High Prediction Research Articles

Multi-view learning framework for predicting unknown types of cancer markers via directed graph neural networks fitting regulatory networks.

The discovery of diagnostic and therapeutic biomarkers for complex diseases, especially cancer, has always been a central and long-term challenge in molecular association prediction research, offering promising avenues for advancing the understanding of complex diseases. To this end, researchers have developed various network-based prediction techniques targeting specific molecular associations. However, limitations imposed by reductionism and network representation learning have led existing studies to narrowly focus on high prediction efficiency within single association type, thereby glossing over the discovery of unknown types of associations. Additionally, effectively utilizing network structure to fit the interaction properties of regulatory networks and combining specific case biomarker validations remains an unresolved issue in cancer biomarker prediction methods. To overcome these limitations, we propose a multi-view learning framework, CeRVE, based on directed graph neural networks (DGNN) for predicting unknown type cancer biomarkers. CeRVE effectively extracts and integrates subgraph information through multi-view feature learning. Subsequently, CeRVE utilizes DGNN to simulate the entire regulatory network, propagating node attribute features and extracting various interaction relationships between molecules. Furthermore, CeRVE constructed a comparative analysis matrix of three cancers and adjacent normal tissues through The Cancer Genome Atlas and identified multiple types of potential cancer biomarkers through differential expression analysis of mRNA, microRNA, and long noncoding RNA. Computational testing of multiple types of biomarkers for 72 cancers demonstrates that CeRVE exhibits superior performance in cancer biomarker prediction, providing a powerful tool and insightful approach for AI-assisted disease biomarker discovery.

Using ensemble learning method and binary decision tree algorithm for drought intensity level classification

This study applies machine learning methods such as Decision Tree (CART) and Random Forest to classify drought intensity based on meteorological data. The goal of the study was to evaluate the effectiveness of these methods for drought classification and their use in water resource management and agriculture. The methodology involved using two machine learning models that analyzed temperature and humidity indicators, as well as wind speed indicators. The models were trained and tested on real meteorological data to assess their accuracy and identify key factors affecting predictions. Results showed that the Random Forest model achieved the highest accuracy of 94.4% when analyzing temperature and humidity indicators, while the Decision Tree (CART) achieved an accuracy of 93.2%. When analyzing wind speed indicators, the models’ accuracies were 91.3% and 93.0%, respectively. Feature importance revealed that atmospheric pressure, temperature at 2 m, and wind speed are key factors influencing drought intensity. One of the study’s limitations was the insufficient amount of data for high drought levels (classes 4 and 5), indicating the need for further data collection. The innovation of this study lies in the integration of various meteorological parameters to build drought classification models, achieving high prediction accuracy. Unlike previous studies, our approach demonstrates that using a wide range of meteorological data can significantly improve drought classification accuracy. Significant findings include the necessity to expand the dataset and integrate additional climatic parameters to improve models and enhance their reliability.

AI-enabled CT-guided end-to-end quantification of total cardiac activity in 18FDG cardiac PET/CT for detection of cardiac sarcoidosis.

[18F]-fluorodeoxyglucose ([18F]FDG) positron emission tomography (PET) plays a central role in diagnosing and managing cardiac sarcoidosis. We propose a fully automated pipeline for quantification of [18F]FDG PET activity using deep learning (DL) segmentation of cardiac chambers on computed tomography (CT) attenuation maps and evaluate several quantitative approaches based on this framework. We included consecutive patients undergoing [18F]FDG PET/CT for suspected cardiac sarcoidosis. DL segmented left atrium, left ventricular(LV), right atrium, right ventricle, aorta, LV myocardium, and lungs from CT attenuation scans. CT-defined anatomical regions were applied to [18F]FDG PET images automatically to target to background ratio (TBR), volume of inflammation (VOI) and cardiometabolic activity (CMA) using full sized and shrunk segmentations. A total of 69 patients were included, with mean age of 56.1 ± 13.4 and cardiac sarcoidosis present in 29 (42%). CMA had the highest prediction performance (area under the receiver operating characteristic curve [AUC] 0.919, 95% confidence interval [CI] 0.858 - 0.980) followed by VOI (AUC 0.903, 95% CI 0.834 - 0.971), TBR (AUC 0.891, 95% CI 0.819 - 0.964), and maximum standardized uptake value (AUC 0.812, 95% CI 0.701 - 0.923). Abnormal CMA (≥1) had a sensitivity of 100% and specificity 65% for cardiac sarcoidosis. Lung quantification was able to identify patients with pulmonary abnormalities. We demonstrate that fully automated volumetric quantification of [18F]FDG PET for cardiac sarcoidosis based on CT attenuation map-derived volumetry is feasible, rapid, and has high prediction performance. This approach provides objective measurements of cardiac inflammation with consistent definition of myocardium and background region.

