Developed Machine Learning Research Articles

Revealing SARS-CoV-2 Mpro mutation cold and hot spots: Dynamic residue network analysis meets machine learning

Deciphering the effect of evolutionary mutations of viruses and predicting future mutations is crucial for designing long-lasting and effective drugs. While understanding the impact of current mutations on protein drug targets is feasible, predicting future mutations due to natural evolution of viruses and environmental pressures remains challenging. Here, we leveraged existing mutation data during the evolution of the SARS-CoV-2 protein drug target main protease (Mpro) to test the predictive power of dynamic residue network (DRN) analysis in identifying mutation cold and hot spots. We conducted molecular dynamics simulations on the Mpro of SARS-CoV-2 (Wuhan strain) and calculated eight DRN metrics (averaged BC, CC, DC, EC, ECC, KC, L, PR), each of which identifies a unique network feature within the protein. The sets of residues with the highest and lowest values for each metric, comprising potential cold and hot spots, were compared to published biochemical analyses and per residue mutation frequencies observed across five SARS-CoV-2 lineages, encompassing a total of 191,878 sequences. Individual DRN metrics displayed only modest power to predict the mutation frequency of individual residues. However, integrating the eight DRN metrics with additional structural and sequence-derived metrics allowed us to develop machine learning models which significantly improved the prediction of residue mutation frequency. While further refinements should enhance accuracy, we demonstrated a robust method to understand pathogen evolution. This approach can also guide the development of long-lasting drugs by targeting functional residues located in and near active site, and allosteric sites, that are less prone to mutations.

Development and validation of machine learning models for young-onset colorectal cancer risk stratification

Incidence of young-onset colorectal cancer (YOCRC, younger than 50) has significantly increased worldwide. The performance of fecal immunochemical test in detecting YOCRC is unsatisfactory. Using routine clinical data, we aimed to develop machine learning (ML) models to identify individuals with high-risk YOCRC who require further colonoscopy. We retrospectively extracted data of 10,874 young individuals. Multiple supervised ML techniques were devised to distinguish individuals with and without CRC, classifiers were trained, internally validated and temporally validated. In internal validation cohort, Random Forest (RF) ML model demonstrated good performance with AUC of 0.859 and highest recall of 0.840. In temporal validation cohort, the RF ML model also exhibited good classification performance, achieving AUC of 0.888 and highest recall of 0.872. RF algorithm-based approach is effective and feasible in YOCRC risk stratification. This could be valuable in assessing the risk of YOCRC so that clinical management, including further colonoscopy, can be subsequently made. (Registration: This study was registered with ClinicalTrials.gov (NCT06342622) on March 15, 2024.).

Comparing ‘fair’ machine learning models for detecting at-risk online gamblers

ABSTRACT Researchers have worked to develop machine learning models that detect at-risk online gamblers, enabling personalized harm prevention tools. However, existing research has not evaluated these models’ potential to reinforce or amplify sociodemographic biases leading to treatment disparity, a recognized issue in the machine learning field. We sought to develop and compare three examples of potentially fair models using online gambling data. In two large samples of transaction data from a provincially owned Canadian gambling website (N1 = 9,145, N2 = 10,716), we developed three machine learning models based on competing concepts of fairness: fairness via unawareness, classification parity, and outcome calibration. We hypothesized that significant relationships existed between reporting a high risk of past-year gambling problems (the dependent variable) and participants’ age and sex. Further, we hypothesized that the three ‘fair’ models would show differing levels of classification performance both in aggregate and within sociodemographic groups. Significant age and sex effects were found, refuting the fairness via unawareness modeling strategy. Superiority across all performance metrics was not present for either of the remaining models. For the fairest practices in any jurisdiction, classification parity and outcome calibration models should be tested in situ, and incorporate the perspectives and preferences of end users who will be affected.

A Distributed VMD-BiLSTM Model for Taxi Demand Forecasting with GPS Sensor Data.

