Optimal Prediction Model Research Articles

Interpretation of fatigue lifetime prediction by machine learning modeling in piston aluminum alloys under different manufacturing and loading conditions

Various input variables, including corrosion time, fretting force, stress, lubrication, heat-treating, and nano-particles, were evaluated by modeling of stress-controlled fatigue lifetimes in AlSi12CuNiMg aluminum alloy of the engine pistons with different machine learning (ML) techniques. Bending fatigue experiments were conducted through cyclic loading with zero mean stress, and then experimental data was predicted by five different ML-based models. Moreover, when the optimal ML prediction model was found, it was analyzed using the Shapley additive explanation (SHAP) values method. Results illustrated that extreme gradient boosting (XGBoost) had superior data for estimations, with average training coefficients of determination of at least 63% and 90%, respectively for fatigue lifetime and its logarithmic value. The SHAP values interpretation of the XGBoost model revealed that fretting force, stress, and corrosion time were the most significant inputs in estimating the logarithm values of fatigue lifetimes, respectively.

IMPROVING THE EVALUATION AND PREDICTION OF PREVENTION AND TREATMENT EFFICIENCY DURING PUBLIC HEALTH EMERGENCIES BY USING THE SBM-BPNN ALGORITHM

Crisis response is the key to reducing the impact and damage of public health emergencies. Previous studies only focused on specific measures for emergency response, but the effectiveness of these measures has seldom been discussed. How about the effective of the crisis response on public health emergencies? What will happen to the effectiveness of crisis response after the transformation of public health emergency measures? To address these questions, a novel research framework was constructed by combining slacks-based measure (SBM) and back propagation neural network (BPNN) algorithms. At Stage I, an efficiency evaluation was conducted on SBM with undesirable outputs to evaluate the prevention and treatment efficiency during public health emergencies, and stochastic frontier analysis was performed to mitigate the influence of environmental factors and statistical noise. At Stage II, the adjusted efficiency was predicted with the BPNN model. The decision-making units were effectively improved by incorporating slack variables, thus enabling the prediction and optimization of optimal cases given the epidemic resources. An empirical analysis of the response of 43 G20 member countries to the COVID-19 pandemic showed that the novel framework could evaluate prevention and treatment efficiency across regions and predict the efficiency and optimal case outputs following shifts in epidemic preventive measures. After evaluation, the mean squared error of the BPNN efficiency prediction model was only 0.0014, whereas that of the BPNN optimal output prediction model was 0.126. Therefore, this novel framework is suitable for evaluating and predicting the effectiveness of crisis response on public health emergencies, which provides necessary assistance for crisis decision-making by the government and emergency management organizations., Keywords: Public health emergencies, COVID-19 pandemic, Crisis response efficiency, Evaluation, Prediction.

Research on recompression supercritical CO2 power cycle system considering performance and stability of main compressor

The supercritical carbon dioxide (sCO2) power cycle has good prospects for use in many energy applications. Ensuring the stable operation of the power cycle is an important issue. The inlet condition of the main compressor in the sCO2 recompression cycle is close to the critical point of CO2 and the physical properties of CO2 change drastically near the critical point, so the study of inlet parameter change behavior of the main compressor is of great significance to the stable operation of the equipment and the system. This paper investigates the performance of an sCO2 recompression power cycle system under off-design conditions with different cooling and heating conditions. Unlike previous studies, this paper uses more accurate compressor and heat exchanger models to simulate the system for off-designed performance. A 1-D optimization prediction model of the compressor and a density-based segmented heat exchanger model are introduced to predict the performance changes of the system. The operating conditions, performance, and stability of the compressor under different cooling and heating conditions were compared. The results show that variations in different cooling and heating conditions change the inlet temperature of the compressor, causing the sCO2 compressor to deviate from optimal operation. As the inlet temperature increases, the power consumption of the compressor increases. At an inlet temperature of 305 K, the power is 1.93 MW (main compressor, MC) and 2.17 MW (recompressor, RC). When the inlet temperature increases to 310 K, the power increases by more than 10 %. The largest change in system efficiency is 3.76 % for a change in cooling conditions, and the largest change in compressor speed is 14.23 % (MC) and 7.79 % (RC) for a change in heating conditions, respectively. The operating condition of the main compressor moves far away from the surge line when its inlet temperature increases, which means the operation of the main compressor is more stable in this situation. The results of this paper contribute to the stable operation of the sCO2 recompression cycle.

