Prediction Performance Of Model Research Articles

Predicting non-suicidal self-injury among Chinese adolescents: The application of ten algorithms of machine learning

Background and aimsHigh non-suicidal self-injury (NSSI) prevalence among adolescents is a global health issue. However, current prediction models for adolescent NSSI rely on a limited set of algorithms, resulting in biased predictions. Therefore, the aim of this study is to develop multiple machine learning models to enhance prediction accuracy and mitigate biases among Chinese adolescents. MethodsA total of 4487 junior and senior high school students in China were recruited. Multiple algorithms were included, such as logistic regression, decision tree, support vector machine, Naive Bayes, multi-layer perceptron, K-nearest neighbors, and ensemble learning algorithm like random forest, bagging, AdaBoost, and stacking to build predictive models. Data processing techniques, including standardization and the synthetic minority oversampling technique, were employed to optimize the predictive model. The model was trained on 70 % of the data, reserving 30 % for testing. ResultsThe ten prediction models achieved a good performance, with area under the receiver operating characteristic curve (AUC) scores above 0.700 in the test set. The stacking and random forest models achieved AUC scores of 0.904 and 0.898, respectively. The prediction performance of the Naive Bayes model was relatively poor. The top five important variables were resilience, bully, suicidal ideation, internet addiction, and depression. ConclusionsThe ensemble machine learning algorithm showed promising results predicting NSSI among adolescents. Such algorithms should be recommended for future NSSI research to enhance predictive accuracy. Identification of important features in NSSI prediction can help develop screening protocols and lay a foundation for clinical diagnosis and intervention in adolescent populations.

An unsupervised learning model based on CT radiomics features accurately predicts axillary lymph node metastasis in breast cancer patients: diagnostic study.

The accuracy of traditional clinical methods for assessing the metastatic status of axillary lymph nodes (ALNs) is unsatisfactory. In this study, the authors propose the use of radiomic technology and three-dimensional (3D) visualization technology to develop an unsupervised learning model for predicting axillary lymph node metastasis in patients with breast cancer (BC), aiming to provide a new method for clinical axillary lymph node assessment in patients with this disease. In this study, we retrospectively analyzed the data of 350 patients with invasive BC who underwent lung-enhanced computed tomography (CT) and axillary lymph node dissection surgery at the Department of Breast Surgery of the Second Xiangya Hospital of Central South University. The authors used 3D visualization technology to create a 3D atlas of ALNs and identified the region of interest for the lymph nodes. Radiomic features were subsequently extracted and selected, and a prediction model for ALNs was constructed using the K-means unsupervised algorithm. To validate the model, the authors prospectively collected data from 128 BC patients who were clinically evaluated as negative at our center. Using 3D visualization technology, we extracted and selected a total of 36 CT radiomics features. The unsupervised learning model categorized 1737 unlabeled lymph nodes into two groups, and the analysis of the radiomic features between these groups indicated potential differences in lymph node status. Further validation with 1397 labeled lymph nodes demonstrated that the model had good predictive ability for axillary lymph node status, with an area under the curve of 0.847 (0.825-0.869). Additionally, the model's excellent predictive performance was confirmed in the 128 axillary clinical assessment negative cohort (cN0) and the 350 clinical assessment positive (cN+) cohort, for which the correct classification rates were 86.72 and 87.43%, respectively, which were significantly greater than those of clinical assessment methods. The authors created an unsupervised learning model that accurately predicts the status of ALNs. This approach offers a novel solution for the precise assessment of ALNs in patients with BC.