With the ubiquitous deployment of mobile and sensor technologies in modes of transportation, taxis have become a significant component of public transportation. However, vacant taxis represent an important waste of transportation resources. Forecasting taxi demand within a short time achieves a supply-demand balance and reduces oil emissions. Although earlier studies have forwarded highly developed machine learning- and deep learning-based models to forecast taxicab demands, these models often face significant computational expenses and cannot effectively utilize large-scale trajectory sensor data. To address these challenges, in this paper, we propose a hybrid deep learning-based model for taxi demand prediction. In particular, the Variational Mode Decomposition (VMD) algorithm is integrated along with a Bidirectional Long Short-Term Memory (BiLSTM) model to perform the prediction process. The VMD algorithm is applied to decompose time series-aware traffic features into multiple sub-modes of different frequencies. After that, the BiLSTM method is utilized to predict time series data fed with the relevant demand features. To overcome the limitation of high computational expenses, the designed model is performed on the Spark distributed platform. The performance of the proposed model is tested using a real-world dataset, and it surpasses existing state-of-the-art predictive models in terms of accuracy, efficiency, and distributed performance. These findings provide insights for enhancing the efficiency of passenger search and increasing the profit of taxicabs.

Machine Learning-Based Prediction for High Health Care Utilizers by Using a Multi-Institutional Diabetes Registry: Model Training and Evaluation.

The cost of health care in many countries is increasing rapidly. There is a growing interest in using machine learning for predicting high health care utilizers for population health initiatives. Previous studies have focused on individuals who contribute to the highest financial burden. However, this group is small and represents a limited opportunity for long-term cost reduction. We developed a collection of models that predict future health care utilization at various thresholds. We utilized data from a multi-institutional diabetes database from the year 2019 to develop binary classification models. These models predict health care utilization in the subsequent year across 6 different outcomes: patients having a length of stay of ≥7, ≥14, and ≥30 days and emergency department attendance of ≥3, ≥5, and ≥10 visits. To address class imbalance, random and synthetic minority oversampling techniques were employed. The models were then applied to unseen data from 2020 and 2021 to predict health care utilization in the following year. A portfolio of performance metrics, with priority on area under the receiver operating characteristic curve, sensitivity, and positive predictive value, was used for comparison. Explainability analyses were conducted on the best performing models. When trained with random oversampling, 4 models, that is, logistic regression, multivariate adaptive regression splines, boosted trees, and multilayer perceptron consistently achieved high area under the receiver operating characteristic curve (>0.80) and sensitivity (>0.60) across training-validation and test data sets. Correcting for class imbalance proved critical for model performance. Important predictors for all outcomes included age, number of emergency department visits in the present year, chronic kidney disease stage, inpatient bed days in the present year, and mean hemoglobin A1c levels. Explainability analyses using partial dependence plots demonstrated that for the best performing models, the learned patterns were consistent with real-world knowledge, thereby supporting the validity of the models. We successfully developed machine learning models capable of predicting high service level utilization with strong performance and valid explainability. These models can be integrated into wider diabetes-related population health initiatives.

Synthetic data in generalizable, learning-based neuroimaging

Abstract Synthetic data has emerged as an attractive option for developing machine learning methods in human neuroimaging, particularly in magnetic resonance imaging (MRI)— a modality where image contrast depends enormously on acquisition hardware and parameters. This retrospective paper reviews a family of recently proposed methods, based on synthetic data, for generalizable machine learning in brain MRI analysis. Central to this framework is the concept of domain randomization, which involves training neural networks on a vastly diverse array of synthetically generated images with random contrast properties. This technique has enabled robust, adaptable models that are capable of handling diverse MRI contrasts, resolutions, and pathologies, while working out-of-the-box, without retraining. We have successfully applied this method to tasks such as whole brain segmentation (SynthSeg), skull-stripping (SynthStrip), registration (SynthMorph, EasyReg), super-resolution and MR contrast transfer (SynthSR). Beyond these applications, the paper discusses other possible use cases and future work in our methodology. Neural networks trained with synthetic data enable the analysis of clinical MRI, including large retrospective datasets, while greatly alleviating (and sometimes eliminating) the need for substantial labeled datasets, and offer enormous potential as robust tools to address various research goals.