Prediction of intraoperative red blood cell transfusion in valve replacement surgery: machine learning algorithm development based on non-anemic cohort.

Our study aimed to develop machine learning algorithms capable of predicting red blood cell (RBC) transfusion during valve replacement surgery based on a preoperative dataset of the non-anemic cohort. A total of 423 patients who underwent valvular replacement surgery from January 2015 to December 2020 were enrolled. A comprehensive database that incorporated demographic characteristics, clinical conditions, and results of preoperative biochemistry tests was used for establishing the models. A range of machine learning algorithms were employed, including decision tree, random forest, extreme gradient boosting (XGBoost), categorical boosting (CatBoost), support vector classifier and logistic regression (LR). Subsequently, the area under the receiver operating characteristic curve (AUC), accuracy, recall, precision, and F1 score were used to determine the predictive capability of the algorithms. Furthermore, we utilized SHapley Additive exPlanation (SHAP) values to explain the optimal prediction model. The enrolled patients were randomly divided into training set and testing set according to the 8:2 ratio. There were 16 important features identified by Sequential Backward Selection for model establishment. The top 5 most influential features in the RF importance matrix plot were hematocrit, hemoglobin, ALT, fibrinogen, and ferritin. The optimal prediction model was CatBoost algorithm, exhibiting the highest AUC (0.752, 95% CI: 0.662-0.780), which also got relatively high F1 score (0.695). The CatBoost algorithm also showed superior performance over the LR model with the AUC (0.666, 95% CI: 0.534-0.697). The SHAP summary plot and the SHAP dependence plot were used to visually illustrate the positive or negative effects of the selected features attributed to the CatBoost model. This study established a series of prediction models to enhance risk assessment of intraoperative RBC transfusion during valve replacement in no-anemic patients. The identified important predictors may provide effective preoperative interventions.

수학학습 정의적 영역을 이용한 대학 이공계 전공 예측 기계학습 분석

This study developed and analyzed models to predict the selection of STEM majors in college based on the affective domain characteristics of students during their middle and high school years, using data from the Korean Education Longitudinal Study 2005. Instrumental motivation, intrinsic motivation towards mathematics, self-efficacy in mathematics, and beliefs about intelligence were used as independent variables for prediction based on the data from the second grade of middle school and the first grade of high school. Logistic models, Multi Layer Perceptron (MLP) models, and Random Forest models were applied, and the optimal prediction model was sought by comprehensively considering various indicators. In the case of using the data from the second grade of middle school, the MLP model showed superior predictive power compared to other models. In the case of using the data from the first grade of high school, all models did not show good predictive power. The model based on the data from the second grade of middle school was deemed to have practical applicability. These results suggest that while the affective domain regarding mathematics is an important factor in the choice of a STEM major, more diverse variables and complex models are needed for predicting majors during high school years. Additionally, while recognizing the limitations of machine learning models, more theoretical discussions are needed on the relationship between major selection and the affective elements of mathematics, and it is hoped that various prediction models will be developed and used in actual practice.

Sleep-phasic heart rate variability predicts stress severity: Building a machine learning-based stress prediction model.