Broadband metamaterial linear polarization converter designed by a hybrid neural network with data augmentation

Deep learning techniques provide a new approach to the design and optimization of electromagnetic metamaterials. This study used a convolutional neural network and long short-term memory (CNN–LSTM) hybrid network to design and optimize a broadband metamaterial reflective linear polarization converter. The data augmentation method was also employed in few-shot learning to reduce optimization costs and improve model prediction performance. With the inverse prediction, a linear polarization converter that perfectly covers the Ku-band was obtained and fabricated with flexible printed circuit technology. Both simulation and experimental results indicate that this network can accurately predict the structural parameters. The polarization converter not only achieves remarkable broadband polarization conversion efficiency spanning the 2.2–18 GHz range but also maintains precise cross-polarization control across the entire Ku-band. The mean polarization conversion ratio in the Ku-band was calculated to be an impressive 99.69%. Finally, the mechanism of polarization conversion and the influence of each structural parameter on its performance further verify the optimality of the inverse design model. The use of CNN–LSTM deep learning methods significantly simplified the design process of electromagnetic metamaterials, reducing design costs while ensuring high design precision and excellent performance.

Development and validation of a prediction model for rebound hyperbilirubinemia: a Chinese neonatal cohort study.

Rebound hyperbilirubinemia (HBB) is still present in as high as 10% of newborn babies. However, the applicability of established prediction models for rebound HBB to Chinese newborns is unclear. This study aimed to establish a model to predict HBB rebound after phototherapy among Chinese neonates. A retrospective cohort study was conducted on 1,035 HBB infants receiving phototherapy. Rebound HBB was defined as total serum bilirubin (TSB) returning to or above the American Academy of Pediatrics (AAP) phototherapy threshold within 72 hours after the end of phototherapy. The predictive effects of previously published two- and three-variable scores were verified. Neonates were randomly assigned in a 6:4 ratio to the training (n=621) group and the testing (n=414) group. All variables in the training set were used to select predictors by least absolute shrinkage and selection operator (LASSO) regression analysis. The internal validation of the prediction model was performed using the testing set. The model's predictive performance was evaluated by area under the curve (AUC), accuracy, sensitivity, and specificity, each with 95% confidence intervals (CIs). Receiver operating characteristic (ROC) and calibration curves were constructed to evaluate the discrimination ability and fitting effect of the prediction model, respectively. Rebound HBB was observed in 210 patients (20.3%). The AUC for the two- and three-variable scores were 0.498 (95% CI: 0.455-0.540) and 0.498 (95% CI: 0.457-0.540), respectively. Predictive factors for the risk of rebound HBB included formula feeding (>3 times/day), standard phototherapy irradiation time, TSB levels and age at termination of phototherapy, neonatal weight, and differences between TSB levels at the phototherapy termination and phototherapy threshold. The prediction model's AUC was 0.935 (95% CI: 0.911-0.958), the sensitivity was 0.880 (95% CI: 0.809-0.950), the specificity was 0.831 (95% CI: 0.790-0.871), and the accuracy was 0.841 (95% CI: 0.805-0.876). The established model performed well in predicting rebound risk among Chinese infants with HBB, which may be beneficial in treating and managing HBB in infants.

Application of a high-performance grey prediction model to predict the cardiovascular disease mortality in elderly Chinese residents

With the acceleration of urbanization and aging process, an increasing number of Chinese senior citizens are suffering from cardiovascular diseases (CVDs). Therefore, reducing the prevalence and mortality of CVD has become one of the priorities of the public health department. And the scientific prediction of CVD morality is of great significance in the prevention and treatment of CVD among the elderly. Given the characteristics of “small sample, complex causes and long duration” of CVD among the senior citizens, grey prediction models which aim at handling the data with “small sample, poor information” are more feasible in forecasting its developing trend. In this paper, a new discrete grey prediction model is designed to forecast the CVD morality among Chinese senior citizens, in which a non-linear function is introduced to expand the model structure and the order and background value coefficient are optimized. In all the three sets of experiments, the errors of the new model are close to but smaller than those of the other three mainstream grey prediction models, indicating that the simulation and prediction performance of the new model is more accurate and stable. The research provides a new modelling approach to predicting the CVD mortality among Chinese senior citizens, which is conducive to the prevention of CVD prevalence and the sustainable development of public health.

