HFTC: a hierarchical fungal taxonomic classification model for ITS sequences using low-dimensional embedding features

Motion trajectory recognition is important for characterizing the moving property of an object. The speed and accuracy of trajectory recognition rely on a compact and discriminative feature representation, and the situations of varying rotation, scaling and translation has to be specially considered. In this paper we propose a novel feature extraction method for trajectories. Firstly a trajectory is represented by a proposed point context, which is a rotation-scale-translation (RST) invariant shape descriptor with a flexible tradeoff between computational complexity and discrimination, yet we prove that it is a complete shape descriptor. Secondly, the shape context is nonlinearly mapped to a subspace by kernel nonparametric discriminant analysis (KNDA) to get a compact feature representation, and thus a trajectory is projected to a single point in a low-dimensional feature space. Experimental results show that, the proposed trajectory feature shows encouraging improvement than state-of-art methods.

Anemia is one of the common adverse reactions after hip fracture surgery. The traditional method to solve anemia is allogeneic transfusion. However, the transfusion may lead to some complications such as septicemia and fever. So far, few studies have reported roles of machine learning in predicting whether blood transfusion is needed or not after hip fracture surgery. Therefore, the purpose of this study is to develop machine learning models to predict the likelihood of postoperative blood transfusion in patients undergoing hip fracture surgery. This study enrolled 1355 patients who underwent hip fracture surgery at the Affiliated Hospital of Qingdao University from January 2016 to December 2021. Among all patients, 210 cases received postoperative blood transfusion. All patients were randomly divided into a training group and a testing group at a ratio of 7:3. In the training group, univariate and multivariate logistic regression analyses were used to determine independent risk factors for the postoperative transfusion. Then, based on these independent risk factors, tenfold cross-validation method was utilized to develop five machine learning models, including logistic, multilayer perceptron (MLP), extreme gradient boosting (XGBoost), random forest (RF), and support vector machine (SVM). The receiver operating characteristic (ROC) curve, area under ROC curve (AUC), and Matthews correlation coefficient (MCC) were generated to evaluate the performance of the models. Calibration plot and decision curve analysis (DCA) were used to test the performance, stability, and clinical applicability of the models. The models were validated using the testing group; and the ROC curve, MCC, calibration plot, and DCA curves were also generated to validate the performance, stability, and clinical applicability of the models. To further verify the robustness of the model, we randomly grabbed 70% of the samples in the testing set, performed 1000 iterations, and calculated the AUC and confidence interval of the five models. Finally, we used SHapley Additive exPlanations (SHAP) to explain these models. Multivariate logistic regression analysis showed that there were 8 independent risk factors, including age, blood transfusion history, albumin (ALB), globulin (GLO), total bilirubin (TBIL), indirect bilirubin (IBIL), hemoglobin (HB), and blood loss > 200ml. We finally selected five independent risk factors including HB, GLO, age, IBIL, and blood loss > 200ml. Based on these five independent risk factors, we generated six characteristic variables, namely HB, HB × HB, HB × blood loss, GLO × HB, age, age × IBIL, and established five machine learning models using a tenfold cross-validation method. In the training group, the AUC values of logistic, RF, MLP, SVM, and XGB were 0.9320, 0.8911, 0.9327, 0.9225, and 0.8825, respectively, and the average AUC was 0.9122 ± 0.0212. The MCC values were 0.65, 0.77, 0.65, 0.66, and 0.68, respectively, and the calibration plot and DCA performed well. In the testing group the AUC values of logistic, RF, MLP, SVM, and XGB were 0.8483, 0.7978, 0.8576, 0.8598, and 0.8216, respectively. The average AUC was 0.8370 ± 0.0238, and the MCC values were 0.41, 0.35, 0.40, 0.41, and 0.41, respectively. The calibration plot and DCA in the testing group also showed good performance. The AUC values and confidence intervals of the 1000-iteration model were: logistic (AUC, min confidence interval [CI]-max confidence interval [CI] 0.848, 0.804-0.903), RF (AUC, minCI-maxCI 0.797, 0.734-0.857), MLP (AUC, minCI-maxCI 0.858, 0.812-0.902), SVM (AUC, minCI-maxCI 0.859, 0.819-0.910), and XGB (AUC, minCI-maxCI 0.821, 0.764-0.894). The model performed well. Finally, according to SHAP, among all five models, HB played the most important role in model prediction and interpretation. The five models we developed all performed well in predicting the likelihood of blood transfusion after hip fracture surgery. Therefore, we believed that the prediction model based on machine learning had great application prospects in clinical practice, which could help clinicians better predict the risk of blood transfusion after hip fracture surgery.