We propose a novel approach for predicting stress severity by measuring sleep phasic heart rate variability (HRV) using a smart device. This device can potentially be applied for stress self-screening in large populations. Using a Holter electrocardiogram (ECG) and a Huawei smart device, we conducted 24-h dual recordings of 159 medical workers working regular shifts. Based on photoplethysmography (PPG) and accelerometer signals acquired by the Huawei smart device, we sorted episodes of cyclic alternating pattern (CAP; unstable sleep), non-cyclic alternating pattern (NCAP; stable sleep), wakefulness, and rapid eye movement (REM) sleep based on cardiopulmonary coupling (CPC) algorithms. We further calculated the HRV indices during NCAP, CAP and REM sleep episodes using both the Holter ECG and smart-device PPG signals. We later developed a machine learning model to predict stress severity based only on the smart device data obtained from the participants along with a clinical evaluation of emotion and stress conditions. Sleep phasic HRV indices predict individual stress severity with better performance in CAP or REM sleep than in NCAP. Using the smart device data only, the optimal machine learning-based stress prediction model exhibited accuracy of 80.3 %, sensitivity 87.2 %, and 63.9 % for specificity. Sleep phasic heart rate variability can be accurately evaluated using a smart device and subsequently can be used for stress predication.

Machine learning reveals diverse cell death patterns in lung adenocarcinoma prognosis and therapy

Cancer cell growth, metastasis, and drug resistance pose significant challenges in the management of lung adenocarcinoma (LUAD). However, there is a deficiency in optimal predictive models capable of accurately forecasting patient prognoses and guiding the selection of targeted treatments. Programmed cell death (PCD) pathways play a pivotal role in the development and progression of various cancers, offering potential as prognostic indicators and drug sensitivity markers for LUAD patients. The development and validation of predictive models were conducted by integrating 13 PCD patterns with comprehensive analysis of bulk RNA, single-cell RNA transcriptomics, and pertinent clinicopathological details derived from TCGA-LUAD and six GEO datasets. Utilizing the machine learning algorithms, we identified ten critical differentially expressed genes associated with PCD in LUAD, namely CHEK2, KRT18, RRM2, GAPDH, MMP1, CHRNA5, TMPRSS4, ITGB4, CD79A, and CTLA4. Subsequently, we conducted a programmed cell death index (PCDI) based on these genes across the aforementioned cohorts and integrated this index with relevant clinical features to develop several prognostic nomograms. Furthermore, we observed a significant correlation between the PCDI and immune features in LUAD, including immune cell infiltration and the expression of immune checkpoint molecules. Additionally, we found that patients with a high PCDI score may exhibit resistance to immunotherapy and standard adjuvant chemotherapy regimens; however, they may benefit from other FDA-supported drugs such as docetaxel and dasatinib. In conclusion, the PCDI holds potential as a prognostic signature and can facilitate personalized treatment for LUAD patients.

Lung cancer detection in perioperative patients' exhaled breath with nanomechanical sensor array

IntroductionBreath analysis using a chemical sensor array combined with machine learning algorithms may be applicable for detecting and screening lung cancer. In this study, we examined whether perioperative breath analysis can predict the presence of lung cancer using a Membrane-type Surface stress Sensor (MSS) array and machine learning. MethodsPatients who underwent lung cancer surgery at an academic medical center, Japan, between November 2018 and November 2019 were included. Exhaled breaths were collected just before surgery and about one month after surgery, and analyzed using an MSS array. The array had 12 channels with various receptor materials and provided 12 waveforms from a single exhaled breath sample. Boxplots of the perioperative changes in the expiratory waveforms of each channel were generated and Mann-Whitney U test were performed. An optimal lung cancer prediction model was created and validated using machine learning. ResultsSixty-six patients were enrolled of whom 57 were included in the analysis. Through the comprehensive analysis of the entire dataset, a prototype model for predicting lung cancer was created from the combination of array five channels. The optimal accuracy, sensitivity, specificity, positive predictive value, and negative predictive value were 0.809, 0.830, 0.807, 0.806, and 0.812, respectively. ConclusionBreath analysis with MSS and machine learning with careful control of both samples and measurement conditions provided a lung cancer prediction model, demonstrating its capacity for non-invasive screening of lung cancer.

Predictability comparison of sizing parameters for postoperative vault after implantable Collamer lens implantation.