Comprehensive assessment of matcha qualities and visualization of constituents using hyperspectral imaging technology

Matcha, made from different tea leaves as raw material, exhibits diverse aromas and flavors. Therefore, there is an urgent need for a rapid, non-destructive method to assess the quality of matcha to ensure that these different characteristics are accurately assessed without compromising the integrity of the product. In this study, hyperspectral imaging technology (HSI) combined with machine learning methods enabled the first visual in situ assessment of matcha quality. The physicochemical contents of matcha were determined chemically. Qualitative and quantitative detection models for different types and grades were developed using HSI (containing Vis-NIR and NIR band). The results showed that hyperspectral data in the Vis-NIR were better than in the NIR band. The accuracy of XGBoost in modelling the classification of matcha grades reached 98.10 %. After feature selection using the random forest (RF) method, partial least squares regression (PLSR) was built to predicted the quality of matcha, which showed high prediction accuracy (test set Rp2 > 0.95). The model uses HSI to visually visualize spatial variations in constitutions (catechins, free amino acids, caffeine, soluble proteins, and soluble sugars) to show compositional differences between different types of matcha, providing a rapid non-destructive method for comprehensive assessment of matcha quality.

Experimental study and life prediction for aero-engine turbine blade considering creep-fatigue interaction effect

As an important failure form of the turbine blade, creep-fatigue interaction damage affects the safe operation and maintenance strategy of aero-engine, and has been the focus of scientific research and the academic community. Firstly, based on the Kachanov-Rabotnov-Lemaitre continuum damage mechanics theory and the nonlinear symmetry of creep damage and fatigue damage, a creep-fatigue life prediction model is constructed considering the interaction effect in this paper. Then, based on the thermal-fluid–solid multi-physical field coupling numerical simulation of the turbine blade, the equivalent method of creep-fatigue load spectrum was explored according to the equal damage criterion and linear damage rule, and the creep-fatigue interaction test of smooth samples of the blade material was conducted to analyze the creep-fatigue fracture morphology. Finally, the stress term in the creep-fatigue life prediction model of blade material is modified by the correction factor α, and the modified creep-fatigue life prediction model of the turbine blade is constructed considering the interaction effect. The results show that the modified creep-fatigue life prediction model considering the interaction effect has a high life prediction ability with an error of 3.9%. The above research has important scientific research value for the life extension design of turbine blades and the improvement of aero-engine maintenance strategy.

Research on Machining Quality Prediction Method Based on Machining Error Transfer Network and Grey Neural Network

Machining quality prediction is the critical link of quality control in parts machining. With the advent of the Industry 4.0 era, intelligent manufacturing and data-driven technologies bring new ideas for quality control in complex machining processes. Quality control is complicated for multi-process, multi-condition, small-batch, and high-precision parts processing requirements. To solve this problem, this paper proposes a machining quality prediction method based on the machining error transfer network and the grey neural network. Initially, by constructing a processing error transfer network, the error transfer law in part processing is described, and the PageRank algorithm and the influence degree of the nodes are used to determine the critical quality features. Additionally, the problem of low prediction accuracy due to small sample data and multiple coupling relationships is solved using the grey neural network algorithm, and a high accuracy prediction of critical quality features is achieved. Finally, the effectiveness and reliability of the method are verified by the case of medium-speed marine diesel engine fuselage processing. The results indicate that this method not only effectively identifies critical quality features in the machining process of complex parts, but it also maintains a high predictive accuracy for these features, even with small samples and limited data.