Machine learning classification of archaea and bacteria identifies novel predictive genomic features

BackgroundArchaea and Bacteria are distinct domains of life that are adapted to a variety of ecological niches. Several genome-based methods have been developed for their accurate classification, yet many aspects of the specific genomic features that determine these differences are not fully understood. In this study, we used publicly available whole-genome sequences from bacteria (N=2546\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$N=2546$$\\end{document}) and archaea (N=109\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$N=109$$\\end{document}). From these, a set of genomic features (nucleotide frequencies and proportions, coding sequences (CDS), non-coding, ribosomal and transfer RNA genes (ncRNA, rRNA, tRNA), Chargaff’s, topological entropy and Shannon’s entropy scores) was extracted and used as input data to develop machine learning models for the classification of archaea and bacteria.ResultsThe classification accuracy ranged from 0.993 (Random Forest) to 0.998 (Neural Networks). Over the four models, only 11 examples were misclassified, especially those belonging to the minority class (Archaea). From variable importance, tRNA topological and Shannon’s entropy, nucleotide frequencies in tRNA, rRNA and ncRNA, CDS, tRNA and rRNA Chargaff’s scores have emerged as the top discriminating factors. In particular, tRNA entropy (both topological and Shannon’s) was the most important genomic feature for classification, pointing at the complex interactions between the genetic code, tRNAs and the translational machinery.ConclusionstRNA, rRNA and ncRNA genes emerged as the key genomic elements that underpin the classification of archaea and bacteria. In particular, higher nucleotide diversity was found in tRNA from bacteria compared to archaea. The analysis of the few classification errors reflects the complex phylogenetic relationships between bacteria, archaea and eukaryotes.

Thermal, mechanical, and electrical properties of Si-stacked nanosheet transistors using machine learning interatomic potentials

Thermal and mechanical properties play a key role in optimizing the performance of nanoelectronic devices. In this study, the lattice thermal conductivity (κL) and elastic constants of Si nanosheets at different sheet thicknesses were determined using recently developed machine learning interatomic potentials (MLIPs). A Si nanosheet with a minimum thickness of 10 atomic layers was used for model training to predict the properties of sheets with greater thicknesses. The training dataset was efficiently constructed using stochastic sampling of the potential energy surface (PES). Density functional theory (DFT) calculations were used to extract
the MLIP, which served as the basis for further analysis. The Moment Tensor Potential (MTP) method was used to obtain the MLIP in this study. The results showed that, at sub-6 nm sheet thickness, the thermal conductivity dropped to ∼ 7 % of its bulk value, whereas some stiffness tensor components dropped to ∼ 3 % of the bulk values. These findings contribute to the understanding of heat transport and mechanical behavior of ultrathin Si nanosheets, which is crucial for designing and optimizing nanoelectronic devices. The technological implications of the extracted parameters on nanosheet field-effect transistor (NS-FET) performance at advanced technology nodes were evaluated using TCAD device simulations.

Machine Learning Models for Predicting Bioavailability of Traditional and Emerging Aromatic Contaminants in Plant Roots.

To predict the behavior of aromatic contaminants (ACs) in complex soil-plant systems, this study developed machine learning (ML) models to estimate the root concentration factor (RCF) of both traditional (e.g., polycyclic aromatic hydrocarbons, polychlorinated biphenyls) and emerging ACs (e.g., phthalate acid esters, aryl organophosphate esters). Four ML algorithms were employed, trained on a unified RCF dataset comprising 878 data points, covering 6 features of soil-plant cultivation systems and 98 molecular descriptors of 55 chemicals, including 29 emerging ACs. The gradient-boosted regression tree (GBRT) model demonstrated strong predictive performance, with a coefficient of determination (R2) of 0.75, a mean absolute error (MAE) of 0.11, and a root mean square error (RMSE) of 0.22, as validated by five-fold cross-validation. Multiple explanatory analyses highlighted the significance of soil organic matter (SOM), plant protein and lipid content, exposure time, and molecular descriptors related to electronegativity distribution pattern (GATS8e) and double-ring structure (fr_bicyclic). An increase in SOM was found to decrease the overall RCF, while other variables showed strong correlations within specific ranges. This GBRT model provides an important tool for assessing the environmental behaviors of ACs in soil-plant systems, thereby supporting further investigations into their ecological and human exposure risks.