Incorporating effects of habitat patches into species distribution models

Abstract Species distribution models (SDMs) are algorithms designed to infer the distribution of species using environmental and biotic variables and have become an important tool for ecologists and conservation biologists seeking to understand the implications of environmental change. Global datasets of environmental variables at resolutions of a few metres are increasingly available. SDMs fitted using such high‐resolution data allow researchers to investigate how local factors affect species occurrences at unprecedented fine spatial scales. As the spatial resolution of SDMs increases, we see a critical need to consider the characteristics of habitat types within or around raster pixels. In particular, we argue that the effects of habitat patches (EHPs, including habitat area, habitat configuration, and habitat diversity), measured focusing on patches or landscapes, have yet to be fully realized in SDMs. We provide guidelines to incorporate EHPs in SDMs. We explain why this development is important, describe approaches to properly conduct such analyses, and discuss pitfalls we foresee in testing EHPs. Synthesis. Ensuring that SDMs incorporating EHPs are properly designed will be key to increasing model predictive performance and to understanding which environmental factors influence the distribution of species at fine spatial scales. At a crucial time for nature conservation, we foresee that this will be a key step forward to understanding and protecting biodiversity.

Construction of a pathomics model for predicting mRNAsi in lung adenocarcinoma and exploration of biological mechanism

ObjectiveThis study aimed to predict the level of stemness index (mRNAsi) and survival prognosis of lung adenocarcinoma (LUAD) using pathomics model. MethodsFrom The Cancer Genome Atlas (TCGA) database, 327 LUAD patients were randomly assigned to a training set (n = 229) and a validation set (n = 98) for pathomics model development and evaluation. PyRadiomics was used to extract pathomics features, followed by feature selection using the mRMR-RFE algorithm. In the training set, Gradient Boosting Machine (GBM) was utilized to establish a model for predicting mRNAsi in LUAD. The model's predictive performance was evaluated using ROC curves, calibration curves, and decision curve analysis (DCA). Prognostic analysis was conducted using Kaplan-Meier curves and cox regression. Additionally, gene enrichment analysis, tumor microenvironment analysis, and tumor mutational burden (TMB) analysis were performed to explore the biological mechanisms underlying the pathomics prediction model. ResultsMultivariable cox analysis (HR = 1.488, 95 % CI 1.012–2.187, P = 0.043) identified mRNAsi as a prognostic risk factor for LUAD. A total of 465 pathomics features were extracted from TCGA-LUAD histopathological images, and ultimately, the most representative 8 features were selected to construct the predictive model. ROC curves demonstrated the significant predictive value of the model for mRNAsi in both the training set (AUC = 0.769) and the validation set (AUC = 0.757). Calibration curves and Hosmer-Lemeshow goodness-of-fit test showed good consistency between the model's prediction of mRNAsi levels and the actual values. DCA indicated a good net benefit of the model. The prediction of mRNAsi levels by the pathomics model is represented using the pathomics score (PS). PS was strongly associated with the prognosis of LUAD (HR = 1.496, 95 % CI 1.008–2.222, P = 0.046). Signaling pathways related to DNA replication and damage repair were significantly enriched in the high PS group. Prediction of immune therapy response indicated significantly reduced Dysfunction in the high PS group (P < 0.001). The high PS group exhibited higher TMB values (P < 0.001). ConclusionsThe predictive model constructed based on pathomics features can forecast the mRNAsi and survival risk of LUAD. This model holds promise to aid clinical practitioners in identifying high-risk patients and devising more optimized treatment plans for patients by jointly employing therapeutic strategies targeting cancer stem cells (CSCs).