This study aims to assess the accuracy of three parameters (white-to-white distance [WTW], angle-to-angle [ATA], and sulcus-to-sulcus [STS]) in predicting postoperative vault and to formulate an optimized predictive model. In this retrospective study, a cohort of 465 patients (comprising 769 eyes) who underwent the implantation of the V4c implantable Collamer lens with a central port (ICL) for myopia correction was examined. Least absolute shrinkage and selection operator (LASSO) regression and classification models were used to predict postoperative vault. The influences of WTW, ATA, and STS on predicting the postoperative vault and ICL size were analyzed and compared. The dataset was randomly divided into training (80%) and test (20%) sets, with no significant differences observed between them. The screened variables included only seven variables which conferred the largest signal in the model, namely, lens thickness (LT, estimated coefficients for logistic least absolute shrinkage of -0.20), STS (-0.04), size (0.08), flat K (-0.006), anterior chamber depth (0.15), spherical error (-0.006), and cylindrical error (-0.0008). The optimal prediction model depended on STS (R2=0.419, RMSE=0.139), whereas the least effective prediction model relied on WTW (R2=0.395, RMSE=0.142). In the classified prediction models of the vault, classification prediction of the vault based on STS exhibited superior accuracy compared to ATA or WTW. This study compared the capabilities of WTW, ATA, and STS in predicting postoperative vault, demonstrating that STS exhibits a stronger correlation than the other two parameters.

UAV hyperspectral analysis of secondary salinization in arid oasis cotton fields: effects of FOD feature selection and SOA-RF.

Secondary salinization is a crucial constraint on agricultural progress in arid regions. The specific mulching irrigation technique not only exacerbates secondary salinization but also complicates field-scale soil salinity monitoring. UAV hyperspectral remote sensing offers a monitoring method that is high-precision, high-efficiency, and short-cycle. In this study, UAV hyperspectral images were used to derive one-dimensional, textural, and three-dimensional feature variables using Competitive adaptive reweighted sampling (CARS), Gray-Level Co-occurrence Matrix (GLCM), Boruta Feature Selection (Boruta), and Brightness-Color-Index (BCI) with Fractional-order differentiation (FOD) processing. Additionally, three modeling strategies were developed (Strategy 1 involves constructing the model solely with the 20 single-band variable inputs screened by the CARS algorithm. In Strategy 2, 25 texture features augment Strategy 1, resulting in 45 feature variables for model construction. Strategy 3, building upon Strategy 2, incorporates six triple-band indices, totaling 51 variables used in the model's construction) and integrated with the Seagull Optimization Algorithm for Random Forest (SOA-RF) models to predict soil electrical conductivity (EC) and delineate spatial distribution. The results demonstrated that fractional order differentiation highlights spectral features in noisy spectra, and different orders of differentiation reveal different hidden information. The correlation between soil EC and spectra varies with the order. 1.9th order differentiation is proved to be the best order for constructing one-dimensional indices; although the addition of texture features slightly improves the accuracy of the model, the integration of the three-waveband indices significantly improves the accuracy of the estimation, with an R2 of 0.9476. In contrast to the conventional RF model, the SOA-RF algorithm optimizes its parameters thereby significantly improving the accuracy and model stability. The optimal soil salinity prediction model proposed in this study can accurately, non-invasively and rapidly identify excessive salt accumulation in drip irrigation under membrane. It is of great significance to improve the growing conditions of cotton, increase the cotton yield, and promote the sustainable development of Xinjiang's agricultural economy, and also provides a reference for the prevention and control of regional soil salinization.

Modeling of flaxseed protein, oil content, linoleic acid, and lignan content prediction based on hyperspectral imaging.