Integrating near-infrared hyperspectral imaging with machine learning and feature selection: Detecting adulteration of extra-virgin olive oil with lower-grade olive oils and hazelnut oil

Objective: In this study, it was aimed to find the method with high classification success among the methods used in the study by comparing the supervised machine learning methods according to the classification performance. Material and Methods: In our study, both the real data set obtained from 302 patients with invasive ductal carcinoma and 24 different data sets obtained by simulation were used to compare the classification performance of support vector machines, random forest and artificial neural networks. The success of classifications of the methods used was compared according to the general accuracy, F-measure, Matthews correlation coefficient, area under the curve (AUC) and discriminant power in breast cancer data. In addition, the difference in training-test accuracy in the simulation data and the significance of this difference were also evaluated. Results: The highest survival classification accuracy (80%) for the test set of stage III patients with invasive ductal carcinoma was obtained from support vector machines (SVM) with the radial-based kernel. The highest values in other performance metrics (F-measure=0.87, Matthews correlation coefficient=0.22, AUC=0.89 and discriminant power=0.52), and the most successful results in simulation data were generally obtained from SVM. Conclusion: SVM had higher accuracy in both the real data set and simulation data than random forest and artificial neural networks.

Assessment of socioeconomic and demographic risk factors for low birth weight using model-agnostic explainable ensembles.

© 2019 Nguyen-Vo et al. Background: Pseudouridine modification is most commonly found among various kinds of RNA modification occurred in both prokaryotes and eukaryotes. This biochemical event has been proved to occur in multiple types of RNAs, including rRNA, mRNA, tRNA, and nuclear/nucleolar RNA. Hence, gaining a holistic understanding of pseudouridine modification can contribute to the development of drug discovery and gene therapies. Although some laboratory techniques have come up with moderately good outcomes in pseudouridine identification, they are costly and required skilled work experience. We propose iPseU-NCP - an efficient computational framework to predict pseudouridine sites using the Random Forest (RF) algorithm combined with nucleotide chemical properties (NCP) generated from RNA sequences. The benchmark dataset collected from Chen et al. (2016) was used to develop iPseU-NCP and fairly compare its performances with other methods. Results: Under the same experimental settings, comparing with three state-of-the-art methods including iPseU-CNN, PseUI, and iRNA-PseU, the Matthew's correlation coefficient (MCC) of our model increased by about 20.0%, 55.0%, and 109.0% when tested on the H. sapiens (H_200) dataset and by about 6.5%, 35.0%, and 150.0% when tested on the S. cerevisiae (S_200) dataset, respectively. This significant growth in MCC is very important since it ensures the stability and performance of our model. With those two independent test datasets, our model also presented higher accuracy with a success rate boosted by 7.0%, 13.0%, and 20.0% and 2.0%, 9.5%, and 25.0% when compared to iPseU-CNN, PseUI, and iRNA-PseU, respectively. For majority of other evaluation metrics, iPseU-NCP demonstrated superior performance as well. Conclusions: iPseU-NCP combining the RF and NPC-encoded features showed better performances than other existing state-of-the-art methods in the identification of pseudouridine sites. This also shows an optimistic view in addressing biological issues related to human diseases.

To ensure safety, autonomy, and context-aware control, reliable state recognition is still a challenge for stair-climbing wheelchairs, which offer mobility in environments with steps, curbs, and landings. This study uses instantaneous Inertial Measurement Unit (IMU) data to develop a simplified, window-free method for classifying wheelchair stair-related activities. Instead of sliding-window preprocessing or temporal sequence modeling, this study uses an 18-channel feature set that includes orientation, gyroscope, accelerometer, magnetometer, linear acceleration, and gravity signals. Stratified evaluation and several metrics, such as accuracy, macro-F1, Matthews Correlation Coefficient (MCC), ROC-AUC, and per-class precision-recall, were used to systematically benchmark eight classifiers: Multinomial Logistic Regression (MLR), Gaussian Naïve Bayes (GNB), K-Nearest Neighbors (KNN), Decision Tree (DT), Random Forest (RF), RBF-kernel Support Vector Classifier (SVC), a compact Multilayer Perceptron (MLP), and an adaptive CatBoost configuration. The results show that CatBoost performed almost flawlessly (Accuracy = 0.999, MCC = 0.999), closely followed by compact MLP. RF, KNN, and SVC formed a solid middle tier. In a window-free regime, feature importance analysis showed that instantaneous gyroscope and linear acceleration made very little contribution, while orientation and magnetometer channels were found to be the most crucial features. These results show that accurate and computationally efficient recognition of stair-climbing wheelchair states is possible by features driven by posture and heading. The proposed method facilitates low-latency embedded deployment and identifies areas for future development, such as lightweight temporal enhancements, angle encoding, and magnetometer calibration.

