cavity green walls and green roofs in RS and double brick walls with VIP insulation and green roofs in DS. The findings of the paper should be interpreted in light of certain limitations but they underscore the effectiveness of combining BIM and ML for sustainable building design, offering actionable insights for enhancing energy efficiency in hot climates.

New insights into the identification of biochemical traits linked to rooting percentage in fig (Ficus carica L.) cuttings

BACKGROUND: The fig (Ficus carica L.) tree known for its tasty and nutritious fruits, is typically propagated by cutting. While previous studies have focused on the effects of different treatments and environmental conditions on fig cutting propagation, little attention has been paid to the specific role and association of biochemical properties in leaves, stem bark and fruit on the rooting process. OBJECTIVE: This research explores the complex relationship between 40 biochemical traits and the rooting ability of fig cuttings. To achieve this objective, various machine learning techniques were employed, such as a random forest model, feature importance analysis, linear regression, and principal component analysis (PCA). RESULTS: The random forest model showed significant predictive ability with a classification accuracy of 100%, supported by a high kappa statistic. Feature importance analysis identified a* (a colorimetric parameter in fruit), fruit trans-ferulic acid and leaf total flavonoids as the most influential traits in determining the rooting ability of cuttings. The robustness of these findings is supported by the high R-squared value (0.9002) and low error metrics (MAE 0.7554 and MSE 0.6980) of the linear regression model built on these important traits. In parallel, PCA indicated that a*, leaf total flavonoids and fruit trans-ferulic acid were the dominant traits in samples with lower rooting percentage. CONCLUSIONS: These identified biomarkers can be effectively used by fig breeders and growers to select and introduce fig cultivars with improved rooting ability.

Exploiting Metal-Organic Frameworks for Vinylidene Fluoride Adsorption: From Force Field Development, Computational Screening to Machine Learning.

Metal-organic frameworks (MOFs) represent a distinctive class of nanoporous materials with considerable potential across a wide range of applications. Recently, a handful of MOFs has been explored for the storage of environmentally hazardous fluorinated gases (Keasler et al. Science 2023, 381, 1455), yet the potential of over 100,000 MOFs for this specific application has not been thoroughly investigated, particularly due to the absence of an established force field. In this study, we develop an accurate force field for nonaversive hydrofluorocarbon vinylidene fluoride (VDF) and conduct high-throughput computational screening to identify top-performing MOFs with high VDF adsorption capacities. Quantitative structure-property relationships are analyzed via machine learning models on the combinations of geometric, chemical, and topological features, followed by feature importance analysis to probe the effects of these features on VDF adsorption. Finally, from detailed structural analysis via radial distribution functions and spatial densities, we elucidate the significance of different interaction modes between VDF and metal nodes in top-performing MOFs. By synergizing force-field development, computational screening, and machine learning, our findings provide microscopic insights into VDF adsorption in MOFs that will advance the development of new nanoporous materials for high-performance VDF storage or capture.

Using machine learning to identify risk factors for short-term complications following thumb carpometacarpal arthroplasty

BackgroundThumb carpometacarpal (CMC) joint osteoarthritis is among the most common degenerative hand diseases. Thumb CMC arthroplasty, or trapeziectomy with or without tendon augmentation, is the most frequently performed surgical treatment and has a strong safety profile. Though adverse outcomes are infrequent, the ability to predict risk for complications has substantial clinical benefits. In the present study, we evaluated a well-known surgical database with machine learning (ML) techniques to predict short-term complications and reoperations after thumb CMC arthroplasty. MethodsA retrospective study was conducted using data from the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) years 2005–2020. Outcomes were 30-day wound and medical complications and 30-day return to the operating room. We used three ML algorithms — a Random Forest (RF), Elastic-Net Regression (ENet), and Extreme Gradient Boosted Tree (XGBoost), and a deep learning Neural Network (NN). Feature importance analysis was performed in the highest performing model for each outcome to identify predictors with the greatest contributions. ResultsWe included a total of 7711 cases. The RF was the best performing algorithm for all outcomes, with an AUC score of 0.61±0.03 for reoperations, 0.55±0.04 for medical complications, and 0.59±0.03 for wound complications. On feature importance analysis, procedure duration was the highest weighted predictor for reoperations. In all outcomes, procedure duration, older age, and female sex were consistently among the top five predictors. ConclusionsWe successfully developed ML algorithms to predict reoperations, wound complications, and medical complications. RF models had the highest performance in all outcomes.

Extent of resection and underlying liver disease influence the accuracy of the preoperative risk assessment with the American College of Surgeons Risk Calculator

BackgroundLiver surgery is associated with a significant risk of postoperative complications, depending on the extent of liver resection and the underlying liver disease. Therefore, adequate patient selection is crucial. This study aimed to assess the accuracy of the American College of Surgeons Risk Calculator (ACS-RC) by considering liver parenchyma quality and the type of liver resection. MethodsPatients who underwent open or minimally invasive liver resection for benign or malignant indications between January 2019 and March 2023 at the University Hospital Basel were included. Brier score and feature importance analysis were performed to investigate the accuracy of the ACS-RC. ResultsA total of 376 patients were included in the study, 214 (57%) who underwent partial hepatectomy, 89 (24%) who underwent hemihepatectomy, and 73 (19%) who underwent trisegmentectomy. Most patients had underlying liver diseases, with 143 (38%) patients having fibrosis, 75 patients (20%) having steatosis, and 61 patients (16%) having cirrhosis. The ACS-RC adequately predicted surgical site infection (Brier score of 0.035), urinary tract infection (Brier score of 0.038), and death (Brier score of 0.046), and moderate accuracy was achieved for serious complications (Brier score of 0.216) and overall complications (Brier score of 0.180). Compared with the overall cohort, the prediction was limited in patients with cirrhosis, fibrosis, and steatosis and in those who underwent hemihepatectomy and trisegmentectomy. The inclusion of liver parenchyma quality improved the prediction accuracy. ConclusionThe ACS-RC is a reliable tool for estimating 30-day postoperative morbidity, particularly for patients with healthy liver parenchyma undergoing partial liver resection. However, accurate perioperative risk prediction should be adjusted for underlying liver disease and extended liver resections.