Comprehensive assessment of the water environment carrying capacity based on machine learning

Human activities and climate change are constantly threatening water environment systems. To alleviate pressure on the water environment and meet the requirements for sustainable societal development, it has become important to study the water environment carrying capacity (WECC). As science and technology have advanced, the use of artificial intelligence has penetrated various disciplines, and the application of machine learning has achieved many good results in the environmental sciences; however, little attention has been given to the WECC. To this end, a WECC assessment model was developed based on a variety of machine learning algorithms, to provide a novel approach and better understanding of the WECC. In this study, 13 natural and anthropogenic factors related to the water environment system in Liaoning Province (China) were selected, and the study period ranged from 2015 to 2019. Four basic machine learning algorithms, namely, random forest (RF), support vector machine (SVM), multilayer perceptron (MLP) and k-nearest neighbour (KNN), were used for batch modelling, and stacking was carried out based on the output results. A comparison and screening of the models revealed that RF and the stacking model had the best model prediction performances. Because RF was relatively simple and convenient, a comprehensive assessment of the regional WECC based on this algorithm was carried out. The results showed that most of the control units with an overloaded WECC are located in the central and northern regions of Liaoning Province, and the control units of overloading can be well identified and the trend of regional WECC status can be observed. To improve model transparency, we also performed an explanatory analysis based on the shapley additive explanations (SHAP). The results showed that precipitation, soil texture, sewage treatment plant, wind speed, and temperature were the most important factors influencing the water environment system. Precipitation is the most influential factor, and with increasing rainfall, the WECC first improves and then gradually deteriorates. For the anthropogenic factors, the WECC generally improves first and then worsens as these factors increase. These research results can be used to provide a powerful and practical new methodology for the comprehensive assessment of the WECC and provide a basis for the development of water environment policies and regulatory implementation in the region.

Prediction and optimization of wastewater treatment process effluent chemical oxygen demand and energy consumption based on typical ensemble learning models

Pollution integration and carbon reduction has become a primary focus in wastewater treatment processes. In this study, water quality and control indicators were used as input features and the dataset was extended using the moving average method. Random Forest, eXtreme Gradient Boosting (XGBoost), and Light Gradient Boosting Machine algorithms were used to predict the effluent chemical oxygen demand (COD) and total energy consumption (TEC). The results indicated that the model prediction performance could be effectively improved when the data were amplified by two times and that the XGBoost model exhibited the best prediction performance for effluent COD and TEC. The Non-dominated Sorting Genetic Algorithm II model was employed for the multi-objective optimization of effluent COD and TEC, resulting in reductions of 15% and 18%, respectively. The ensemble learning model proposed in this study to achieve synergy between water quality improvement and energy saving is practical.

Comparative study of kernel- and deep learning-based proxy models for nonlinearly constrained life-cycle production optimization

This study presents a comparative study for the application of deep learning–based and kernel-based proxy models in hydrocarbon production optimization with nonlinear constraints, comparing their performance against high-fidelity simulators (HFS) in terms of computational efficiency and the quality of the results achieved. One of the proxy models employed is referred to as embed to control and observe (E2CO), a deep learning–based model. The other model is a kernel-based proxy model known as least-squares support-vector regression (LS-SVR). Both proxy models can predict production/injection outputs for each well. We use the sequential quadratic programming (SQP) method for nonlinearly constrained production optimization. Our optimization objective revolves around the net present value (NPV), while the nonlinear state constraints involve field liquid production rate (FLPR) and field water production rate (FWPR). NPV, FLPR, and FWPR are estimated using two different types of proxy models. The gradient of the objective function and the Jacobian matrix of constraints are computed analytically for LS-SVR. In contrast, the method of stochastic simplex approximated gradient (StoSAG) is employed for optimization with E2CO and HFS. The reservoir model considered in this study is a two-phase, three-dimensional reservoir characterized by channelized and heterogeneous permeability, sourced from the SPE10 benchmark case. We optimize well controls to maximize NPV in an oil-water waterflooding scenario. All proxy models can achieve optimal NPV results like those obtained through HFS but with significantly lower computational resources. Both proxy models are found to be approximately 15-fold more efficient than using HFS when the number of simulation runs is compared. Forward prediction results show that both proxy models outperform numerical simulator, and the E2CO can predict up to 2400 times faster than HFS whereas speedups gained over HFS go up to 5000-fold with LS-SVR. The training time required for E2CO is at least an order of magnitude longer than that for LS-SVR. The main contributions of the paper are (i) development of a novel methodology that uses an E2CO proxy model within a gradient-based iterative retraining proxy approach for life-cycle production optimization with nonlinear state constraints, (ii) comparison of this methodology in terms of computational cost and accuracy with an LS-SVR proxy model, and (iii) new insights into the accuracy and prediction performances of these machine learning-based proxy models for 3D oil-water systems as well as their efficiency in nonlinearly constrained production optimization for waterflooding applications.