Thermal transport in C6N7 monolayer: a machine learning based molecular dynamics study

The successful synthesis of a novel C6N7carbon nitride monolayer offers expansive prospects for applications in the fields of semiconductors, sensors, and gas separation technologies, in which the thermal transport properties of C6N7are crucial for optimizing the functionality and reliability of these applications. In this work, based on our developed machine learning potential (MLP), molecular dynamics (MD) simulations including homogeneous non-equilibrium, non-equilibrium, and their respective spectral decomposition methods are performed to investigate the effects of phonon transport, temperature, and length on the thermal conductivity of C6N7monolayer. Our results reveal that low-frequency and in-plane phonon modes dominate the thermal conductivity. Notably, thermal conductivity decreases with an increase in temperature due to temperature-induced increase in phonon-phonon scattering of in-plane phonon modes, while it increases with an extension in sample length. Our findings based on MD simulations with MLP contribute new insights into the lattice thermal conductivity of holey carbon nitride compounds, which is helpful for the development of next-generation electronic and photonic devices.

Machine Learning Model for Predicting Risk Factors of Prolonged Length of Hospital Stay in Patients with Aortic Dissection: a Retrospective Clinical Study.

The length of hospital stay (LOS) is crucial for assessing medical service quality. This study aimed to develop machine learning models for predicting risk factors of prolonged LOS in patients with aortic dissection (AD). The data of 516 AD patients were obtained from the hospital's medical system, with 111 patients in the prolonged LOS (> 30days) group based on three quarters of the LOS in the entire cohort. Given the screened variables and prediction models, the XGBoost model demonstrated superior predictive performance in identifying prolonged LOS, due to the highest area under the receiver operating characteristic curve, sensitivity, and F1-score in both subsets. The SHapley Additive exPlanation analysis indicated that high density lipoprotein cholesterol, alanine transaminase, systolic blood pressure, percentage of lymphocyte, and operation time were the top five risk factors associated with prolonged LOS. These findings have a guiding value for the clinical management of patients with AD.

Machine learning for predicting device-associated infection and 30-day survival outcomes after invasive device procedure in intensive care unit patients

This study aimed to preliminarily develop machine learning (ML) models capable of predicting the risk of device-associated infection and 30-day outcomes following invasive device procedures in intensive care unit (ICU) patients. The study utilized data from 8574 ICU patients who underwent invasive procedures, sourced from the Medical Information Mart for Intensive Care (MIMIC)-IV version 2.2 database. Patients were allocated into training and validation datasets in a 7:3 ratio. Seven ML models were employed for predicting device-associated infections, while five models were used for predicting 30-day survival outcomes. Model performance was primarily evaluated using the receiver operating characteristic (ROC) curve for infection prediction and the survival model’s concordance index (C-index). Top-performing models progressively reduced the number of variables based on their importance, thereby optimizing practical utility. The inclusion of all variables demonstrated that extreme gradient boosting (XGBoost) and extra survival trees (EST) models yielded superior discriminatory performance. Notably, when restricted to the top 10 variables, both models maintained performance levels comparable to when all variables were included. In the validation cohort, the XGBoost model, with the top 10 variables, achieved an area under the curve (AUC) of 0.810 (95% CI 0.808–0.812), an area under the precision-recall curve (AUPRC) of 0.226 (95% CI 0.222–0.230), and a Brier score (BS) of 0.053 (95% CI 0.053–0.054). The EST model, with the top 10 variables, reported a C-index of 0.756 (95% CI 0.754–0.757), a time-dependent AUC of 0.759 (95% CI 0.763–0.775), and an integrated Brier score (IBS) of 0.087 (95% CI 0.087–0.087). Both models are accessible via a web application. The internally evaluated XGBoost and EST models demonstrated exceptional predictive accuracy for device-associated infection risks and 30-day survival outcomes post-invasive procedures in ICU patients. Further validation is required to confirm the clinical utility of these two models in future studies.