Protein, oil content, linoleic acid, and lignan are several key indicators for evaluating the quality of flaxseed. In order to optimize the testing methods for flaxseed's nutritional quality and enhance the efficiency of screening high-quality flax germplasm resources, we selected 30 flaxseed species widely cultivated in Northwest China as the subjects of our study. Firstly, we gathered hyperspectral information regarding the seeds, along with data on protein, oil content, linoleic acid, and lignan, and utilized the SPXY algorithm to classify the sample set. Subsequently, the spectral data underwent seven distinct preprocessing methods, revealing that the PLSR model exhibited superior performance after being processed with the SG smoothing method. Feature wavelength extraction was carried out using the Successive Projections Algorithm (SPA) and the Competitive Adaptive Reweighted Sampling (CARS). Finally, four quantitative analysis models, namely Partial Least Squares Regression (PLSR), Support Vector Regression (SVR), Multiple Linear Regression (MLR), and Principal Component Regression (PCR), were individually established. Experimental results demonstrated that among all the models for predicting protein content, the SG-CARS-MLR model predicted the best, with and of 0.9563 and 0.9336, with the corresponding Root Mean Square Error Correction (RMSEC) and Root Mean Square Error Prediction (RMSEP) of 0.4892 and 0.5616, respectively. In the optimal prediction models for oil content, linoleic acid and lignan, the was 0.8565, 0.8028, 0.9343, and the RMSEP was 0.8682, 0.5404, 0.5384, respectively. The study results show that hyperspectral imaging technology has excellent potential for application in the detection of quality characteristics of flaxseed and provides a new option for the future non-destructive testing of the nutritional quality of flaxseed.

Individualized driver amplitude in liver MR elastography: a linear regression study.

Current liver magnetic resonance elastography (MRE) scans often require adjustments to driver amplitude to produce acceptable images. This could lead to time wastage and the potential loss of an opportunity to capture a high-quality image. To construct a linear regression model of individualized driver amplitude to improve liver MRE image quality. Data from 95 liver MRE scans of 61 participants, including abdominal missing volume ratio (AMVR), breath-holding status, the distance from the passive driver on the skin surface to the liver edge (Dd-l), body mass index (BMI), and lateral deflection of the passive driver with respect to the human sagittal plane (Angle α), were continuously collected. The Spearman correlation analysis and lasso regression were conducted to screen the independent variables. Multiple linear regression equations were developed to determine the optimal amplitude prediction model. The optimal formula for linear regression models: driver amplitude (%) = -16.80 + 78.59 × AMVR - 11.12 × breath-holding (end of expiration = 1, end of inspiration = 0) + 3.16 × Dd-l + 1.94 × BMI + 0.34 × angle α, with the model passing the F test (F = 22.455, P <0.001) and R2 value of 0.558. The individualized amplitude prediction model based on AMVR, breath-holding status, Dd-l, BMI, and angle α is a valuable tool in liver MRE examination.

Validated Clinical Score to Predict Gastroesophageal Reflux in Patients With Chronic Laryngeal Symptoms: COuGH RefluX

Background & AimsDiscerning whether laryngeal symptoms result from gastroesophageal reflux is clinically challenging and a reliable tool to stratify patients is needed. We aimed to develop and validate a model to predict likelihood of gastroesophageal reflux disease (GERD) among patients with chronic laryngeal symptoms. MethodsThis multicenter international study collected data from adults with chronic laryngeal symptoms that underwent objective testing (upper GI endoscopy and/or ambulatory reflux monitoring) between 3/2018-5/2023. The training phase identified a model with optimal receiver operating characteristic curves, and beta coefficients informed a weighted model. The validation phase assessed performance characteristics of the weighted model. Results856 adults, 304 in training cohort and 552 in the validation cohort, were included. In the training phase the optimal predictive model [AUC 0.68 (95% CI 0.62, 0.74)] - the COuGH RefluX score - consisted of Cough, Overweight/obesity, Globus, Hiatal Hernia, Regurgitation, and male seX with a lower threshold of 2.5 and upper threshold of 5.0 to predict proven GERD. In the validation phase, the COuGH RefluX score had an AUC of 0.67 (0.62, 0.71) with 79% sensitivity and 81% specificity for proven GERD. ConclusionsThe externally validated COuGH RefluX score is a clinically practical model to predict likelihood of proven GERD. The score classifies most patients with chronic laryngeal symptoms as low/high likelihood of proven GERD, with only 38% remaining as indeterminate. Thus, the COuGH RefluX score can guide diagnostic strategies and reduce inappropriate proton pump inhibitor use or testing for patients referred for evaluation of chronic laryngeal symptoms.