Machine-Learning-Based Path Loss Prediction for Vehicle-to-Vehicle Communication in Highway Environments

Vehicle-to-vehicle (V2V) communication, which plays an important role in intelligent transportation systems, has been statistically proven to improve traffic efficiency and reduce the probability of accidents. In real-world applications, it is critical to accurately estimate the path loss parameter in communication channels due to the variable and complex propagation environments often encountered in inter-vehicle communication scenarios. This paper presents a study on various machine learning methods to improve path loss estimation in V2V communication using a dataset (192,000 observations) obtained from field measurements of highway environments in the Trabzon and Gümüşhane provinces in Türkiye. For this purpose, path loss estimation was carried out with different machine learning algorithms such as Artificial Neural Networks, Random Forest, Linear Regression, Gradient Boosting, Support Vector Regression, and AdaBoost by using various environmental and system features. Then, performance comparisons were conducted between machine learning methods and traditional empirical approaches such as log-distance, two-ray, and log-ray. Examining the outputs reveals that machine learning methods outperform traditional methods and yield results quickly. As a result, the Random Forest and Gradient Boosting methods demonstrated the highest prediction performances, with R2 values of 0.97 and 0.96, MAE values of 0.0557 and 0.0701, and RMSE values of 0.0774 and 0.0964, respectively, outperforming both empirical methods, other machine learning techniques, and the existing studies based on V2V. Overall, our study provides significant contributions to the existing literature by providing a comprehensive parameter set for highway environments, examining the path loss prediction performance of machine learning models with different capabilities, and comparing them with traditional methods. This study not only fills a critical gap in the existing literature but also highlights the necessity, efficiency, and originality of machine learning approaches for improving reliable V2V communication systems.

Reduced-order reconstruction of discrete grey forecasting model and its application

Discrete grey forecasting models based on an accumulative operator have been widely used in many practical fields. With the development of grey forecasting models, it is a problem to be solved to further analyze internal mechanisms and unify the structures. This paper aims to reconstruct the model from a perspective of sequence characteristics and simplify the modeling steps under the condition of ensuring the accuracy of the model. First, this paper analyzes dynamic sequence evolution hidden and mines relationship between the structure and original sequence features contained in discrete grey forecasting model. Then, the reconstruction is carried out to prove the equivalence and quantitative relation between reduced-order model and original model. Under order recursive estimation, new parameters are addressed. Finally, theoretical correctness is verified by large-scale numerical simulation. Moreover, the reduced-order model is applied for prediction on the peak of battery incremental capacity and capacity degradation. Results show the effectiveness and superior prediction performance of the reduced-order model, where MAPEs of grey forecasting models have controlled under 4%.

Population pharmacokinetics study of tacrolimus in liver transplant recipients: a comparison between patients with or without liver cancer before surgery.