Using animal-mounted sensor technology and machine learning to predict time-to-calving in beef and dairy cows

The EGNOS NOTAM proposals provision is a key enabler for the publication of RNP approaches down to LPV minima in Europe, as it provides the different AIS (Aeronautical Information Service) the required information to establish a NOTAM service covering this type of procedures, in line with the ICAO standards. Currently, more than 400 APV-I and 150 LPV-200 SBAS procedures are published in Austria, Belgium, Croatia, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Lithuania, Luxembourg, Malta, Montenegro, Netherlands, Norway, Poland, Portugal, Romania, Serbia, Slovak Republic, Spain, Sweden, Switzerland, UK and the Bailiwick of Guernsey. The EURONOTAM tool is responsible for generating EGNOS NOTAM proposals, implementing an analytical service volume that propagates the orbits of the satellites (based on the GPS almanacs) and computes the satellites in view for each configured airport. Once the satellite visibility is computed, the prediction module estimates the residual errors for each Line Of Sight (LoS), taking into account the number of EGNOS Ranging Integrity Monitoring Stations (RIMS) in view (for the User Differential Range Error —UDRE— computation) and the number of Ionospheric Pierce Points surrounding the neighbouring Ionospheric Grid Points (for the Grid Ionospheric Vertical Error —GIVE— estimation). With these estimations, the tool calculates the Protection Levels for each airport and determines if a new NOTAM proposal is to be issued in case the service is not available to support APV-I or LPV-200 operations in a specific location. Thus, the accuracy of the UDRE and GIVE estimations is essential in order to precisely predict the EGNOS availability events. Taking this into account, the present study is aimed at analysing the feasibility and the potential improvement that could be obtained, by using Machine Learning (ML) or Deep Learning (DL) techniques to enhance or replace the current analytical model. More specifically, the different “classic” ML models which will be addressed are decision trees, random forests, support vector machines and extreme gradient boosting (XGBoost). On top of these ML models, DL techniques are also covered by this study. The outputs of these models will be used to estimate the Protection Levels, and its applicability to the current EURONOTAM software will be discussed. The study will start with the analysis of the input dataset used to train the model. Aimed at analysing the impact of EGNOS station outages on the system performance and the capacity of the model to predict such impact, the input dataset will be composed of a minimum of 41 scenarios, including single and multiple station outages. For the UDRE estimation, more than 1.1 million samples, including the following features, will be considered: satellite coordinates, satellite dilution of precision (DOP) (based on the number of station in view), number of RIMS-A and RIMS-C on sight, satellite family, slot number and RIMS status. Similar features will be considered for the GIVE, with more than 3.5 million samples, including as well information on the number and distribution of the IPPs surrounding each IGP. Feature importance will be evaluated using SHAP values. SHAP values are calculated for each input feature and for each sample, and they measure the contribution of that feature to the predicted value for the sample. Summing the absolute SHAP values for all samples gives a measure of how the output is driven by that feature globally. This will be used to identify which are the most important features (and eventually, consider updating EURONOTAM model to consider these features – e.g. DOP) and discard the unimportant ones from the ML model. The real UDREi and GIVEi values used for the model training and testing will be obtained from the real EGNOS messages downloaded from the EDAS FTP. After preparing the input data set and splitting it into the training and testing datasets, the model hyperparameters will be tuned through a grid search with 5 fold cross-validation (for each set of values, 4/5 of the original training set will be used for actually training the model and the remaining 1/5 will be used for assessing the performance). The parameters resulting in the best averaged model performance will be selected. The resulting model will be evaluated with the testing dataset, using the following metrics: - Binary Accuracy: ratio between number of samples correctly classified as monitored or not-monitored vs the total number of samples - Total Accuracy: ratio between the number of correctly classified samples (considering the 16 possible labels) and the total number of samples. - Matthews correlation coefficient (MCC): A coefficient of +1 represents a perfect prediction and 0 represents an average random prediction. - Mean & RMS error on UDRE or GIVE value (UDREi / GIVEi converted to UDRE /GIVE for the assessment). The results obtained so far seem to indicate that Machine learning models can significantly improve the estimates obtained from the current analytical model (geometric-based). For instance, for one of the ML models tested, the system is able to correctly classify the 86% of the UDREi labels versus the 76% obtained with the current model, meaning a 13% improvement. In terms of binary accuracy, the ML model properly estimates the monitored / not-monitored status for the 99% of the samples tested. The RMS and mean absolute error are decreased by a 45% and 55% respectively. For the GIVE estimation, the improvement in the classification problem is more remarkable, with a 30% enhancement with respect to the current value. The RMS and mean absolute errors decrease as well by a 36% and a 50% factor respectively. The study will determine if Machine learning models can significantly improve the estimates obtained from the current analytical model (geometric-based). The study will also evaluate Protection Levels predicted with the usage of ML-estimated UDRE and GIVE values, where an important improvement in terms of prediction accuracy is expected. The analysis will be completed with a brief discussion on the available software libraries enabling the embedding of ML trained models in existing code and the capability of the model to predict situations that are very different from the ones actually trained. This is particularly challenging in this study, due to the lack of a full set of samples covering all the non-nominal conditions, since potential future situations might have never happened before and cannot be therefore, trained. Future steps will focus on this topic, and aim at finding more robust machine learning models that are able to deal with non-nominal conditions while still outperforming traditional methods under nominal conditions.