Major Depressive Disorder Prediction Based on Sleep-Wake Disorders Symptoms in US Adolescents: A Machine Learning Approach from National Sleep Research Resource.

There is substantial evidence from previous studies that abnormalities in sleep parameters associated with depression are demonstrated in almost all stages of sleep architecture. Patients with symptoms of sleep-wake disorders have a much higher risk of developing major depressive disorders (MDD) compared to those without. The aim of the present study is to establish and compare the performance of different machine learning models based on sleep-wake disorder symptoms data and to select the optimal model to interpret the importance of sleep-wake disorder symptoms to predict MDD occurrence in adolescents. We derived data for this work from 2020 to 2021 Assessing Nocturnal Sleep/Wake Effects on Risk of Suicide Phase I Study from National Sleep Research Resource. Using demographic and sleep-wake disorder symptoms data as predictors and the occurrence of MDD measured base on the center for epidemiologic studies depression scale as an outcome, the following six machine learning predictive models were developed: eXtreme Gradient Boosting model (XGBoost), Light Gradient Boosting mode, AdaBoost, Gaussian Naïve Bayes, Complement Naïve Bayes, and multilayer perceptron. The models' performance was assessed using the AUC and other metrics, and the final model's predictor importance ranking was explained. XGBoost is the optimal predictive model in comprehensive performance with the AUC of 0.804 in the test set. All sleep-wake disorder symptoms were significantly positively correlated with the occurrence of adolescent MDD. The insomnia severity was the most important predictor compared with the other predictors in this study. This machine learning predictive model based on sleep-wake disorder symptoms can help to raise the awareness of risk of symptoms between sleep-wake disorders and MDD in adolescents and improve primary care and prevention.

Low-altitude small target detection in sea clutter background based on improved CEEMDAN-IZOA-ELM

To effectively detect low-altitude small targets under complex sea surface environment, an innovative method has been developed. This method harnesses the chaotic characteristics of sea clutter and employs a combination of Adaptive Noise Complete Ensemble Empirical Modal Decomposition (CEEMDAN), Adaptive Wavelet Thresholding (AWT), and Polynomial Fitting Filtering (SG) for denoising sea clutter data. Subsequently, the Improved Zebra Optimization Algorithm-Extreme Learning Machine (IZOA-ELM) detector is utilized to identify low-altitude small targets amidst the sea clutter background. To begin, the CEEMDAN method is applied to disentangle the measured sea clutter data into a set of Intrinsic Mode Functions (IMFs). Afterwords, the Refined Composite Multiscale Dispersion Entropy (RCMDE) is computed for each individual IMF. This process categorizes the IMFs into three distinct components: noise-dominant, signal-noise mixture, and signal-dominant segments. The noise-dominate of IMF component is subjected to denoising through AWT, the signal-noise mixture of IMF components are processed using SG filtering, while the signal-dominant of IMF remains unaltered. The denoised sea clutter signal is reconstructed by concatenating the denoised and unprocessed IMFs. Based on the chaotic nature of sea clutter signals, first-order sea clutter data is transformed into high-dimensional data through phase space reconstruction. The initial weights and thresholds of the ELM are optimized through the IZOA to establish an optimal prediction model. This model is then used to detect small, low-altitude targets by analyzing the prediction error. The algorithm's effectiveness in noise removal is validated using IPIX and SPRR measured sea clutter data, demonstrating a significant improvement in the root mean square of prediction error (RMSE) by one order of magnitude after denoising compared to the pre-denoising state. Furthermore, we observed that the IZOA-ELM method can be effectively applied to detect small targets at low altitudes across various sea conditions. However, when the sea state is complex and greatly affected by the surrounding noise, an effective approach is to first employ CEEMDAN-AWT-SG to denoise the original signal, and then utilize IZOA-ELM for target detection.