The main challenge for immunosuppressive therapy using tacrolimus in liver transplantation is the considerable variability in its oral bioavailability and the narrow treatment range. Many population pharmacokinetic (PopPK) models have been established to precisely estimate the PK variability of tacrolimus in liver transplant recipients. However, it remains unclear whether there is a significant difference in the PK behavior of tacrolimus between patients with or without liver cancer before surgery. Therefore, we aimed to compare the differences of PK parameters and simulate exposures of tacrolimus between populations preoperatively diagnosed with liver cancer or not by PopPK modeling. In total, 802 blood concentrations of tacrolimus from 196 patients (118 liver cancer and 78 non-liver-cancer samples) were included in this study. Demographic data and clinical parameters were integrated to perform a PopPK analysis using the nonlinear mixed-effects modeling approach. Potential covariates were evaluated by using a stepwise method. Goodness-of-fit plot and bootstrap were performed to assess the model stability and predictive performance. Simulations were introduced to optimize dosing regimens of both the liver cancer and non-liver-cancer groups according to the guidance. The PK of tacrolimus was best described by a one-compartment model with first-order absorption and linear elimination, with weight and direct bilirubin as the significant covariates. In the process of constructing the basic model, we tried to separately estimate the PK parameters in liver cancer and non-liver-cancer populations. The results showed that the PK parameters in the two populations were similar, and the individual variation in Ka in non-liver-cancer subjects was large. Hence, the final model did not distinguish between the two populations. Moreover, a minor increase of less than 10% was observed in the simulated exposure in the patients preoperatively diagnosed with liver cancer compared with that in non-liver-cancer groups. The established PopPK model was capable of optimizing tacrolimus dosing in whole populations who underwent liver transplantation. Although a minimal difference was found in tacrolimus exposure between the liver cancer and non-liver-cancer groups, more research is warranted to explore the differences between the two populations in the future, given the potential limitations of this study.

Natural gas consumption forecasting using a novel two-stage model based on improved sparrow search algorithm

The foundation of natural gas intelligent scheduling is the accurate prediction of natural gas consumption (NGC). However, its volatility, brings difficulties and challenges in accurately predicting NGC. To address this problem, an improved model is developed combining improved sparrow search algorithm (ISSA), long short-term memory (LSTM), and wavelet transform (WT). First, the performance of ISSA is tested. Second, the NGC is divided into several high- and low-frequency components applying different layers of Coilfets’, Fejer-Korovkins’, Symletss’, Haars’, and Discretes’ orders. In addition, the LSTM is applied to forecast the decomposed components in view of the one- and multi-step, and its hyper-parameters are optimized by ISSA. At last, the final prediction results are reconstructed. The research results indicate that: 1) Comparing to other machine algorithms (e.g., fuzzy neural network), the convergence speed and stability of ISSA are stronger in view of standard deviation and mean; 2) The prediction performance of the developed model is better than that of other forecasting models; 3) The forecasting performance of the single-step forecasting is superior to that of the two-, three-, and four- step; 4) The computational load of the proposed prediction model is the highest compared to other models, and the prediction accuracy is still excellent on the extended time series.

Machine learning model to predict rate constants for sonochemical degradation of organic pollutants

In this study, machine learning (ML) algorithms were employed to predict the pseudo-1st-order reaction rate constants for the sonochemical degradation of aqueous organic pollutants under various conditions. A total of 618 sets of data, including ultrasonic, solution, and pollutant characteristics, were collected from 89 previous studies. Considering the difference between the electrical power (Pele) and calorimetric power (Pcal), the collected data were divided into two groups: data with Pele and data with Pcal. Eight input variables, including frequency, power density, pH, temperature, initial concentration, solubility, vapor pressure, and octanol–water partition coefficient (Kow), and one target variable of the degradation rate constant, were selected for ML. Statistical analysis was conducted, and outliers were determined separately for the two groups. ML models, including random forest (RF), extreme gradient boosting (XGB), and light gradient boosting machine (LGB), were used to predict the pseudo-1st-order reaction rate constants for the removal of aqueous pollutants. The prediction performance of the ML models was evaluated using different metrics, including the root mean squared error (RMSE), mean absolute error (MAE), and R squared (R2). A significantly higher prediction performance was obtained using data without outliers and augmented data. Consequently, all the applied ML models could be used to predict the sonochemical degradation of aqueous pollutants, and the XGB model showed the highest accuracy in predicting the rate constants. In addition, the power density and frequency were the most influential factors among the eight input variables in prediction with the Shapley additive explanation (SHAP) values method. The degradation rate constants of the two pollutants over a wide frequency range (20–1,000 kHz) were predicted using the trained ML model (XGB) and the prediction results were analyzed.