Unraveling the Distribution of Black Carbon in Chinese Forest Soils Using Machine Learning Approaches

AbstractBlack carbon (BC) is a highly persistent yet poorly understood component of forest soil carbon reservoirs, while its inventory, distribution, and determining factors in forest soils on a large geographic scale remain unclear. Here, we characterized soil BC across 68 Chinese forest sites using benzene polycarboxylic acid method and developed machine learning (ML) models to predict and interpret potential impacts of soil organic matter (SOM) properties, soil physiochemical properties, meteorological conditions, wildfire history, and microbial diversity on BC. Results revealed that SOM properties were the most critical in predicting BC, complemented by the negative impact of mean annual temperature and alkaline mineral composition. The superior prediction accuracy for BC with higher condensed aromaticity (more benzene hexa‐ and penta‐carboxylic acid monomers) likely results from its simpler sources and greater resistance to transformation. This study introduces an effective ML model for predicting and interpreting soil BC inventory to better understand BC cycling.

Limits on inferring T cell specificity from partial information

A key challenge in molecular biology is to decipher the mapping of protein sequence to function. To perform this mapping requires the identification of sequence features most informative about function. Here, we quantify the amount of information (in bits) that T cell receptor (TCR) sequence features provide about antigen specificity. We identify informative features by their degree of conservation among antigen-specific receptors relative to null expectations. We find that TCR specificity synergistically depends on the hypervariable regions of both receptor chains, with a degree of synergy that strongly depends on the ligand. Using a coincidence-based approach to measuring information enables us to directly bound the accuracy with which TCR specificity can be predicted from partial matches to reference sequences. We anticipate that our statistical framework will be of use for developing machine learning models for TCR specificity prediction and for optimizing TCRs for cell therapies. The proposed coincidence-based information measures might find further applications in bounding the performance of pairwise classifiers in other fields.

Predicting 24-hour intraocular pressure peaks and averages with machine learning.

Predicting 24-hour peak and average intraocular pressure (IOP) is essential for the diagnosis and management of glaucoma. This study aimed to develop and assess a machine learning model for predicting 24-hour peak and average IOP, leveraging advanced techniques to enhance prediction accuracy. We also aimed to identify relevant features and provide insights into the prediction results to better inform clinical practice. In this retrospective study, electronic medical records from January 2014 to May 2024 were analyzed, incorporating 24-hour IOP monitoring data and patient characteristics. Predictive models based on five machine learning algorithms were trained and evaluated. Five time points (10:00 AM, 12:00 PM, 2:00 PM, 4:00 PM, and 6:00 PM) were tested to optimize prediction accuracy using their combinations. The model with the highest performance was selected, and feature importance was assessed using Shapley Additive Explanations. This study included data from 517 patients (1,034 eyes). For predicting 24-hour peak IOP, the Random Forest Regression (RFR) model utilizing IOP values at 10:00 AM, 12:00 PM, 2:00 PM, and 4:00 PM achieved optimal performance: MSE 5.248, RMSE 2.291, MAE 1.694, and R2 0.823. For predicting 24-hour average IOP, the RFR model using IOP values at 10:00 AM, 12:00 PM, 4:00 PM, and 6:00 PM performed best: MSE 1.374, RMSE 1.172, MAE 0.869, and R2 0.918. The study developed machine learning models that predict 24-hour peak and average IOP. Specific time point combinations and the RFR algorithm were identified, which improved the accuracy of predicting 24-hour peak and average intraocular pressure. These findings provide the potential for more effective management and treatment strategies for glaucoma patients.

Fractal dimension and lacunarity measures of glioma subcomponents are discriminative of the grade of gliomas and IDH status.