Remote Interactions between tropical cyclones: The case of Hurricane Michael and Leslie's high predictability uncertainty

The study explores Hurricane Michael's impact on Hurricane Leslie's trajectory predictability using ECMWF and NCEP ensemble systems. A clustering method focused on tropical cyclones is used to identify potential paths for Leslie: Cluster 1 accurately predicted Leslie's direction towards the Iberian Peninsula, whereas Clusters 2 and 3 indicated a southern recurve near the Canary Islands. Analysis of potential vorticity and irrotational wind at upper levels showed a significant interaction between Michael, ridge, and trough across the jet stream from +12 h after initialization. Cluster 1 showed a stronger Michael promoting upper-level wind divergence greatest, modifying the jet stream configuration around the ridge and downstream. Alterations in the jet stream's configuration, functioning as a waveguide, propagated downstream, guiding Leslie towards the Iberian Peninsula. Clusters 2 and 3 indicated the trough's failure to incorporate Leslie, resulting in a recurve of the trajectory around the Azores anticyclone. This research enhances comprehension of the interaction between two tropical cyclones via synoptic Rossby wave flow. Moreover, the conceptual framework can aid operational meteorologists in identifying the sources of uncertainty, particularly in track forecasts under synoptic conditions analogous to those examined in this study.

Physics-informed neural networks for biopharmaceutical cultivation processes: Consideration of varying process parameter settings.

We present a new modeling approach for the study and prediction of important process outcomes of biotechnological cultivation processes under the influence of process parameter variations. Our model is based on physics-informed neural networks (PINNs) in combination with kinetic growth equations. Using Taylor series, multivariate external process parameter variations for important variables such as temperature, seeding cell density and feeding rates can be integrated into the corresponding kinetic rates and the governing growth equations. In addition to previous approaches, PINNs also allow continuous and differentiable functions as predictions for the process outcomes. Accordingly, our results show that PINNs in combination with Taylor-series expansions for kinetic growth equations provide a very high prediction accuracy for important process variables such as cell densities and concentrations as well as a detailed study of individual and combined parameter influences. Furthermore, the proposed approach can also be used to evaluate the outcomes of new parameter variations and combinations, which enables a saving of experiments in combination with a model-driven optimization study of the design space.

Ultrasound-based radiomics nomogram for predicting HER2-low expression breast cancer.

Accurate preoperative identification of Human epidermal growth factor receptor 2 (HER2) low expression breast cancer (BC) is critical for clinical decision-making. Our aim was to use machine learning methods to develop and validate an ultrasound-based radiomics nomogram for predicting HER2-low expression in BC. In this retrospective study, 222 patients (108 HER2-0 expression and 114 HER2-low expression) with BC were included. The enrolled patients were randomly divided into a training cohort and a test cohort with a ratio of 8:2. The tumor region of interest was manually delineated from ultrasound image, and radiomics features were subsequently extracted. The features underwent dimension reduction using the least absolute shrinkage and selection operator (LASSO) algorithm, and rad-score were calculated. Five machine learning algorithms were applied for training, and the algorithm demonstrating the best performance was selected to construct a radiomics (USR) model. Clinical risk factors were integrated with rad-score to construct the prediction model, and a nomogram was plotted. The performance of the nomogram was assessed using receiver operating characteristic curve and decision curve analysis. A total of 480 radiomics features were extracted, out of which 11 were screened out. The majority of the extracted features were wavelet features. Subsequently, the USR model was established, and rad-scores were computed. The nomogram, incorporating rad-score, tumor shape, border, and microcalcification, achieved the best performance in both the training cohort (AUC 0.89; 95%CI 0.836-0.936) and the test cohort (AUC 0.84; 95%CI 0.722-0.958), outperforming both the USR model and clinical model. The calibration curves showed satisfactory consistency, and DCA confirmed the clinical utility of the nomogram. The nomogram model based on ultrasound radiomics exhibited high prediction value for HER2-low BC.