Fast prediction of personalized abdominal organ doses from CT examinations by radiomics feature-based machine learning models

The X-rays emitted during CT scans can increase solid cancer risks by damaging DNA, with the risk tied to patient-specific organ doses. This study aims to establish a new method to predict patient specific abdominal organ doses from CT examinations using minimized computational resources at a fast speed. The CT data of 247 abdominal patients were selected and exported to the auto-segmentation software named DeepViewer to generate abdominal regions of interest (ROIs). Radiomics feature were extracted based on the selected CT data and ROIs. Reference organ doses were obtained by GPU-based Monte Carlo simulations. The support vector regression (SVR) model was trained based on the radiomics features and reference organ doses to predict abdominal organ doses from CT examinations. The prediction performance of the SVR model was tested and verified by changing the abdominal patients of the train and test sets randomly. For the abdominal organs, the maximal difference between the reference and the predicted dose was less than 1 mGy. For the body and bowel, the organ doses were predicted with a percentage error of less than 5.2%, and the coefficient of determination (R2) reached up to 0.9. For the left kidney, right kidney, liver, and spinal cord, the mean absolute percentage error ranged from 5.1 to 8.9%, and the R2 values were more than 0.74. The SVR model could be trained to achieve accurate prediction of personalized abdominal organ doses in less than one second using a single CPU core.

A general model for prediction of the CO2 equilibrium solubility in aqueous tertiary amine systems

AbstractWe have developed a general model to predict CO2 equilibrium solubility in aqueous tertiary amine solutions by relating the equilibrium constant (K1) to all relevant parameters in a logical function form. Testing our model, we measured CO2 equilibrium solubility data for N‐methylmorpholine (NMM) and N‐ethylmorpholine (NEM) across various conditions. Comparison with six existing models reveals our general model's superior predictive performance not only for NMM and NEM but also for an additional 10 tertiary amine solutions from literature, indicating its universality. Comprehensively considering the CO2 equilibrium solubility, amine dissociation constant (pKa) and the CO2 absorption heat, it is found that NMM and NEM may be promising desorption promoters enabling to reduce the energy cost. In short, it is expected the general model can be applied to more other tertiary amine systems.

Peritumoral edema enhances MRI-based deep learning radiomic model for axillary lymph node metastasis burden prediction in breast cancer

To investigate whether peritumoral edema (PE) could enhance deep learning radiomic (DLR) model in predicting axillary lymph node metastasis (ALNM) burden in breast cancer. Invasive breast cancer patients with preoperative MRI were retrospectively enrolled and categorized into low (< 2 lymph nodes involved (LNs+)) and high (≥ 2 LNs+) burden groups based on surgical pathology. PE was evaluated on T2WI, and intra- and peri-tumoral radiomic features were extracted from MRI-visible tumors in DCE-MRI. Deep learning models were developed for LN burden prediction in the training cohort and validated in an independent cohort. The incremental value of PE was evaluated through receiver operating characteristic (ROC) analysis, confirming the improvement in the area under the curve (AUC) using the Delong test. This was complemented by net reclassification improvement (NRI) and integrated discrimination improvement (IDI) metrics. The deep learning combined model, incorporating PE with selected radiomic features, demonstrated significantly higher AUC values compared to the MRI model and the DLR model in the training cohort (n = 177) (AUC: 0.953 vs. 0.849 and 0.867, p < 0.05) and the validation cohort (n = 111) (AUC: 0.963 vs. 0.883 and 0.882, p < 0.05). The complementary analysis demonstrated that PE significantly enhances the prediction performance of the DLR model (Categorical NRI: 0.551, p < 0.001; IDI = 0.343, p < 0.001). These findings were confirmed in the validation cohort (Categorical NRI: 0.539, p < 0.001; IDI = 0.387, p < 0.001). PE improved preoperative ALNM burden prediction of DLR model, facilitating personalized axillary management in breast cancer patients.