As the reliance on digital networks grows, the need for robust security measures becomes paramount. It offers a thorough analysis of creating and assessing a Network Intrusion Detection System (NIDS) incorporating various machine learning algorithms. The selected algorithms consist of K Nearest Neighbour, Naïve Bayes, Decision Tree, Support Vector Machine, and Logistic Regression. Moreover, widely recognised ensemble learning models like CatBoost, AdaBoost, Gradient Boost, Random Forest, and Extreme Gradient Boost. The core objective is to improve the efficiency of intrusion detection by evaluating various machine-learning models. Utilizes feature selection techniques based on chi-square to enhance the relevance and effectiveness of the selected features. The evaluation metrics, particularly Matthews’ Correlation Coefficient (MCC) and ROC Curves, are used in assessing the models’ efficacy. They offer valuable insights into the models’ capability to differentiate between normal and malicious network activities. This study aims to advance cybersecurity by comparing classical machine learning and ensemble models for network intrusion detection. Significantly, our results underscore XGBoost’s effectiveness with 0.996 accuracy and 0.993 MCC, in enhancing digital network security.

Background Early detection of oppositional defiant behavior is warranted for timely intervention in children at risk. This study aimed to build a predictive model of persistent oppositional defiant behavior based on a machine learning algorithm. Methods With nationwide cohort data collected from 2012 to 2017, a tree-based ensemble model, random forest, was exploited to build a predictive model for persistent oppositional defiant behavior. The persistent oppositional defiant behavior was defined by the presence of oppositional defiant behavior for all the five years. The area under the receiver operating characteristic curve (AUC), overall accuracy, sensitivity, specificity, and Matthew’s correlation coefficients (MCC) were measured. Results Data of 1,323 children were used for building the machine learning-based predictive model. The baseline mean ± standard deviation month-age of the participants was 51.0 ± 1.2 months. The proportion of persistent oppositional defiant behavior was 0.98% (13/1323). In the hold-out test set, the overall accuracy, AUC, sensitivity, specificity, and MCC were 0.955, 0.982, 1.000, 0.954, and 0.417, respectively. Conclusion Our study demonstrated that the machine learning-based approach is useful for predicting persistent oppositional defiant behavior in preschool-aged children.

RNA editing is a post-transcriptional RNA process that provides RNA and protein complexity for regulating gene expression in eukaryotes. It is challenging to predict RNA editing by computational methods. In this study, we developed a novel method to predict RNA editing based on a random forest method. A careful feature selection procedure was performed based on the Maximum Relevance Minimum Redundancy (mRMR) and Incremental Feature Selection (IFS) algorithms. Eighteen optimal features were selected from the 77 features in our dataset and used to construct a final predictor. The accuracy and MCC (Matthews correlation coefficient) values for the training dataset were 0.866 and 0.742, respectively; for the testing dataset, the accuracy and MCC were 0.876 and 0.576, respectively. The performance was higher using 18 features than all 77, suggesting that a small feature set was sufficient to achieve accurate prediction. Analysis of the 18 features was performed and may shed light on the mechanism and dominant factors of RNA editing, providing a basis for future experimental validation.