Since the overall glioma mass and its subcomponents-enhancing region (malignant part of the tumor), non-enhancing (less aggressive tumor cells), necrotic core (dead cells), and edema (water deposition)-are complex and irregular structures, non-Euclidean geometric measures such as fractal dimension (FD or "fractality") and lacunarity are needed to quantify their structural complexity. Fractality measures the extent of structural irregularity, while lacunarity measures the spatial distribution or gaps. The complex geometric patterns of the glioma subcomponents may be closely associated with the grade and molecular landscape. Therefore, we measured FD and lacunarity in the glioma subcomponents and developed machine learning models to discriminate between tumor grades and isocitrate dehydrogenase (IDH) gene status. 3D fractal dimension (FD3D) and lacunarity (Lac3D) were measured for the enhancing, non-enhancing plus necrotic core, and edema-subcomponents using preoperative structural-MRI obtained from the The Cancer Genome Atlas (TCGA) and University of California San Francisco Preoperative Diffuse Glioma MRI (UCSF-PDGM) glioma cohorts. The FD3D and Lac3D measures of the tumor-subcomponents were then compared across glioma grades (HGGs: high-grade gliomas vs. LGGs: low-grade gliomas) and IDH status (mutant vs. wild type). Using these measures, machine learning platforms discriminative of glioma grade and IDH status were developed. Kaplan-Meier survival analysis was used to assess the prognostic significance of FD3D and Lac3D measurements. HGG exhibited significantly higher fractality and lower lacunarity in the enhancing subcomponent, along with lower fractality in the non-enhancing subcomponent compared to LGG. This suggests that a highly irregular and complex geometry in the enhancing-subcomponent is a characteristic feature of HGGs. A comparison of FD3D and Lac3D between IDH-wild type and IDH-mutant gliomas revealed that mutant gliomas had ~2.5-fold lower FD3D in the enhancing-subcomponent and higher FD3D with lower Lac3D in the non-enhancing subcomponent, indicating a less complex and smooth enhancing subcomponent, and a more continuous non-enhancing subcomponent as features of IDH-mutant gliomas. Supervised ML models using FD3D from both the enhancing and non-enhancing subcomponents together demonstrated high-sensitivity in discriminating glioma grades (~97.9%) and IDH status (~94.4%). A combined fractal estimation of the enhancing and non-enhancing subcomponents using MR images could serve as a non-invasive, precise, and quantitative measure for discriminating glioma grade and IDH status. The combination of 2-hydroxyglutarate-magnetic resonance spectroscopy (2HG-MRS) with FD3D and Lac3D quantification may be established as a robust imaging signature for glioma subtyping.

Practical application of machine learning in energy and thermal management: Long-term data analysis of solar-assisted AC systems in portable cabins in Kuwait and Australia

Machine learning techniques are advancing rapidly in theoretical grounds, but their practical application and real evaluation in engineering disciplines are limited. This study investigates one-year of experimental data for indoor air conditions and energy monitoring of two identical portable cabins in Kuwait. Additionally, a 9-month period of solar photovoltaic energy production is experimentally obtained for assessing energy saving aspects of a solar assisted air conditioning system in one of the cabins. A transient system simulation tool model is developed and validated for the portable cabins. Both one-year experimental data and the validated simulation data are used for developing machine learning models for six case studies in Kuwait and Australia. In simulations, there are issues related to lack of access to real-time weather data from a locally installed weather station. In experimental works, there were issues related to disruption of data collection due to power or internet shutdowns. To resolve these issues, nineteen regression models are investigated. It was found that all models performed very well although the Matern 5/2 and Exponential gaussian process regression model performed slightly better, with all models achieving high accuracy, as indicated by R-squared values close to 1.0 in all six case studies. Overall, it was found that the photovoltaic solar-assisted air conditioning system could save annually 26.2 % and 75.7 % energy in Kuwait and Rockhampton, Australia, respectively.

Machine Learning-Based Investigation of Atomic Packing Effects: Chemical Pressures at the Extremes of Intermetallic Complexity.