A dual-radiomics model for overall survival prediction in early-stage NSCLC patient using pre-treatment CT images.

Radiation therapy (RT) is one of the primary treatment options for early-stage non-small cell lung cancer (ES-NSCLC). Therefore, accurately predicting the overall survival (OS) rate following radiotherapy is crucial for implementing personalized treatment strategies. This work aims to develop a dual-radiomics (DR) model to (1) predict 3-year OS in ES-NSCLC patients receiving RT using pre-treatment CT images, and (2) provide explanations between feature importanceand model prediction performance. The publicly available TCIA Lung1 dataset with 132 ES-NSCLC patients received RT were studied: 89/43 patients in the under/over 3-year OS group. For each patient, two types of radiomic features were examined: 56 handcrafted radiomic features (HRFs) extracted within gross tumor volume, and 512 image deep features (IDFs) extracted using a pre-trained U-Net encoder. They were combined as inputs to an explainable boosting machine (EBM) model for OS prediction. The EBM's mean absolute scores for HRFs and IDFs were used as feature importance explanations. To evaluate identified feature importance, the DR model was compared with EBM using either (1) key or (2) non-key feature type only. Comparison studies with other models, including supporting vector machine (SVM) and random forest (RF), were also included. The performance was evaluated by the area under the receiver operating characteristic curve (AUCROC), accuracy, sensitivity, and specificity with a 100-fold Monte Carlo cross-validation. The DR model showed highestperformance in predicting 3-year OS (AUCROC=0.81 ± 0.04), and EBM scores suggested that IDFs showed significantly greater importance (normalized mean score=0.0019) than HRFs (score=0.0008). The comparison studies showed that EBM with key feature type (IDFs-only demonstrated comparable AUCROC results (0.81 ± 0.04), while EBM with non-key feature type (HRFs-only) showed limited AUCROC (0.64 ± 0.10). The results suggested that feature importance score identified by EBM is highly correlated with OS prediction performance. Both SVM and RF models were unable to explain key feature type while showing limited overall AUCROC=0.66 ± 0.07 and 0.77 ± 0.06, respectively. Accuracy, sensitivity, and specificity showed a similar trend. In conclusion, a DR model was successfully developed to predict ES-NSCLC OS based on pre-treatment CT images. The results suggested that the feature importance from DR model is highly correlated to the model prediction power.

Prediction model of moisture content in spray drying of ceramic slurry based on IMFO-BPNN

The final particle moisture content of ceramic slurry spray drying affects the processing, product quality, and process energy consumption. While the real-time moisture content determination is very difficult for spray drying process due to nonlinear, severe lag, interference factors, and complexity of the systems, the high temperature, high humidity drying environment and short drying process. In current work, an improved moth optimization (IMFO) algorithm combined with backpropagation neural network (BPNN) was proposed to predict the moisture content of ceramic slurry particles during spray drying. The performance of the model is trained and tested. Results demonstrate that IMFO has better convergence ability and speed compared to other algorithms such as particle swarm optimization (PSO) and gravitational search algorithm (GSA). The IMFO-BPNN model achieves a MAE of 0.0293, RMSE of 0.0383, and R2 value of 0.9113, outperforming the prediction performance of BPNN and MFO-BPNN models. The absolute error rate of the IMFO-BPNN model (0–5%) is lower compared to BPNN (5–10%) and MFO-BPNN model (>10%), showcasing superior accuracy in predicting moisture content. This mathematical model established in the study provides an efficient, accurate, and nondestructive method for predicting the drying endpoint of ceramic slurry spray drying.