Value of multi-center 18F-FDG PET/CT radiomics in predicting EGFR mutation status in lung adenocarcinoma.

Accurate, noninvasive, and reliable assessment of epidermal growth factor receptor (EGFR) mutation status and EGFR molecular subtypes is essential for treatment plan selection and individualized therapy in lung adenocarcinoma (LUAD). Radiomics models based on 18F-FDG PET/CT have great potential in identifying EGFR mutation status and EGFR subtypes in patients with LUAD. The validation of multi-center data, model visualization, and interpretation are significantly important for the management, application and trust of machine learning predictive models. However, few EGFR-related research involved model visualization and interpretation, and multi-center trial. To develop explainable optimal predictive models based on handcrafted radiomics features (HRFs) extracted from multi-center 18F-FDG PET/CT to predict EGFR mutation status and molecular subtypes in LUAD. Baseline 18F-FDG PET/CT images of 383 LUAD patients from three hospitals and one public data set were collected. Further, 1808 HRFs were extracted from the primary tumor regions using Pyradiomics. Predictive models were built based on cross-combination of seven feature selection methods and seven machine learning algorithms. Yellowbrick and explainable artificial intelligence technology were used for model visualization and interpretation. Receiver operating characteristic curve, classification report and confusion matrix were used for model performance evaluation. Clinical applicability of the optimal models was assessed by decision curve analysis. STACK feature selection method combined with light gradient boosting machine (LGBM) reached optimal performance in identifying EGFR mutation status ([area under the curve] AUC=0.81 in the internal test cohort; AUC=0.62 in the external test cohort). Random forest feature selection method combined with LGBM reached optimal performance in predicting EGFR mutation molecular subtypes (AUC=0.89 in the internal test cohort; AUC=0.61 in the external test cohort). Explainable machine learning models combined with radiomics features extracted from multi-center/scanner 18F-FDG PET/CT have certain potential to identify EGFR mutation status and subtypes in LUAD, which might be helpful to the treatment of LUAD.

P953 Establishment and Internal Validation of a Clinical Risk Score to Predict the Efficacy of Adalimumab in Crohn's Disease

Abstract Background Adalimumab (ADA) plays a crucial role in treating Crohn's disease (CD), but not all patients benefit from it. This study aimed to analyze the factors affecting the treatment effectiveness of ADA for CD and then establish an efficacy prediction model with internal validation. Methods Retrospectively collected data from CD patients receiving ADA at Xijing Hospital from October 2020 to February 2023. Collect clinical data of patients and use FITC-labeled ADA to perform immunofluorescence staining on paraffin tissue sections of CD patients before ADA treatment. The data were randomly split into training and validation. Single-factor and multiple-factor logistic regression models were used to analyze the factors influencing the 12-week clinical remission of CD patients treated with ADA in the training, determining the optimal predictive model. A line chart prediction model was established and the model's discrimination and calibration were tested. Finally, the model was internally validated in the validation set. Results A total of 68 patients were included in the study. The clinical remission rate of CD patients treated with ADA at 12 weeks was 55.88% (38/68). The results showed that baseline endoscopic stenosis [OR=0.019, 95% confidence interval (95% CI) 0.019-0.993], disease course [OR=0.966, 95% CI 0.940-0.993], and the number of FITC-labeled ADA-positive cells [OR=1.109, 95% CI 1.025-1.201] were factors affecting the therapeutic effect of ADA in CD patients (p&amp;lt;0.05) which used to establish the efficacy prediction model. The AUC of the prediction model in the training and validation were 0.868 (95% CI 0.769-0.977) and 0.802 (95% CI 0.604-1) (p&amp;lt;0.05), respectively. The clinical decision curve analysis showed that the prediction model had significant clinical benefits in both the training and validation. Conclusion The prediction model can provide reference for clinicians to make treatment decisions.