Intermetallic phases represent a domain of emergent behavior, in which atoms with packing and electronic preferences can combine into complex geometrical arrangements whose long-range order involves repeat patterns containing thousands of atoms or is incompatible with a 3D unit cell. The formation of such arrangements points to unexplained driving forces within these systems that, if understood, could be harnessed in the design of new metallic materials. DFT-chemical pressure (CP) analysis has emerged as an approach to visualize how atomic packing tensions within simpler crystal structures can drive this complexity and create potential functionality. However, the applications of this method have hitherto been limited in scope by its dependence on resource-intensive electronic structure calculations. In this Article, we develop machine learning (ML)-based implementation of the CP approach, drawing on the collection of DFT-CP schemes in the Intermetallic Reactivity Database. We illustrate the method with comparisons of ML-CP and DFT-CP schemes for a series of examples, before demonstrating its application with an exploration of one of the quintessential instances of complexity in intermetallic chemistry, Mg2Al3, whose high-temperature unit cell is a 2.8 nm cube containing 1227 atoms. An analysis of its ML-CP-derived interatomic pressures traces the origins of the structure to simple matching rules for the assembly of Frank-Kasper polyhedra. The ML-CP model can be immediately employed on other intermetallic systems, through either its web interface or a command-line version, with just a crystallographic information file.

WiFi-Based Human Identification with Machine Learning: A Comprehensive Survey.

In the modern world of human-computer interaction, notable advancements in human identification have been achieved across fields like healthcare, academia, security, etc. Despite these advancements, challenges remain, particularly in scenarios with poor lighting, occlusion, or non-line-of-sight. To overcome these limitations, the utilization of radio frequency (RF) wireless signals, particularly wireless fidelity (WiFi), has been considered an innovative solution in recent research studies. By analyzing WiFi signal fluctuations caused by human presence, researchers have developed machine learning (ML) models that significantly improve identification accuracy. This paper conducts a comprehensive survey of recent advances and practical implementations of WiFi-based human identification. Furthermore, it covers the ML models used for human identification, system overviews, and detailed WiFi-based human identification methods. It also includes system evaluation, discussion, and future trends related to human identification. Finally, we conclude by examining the limitations of the research and discussing how researchers can shift their attention toward shaping the future trajectory of human identification through wireless signals.

DSA Quantitative Analysis and Predictive Modeling of Obliteration in Cerebral AVM following Stereotactic Radiosurgery.

Stereotactic radiosurgery is a key treatment modality for cerebral AVMs, particularly for small lesions and those located in eloquent brain regions. Predicting obliteration remains challenging due to evolving treatment paradigms and complex AVM presentations. With digital subtraction angiography (DSA) being the gold standard for outcome evaluation, radiomic approaches offer potential for more objective and detailed analysis. We aimed to develop machine learning modeling using DSA quantitative features for post-SRS obliteration prediction. A prospective registry of patients with cerebral AVMs was screened to include patients with digital prestereotactic radiosurgery DSA. Anterior-posterior and lateral views were retrieved and manually segmented. Quantitative features were computed from the lesion ROI. Following feature selection, machine learning models were developed to predict unsuccessful 2-year total obliteration using processed radiomics features in comparison with clinical and radiosurgical features. When we evaluated through area under the receiver operating characteristic curve (AUROC), accuracy, area under the precision-recall curve F1, recall, and precision, the best performing model predictions on the test set were interpreted using the Shapley additive explanations approach. DSA images of 100 included patients were retrieved and analyzed. The best-performing clinical radiosurgical model was a gradient boosting classifier with an AUROC of 68% and a recall of 67%. When we used radiomics variables as input, the AdaBoost classifier had the best evaluation metrics with an AUROC of 79% and a recall of 75%. The most important clinico-radiosurgical features, ranked by model contribution, were lesion volume, patient age, treatment dose rate, the presence of seizure at presentation, and prior resection. The most important ranked radiomics features were the following: gray-level size zone matrix, gray-level nonuniformity, kurtosis, sphericity, skewness, and gray-level dependence matrix dependence nonuniformity. The combination of radiomics with machine learning is a promising approach for predicting cerebral AVM obliteration status following stereotactic radiosurgery. DSA could enhance prognostication of stereotactic radiosurgery-treated AVMs due to its high spatial resolution. Model interpretation is essential for building transparent models and establishing clinically valid radiomic signatures.