In the field of network security, botnet attacks pose a significant challenge, exploiting networks of infected devices to launch sophisticated threats. As these attacks evolve, the need for effective detection methods becomes increasingly critical. This study proposes an explainable Machine Learning (ML) model that aims to identify and categorize botnet attacks, and it investigates the efficiency of different Explainable Artificial Intelligence (XAI) techniques for Intrusion Detection Systems (IDS). To train our models, we employed a subset of the NCC-2 dataset, which includes a mix of normal traffic and seven different types of botnet attacks across three sensors. The ML techniques selected for this research are Random Forest (RF), Extra Trees (ET), Decision Tree (DT), and K-Nearest Neighbors (KNN), with a GridSearch cross-validation approach for optimal hyperparameter tuning. We also explored the effects of class balance through Synthetic Minority Oversampling (SMOTE) and Random Undersampling. The models’ performance was rigorously tested using the Matthews Correlation Coefficient (MCC) and Macro F1-score, with the ET model demonstrating superior results of 99% MCC and 97% F1-score, respectively. To enhance the interpretability of the ET model’s decision-making process, we integrated three XAI techniques: SHapley Additive exPlanations (SHAP), Dalex, and Local Interpretable Model-agnostic Explanations (LIME), and evaluated their efficiency. The LIME and Dalex techniques showed efficient construction times for IDS.

Citrullination, an important post-translational modification of proteins, alters the molecular weight and electrostatic charge of the protein side chains. Citrulline, in protein sequences, is catalyzed by a class of Peptidyl Arginine Deiminases (PADs). Dependent on Ca2+, PADs include five isozymes: PAD 1, 2, 3, 4/5, and 6. Citrullinated proteins have been identified in many biological and pathological processes. Among them, abnormal protein citrullination modification can lead to serious human diseases, including multiple sclerosis and rheumatoid arthritis. It is important to identify the citrullination sites in protein sequences. The accurate identification of citrullination sites may contribute to the studies on the molecular functions and pathological mechanisms of related diseases. In this study, after an encoded training set (containing 116 positive and 348 negative samples) into the feature matrix, the mRMR method was used to analyze the 941- dimensional features which were sorted on the basis of their importance. Then, a predictive model based on a self-normalizing neural network (SNN) was proposed to predict the citrullination sites in protein sequences. Incremental Feature Selection (IFS) and 10-fold cross-validation were used as the model evaluation method. Three classical machine learning models, namely random forest, support vector machine, and k-nearest neighbor algorithm, were selected and compared with the SNN prediction model using the same evaluation methods. SNN may be the best tool for citrullination site prediction. The maximum value of the Matthews Correlation Coefficient (MCC) reached 0.672404 on the basis of the optimal classifier of SNN. The results showed that the SNN-based prediction methods performed better when evaluated by some common metrics, such as MCC, accuracy, and F1-Measure. SNN prediction model also achieved a better balance in the classification and recognition of positive and negative samples from datasets compared with the other three models.

Landslides are among the most hazardous geomorphological processes, especially in mountainous terrains where they threaten lives, disrupt infrastructure, and degrade ecosystems. The Chamoli district in Uttarakhand, India, with its rugged topography, erratic rainfall, and anthropogenic disturbances, is highly susceptible to such events. This research focuses on generating a detailed landslide susceptibility map for Chamoli using four machine learning algorithms Naïve Bayes (NB), K-Nearest Neighbors (KNN), Random Forest (RF), and Extreme Gradient Boosting (XGBoost). A total of sixteen causative factors were selected and processed using Geographic Information System (GIS) techniques. To avoid multicollinearity and ensure the reliability of input variables, statistical validation was performed. The landslide inventory, consisting of 778 past events, was split into 70% training and 30% testing datasets for model development and evaluation. The models were assessed using statistical indicators such as sensitivity, specificity, precision, accuracy, F1-score, Matthews Correlation Coefficient (MCC), and the Area Under the Curve (AUC) of the Receiver Operating Characteristic (ROC). Among the models, XGBoost showed the highest performance (AUC = 0.95), followed by RF (0.83), KNN (0.79), and NB (0.78). The susceptibility analysis revealed that XGBoost categorized 17.53% of the area as highly vulnerable, outperforming RF (14.08%), KNN (14.00%), and NB (4.55%). The enhanced accuracy of XGBoost and RF stems from their ensemble learning approach, which effectively captures nonlinear relationships and mitigates overfitting. The resulting susceptibility maps are crucial tools for risk management, infrastructure planning, and sustainable development in hazard-prone Himalayan regions. This study reinforces the importance of machine learning in natural hazard assessment and suggests that future efforts should incorporate real-time data, finer-resolution remote sensing, and deep learning methods to further refine prediction accuracy.