A double‐layer neural network wind speed prediction framework based on training set segmentation and error correction

AbstractHighly accurate wind speed prediction is one of the effective ways to improve wind energy utilization. Therefore, this study proposes a novel double‐layer neural network (D‐NN) framework based on feedforward neural networks. First, a feedforward neural network with an extremely short training time is used as the basic learner to ensure that the D‐NN framework has a short training and optimization time. Then, the double‐layer neural network automatically segments and adjusts the length of the training set using an improved error correction method, thus improving the model's ability to capture sequence features and enhancing the model's generalization performance in a non‐directly‐connected superposition manner. Finally, a genetic algorithm is used to automatically optimize the hyperparameters of each sub‐neural network in the D‐NN and the optimal training set length of each sub‐neural network, resulting in the final optimized prediction model. The accuracy and reliability of the prediction models designed based on the D‐NN framework are verified by comparing several different models. The results of the five different prediction evaluation metrics show that the D‐NN framework can greatly improve the accuracy of wind speed prediction.

Spatio-temporal analysis and risk modeling of foot-and-mouth disease outbreaks in China

FMD is an acute contagious disease that poses a significant threat to the health and safety of cloven-hoofed animals in Asia, Europe, and Africa. The impact of FMD exhibits geographical disparities within different regions of China. The present investigation undertook an exhaustive analysis of documented occurrences of bovine FMD in China, spanning the temporal range from 2011 to 2020. The overarching objective was to elucidate the temporal and spatial dynamics underpinning these outbreaks. Acknowledging the pivotal role of global factors in FMD outbreaks, advanced machine learning techniques were harnessed to formulate an optimal prediction model by integrating comprehensive meteorological data pertinent to global FMD. Random Forest algorithm was employed with top three contributing factors including Isothermality(bio3), Annual average temperature(bio1) and Minimum temperature in the coldest month(bio6), all relevant to temperature. By encompassing both local and global factors, our study provides a comprehensive framework for understanding and predicting FMD outbreaks. Furthermore, we conducted a phylogenetic analysis to trace the origin of Foot-and-mouth disease virus (FMDV), pinpointing India as the country posing the greatest potential hazard by leveraging the spatio-temporal attributes of the collected data. Based on this finding, a quantitative risk model was developed for the legal importation of live cattle from India to China. The model estimated an average probability of 0.002254% for FMDV-infected cattle imported from India to China. TA sensitivity analysis identified two critical nodes within the model: he possibility of false negative clinical examination in infected cattle at destination (P5) and he possibility of false negative clinical examination in infected cattle at source(P3). This comprehensive approach offers a thorough evaluation of FMD landscape within China, considering both domestic and global perspectives, thereby augmenting the efficacy of early warning mechanisms.

Explainable Machine Learning-Based Prediction Model for Diabetic Nephropathy.

The aim of this study is to analyze the effect of serum metabolites on diabetic nephropathy (DN) and predict the prevalence of DN through a machine learning approach. The dataset consists of 548 patients from April 2018 to April 2019 in the Second Affiliated Hospital of Dalian Medical University (SAHDMU). We select the optimal 38 features through a least absolute shrinkage and selection operator (LASSO) regression model and a 10-fold cross-validation. We compare four machine learning algorithms, including extreme gradient boosting (XGB), random forest, decision tree, and logistic regression, by AUC-ROC curves, decision curves, and calibration curves. We quantify feature importance and interaction effects in the optimal predictive model by Shapley additive explanation (SHAP) method. The XGB model has the best performance to screen for DN with the highest AUC value of 0.966. The XGB model also gains more clinical net benefits than others, and the fitting degree is better. In addition, there are significant interactions between serum metabolites and duration of diabetes. We develop a predictive model by XGB algorithm to screen for DN. C2, C5DC, Tyr, Ser, Met, C24, C4DC, and Cys have great contribution in the model and can possibly be biomarkers for DN.