Prediction of the minimum miscibility pressure for CO2 flooding based on a physical information neural network algorithm

Abstract The minimum miscibility pressure (MMP) is a crucial parameter in assessing the miscibility of CO2 displacement and evaluating the effectiveness of oil displacement. Traditional methods for calculating MMP are intricate and time-consuming, involving numerous related parameters. Therefore, precise and efficient determination of MMP is highly significant in the development of CO2-driven reservoirs. This study first utilized the Pearson correlation coefficient to analyse the correlation factor mechanism of 36 sets of fine-tube experimental data. Subsequently, the physical information neural network prediction model was employed with reservoir temperature, crude oil composition, and injected gas type as input parameters. The PRI state equation and Glaso correlation equation drove the model, with parameter optimization and training conducted under both physical and data driving. The model demonstrates high prediction accuracy and strong generalization ability. Finally, Validation of the model was performed using fine-tube experimental data from 5 other wells, revealing a relatively small relative deviation between calculated and experimental values, with an average coefficient of determination of 0.95 and an average relative error of 4.42%. The prediction accuracy was improved by about 75% compared to other machine learning algorithms. This model holds potential for application in on-site reservoir development, enhancing the measurement accuracy of the minimum miscible pressure of pure CO2 flooding, greatly shortening the design cycle of reservoir development, expediting the process of reservoir development, and providing technical guidance for improving oil and gas recovery and pure CO2 flooding exploration and development.

Chemical Graph-Based Transformer Models for Yield Prediction of High-Throughput Cross-Coupling Reaction Datasets.

The chemical reaction yield is an important factor to determine the reaction conditions. Recently, many data-driven models for yield prediction using high-throughput experimentation datasets have been reported. In this study, we propose a neural network architecture based on the chemical graphs of the reaction components to predict the reaction yield. The proposed model is the sequential combination of a message-passing neural network and a transformer encoder (MPNN-Transformer). The reaction components are converted to molecular matrices by the first network, followed by the interplay of the reaction components in the second network after adding the embeddings of the compound roles in the chemical reaction. The predictive ability of the proposed models was compared with state-of-the-art yield prediction models using two high-throughput experimental datasets: the Buchwald-Hartwig cross-coupling (BHC) and Suzuki-Miyaura cross-coupling (SMC) reaction datasets. Overall, the MPNN-Transformer models showed high prediction accuracy for the BHC reaction datasets and some of the extrapolation-oriented SMC reaction datasets. These models also performed well when the training dataset size was relatively large. Furthermore, analyzing the poorly predicted reactions for the BHC reaction dataset revealed a limitation of the data-driven yield prediction approach based on the chemical structural similarity.

Seismic response prediction for buildings with sliding live loads using neural networks

The seismic response of primary structures (PS) can be significantly influenced by live loads, particularly when these loads consist of stacked bodies that are capable of sliding during seismic events. Conventional design methods often fail to account for the energy dissipation resulting from such sliding, leading to overly conservative structural estimates. This study examines the effects of sliding live loads on the seismic response of a PS, where two rigid bodies, referred to as secondary bodies (SBs), are stacked as live loads on a PS. The interaction between the lower SB and the PS, as well as between the SBs, is modelled using Coulomb's friction model. The governing equations are solved using the fourth-order Runge-Kutta method. Seismic Zones III and V from IS 1893:2016 are considered as hazard levels. A parametric investigation explores the impact of varying dynamic parameters of both the PS and SBs on the seismic response. The results demonstrate significant energy dissipation within the stack due to sliding, which, if not considered, results in conservative displacement estimates. The study also evaluates how friction coefficients, mass ratios, and excitation levels affect the stack's energy dissipation capacity. A novel method to compute the modified primary structural period (Tnew) for design purposes is proposed, and an artificial neural network (ANN) model is developed, achieving a high prediction accuracy (R2 = 99 %). Sensitivity analysis is performed to assess the influence of input parameters on the output. This research highlights the need to include sliding loads in seismic design.

PyGRF: An Improved Python Geographical Random Forest Model and Case Studies in Public Health and Natural Disasters

ABSTRACTGeographical random forest (GRF) is a recently developed and spatially explicit machine learning model. With the ability to provide more accurate predictions and local interpretations, GRF has already been used in many studies. The current GRF model, however, has limitations in its determination of the local model weight and bandwidth hyperparameters, potentially insufficient numbers of local training samples, and sometimes high local prediction errors. Also, implemented as an R package, GRF currently does not have a Python version which limits its adoption among machine learning practitioners who prefer Python. This work addresses these limitations by introducing theory‐informed hyperparameter determination, local training sample expansion, and spatially weighted local prediction. We also develop a Python‐based GRF model and package, PyGRF, to facilitate the use of the model. We evaluate the performance of PyGRF on an example dataset and further demonstrate its use in two case studies in public health and natural disasters.

Construction and evaluation of sarcopenia risk prediction model for patients with diabetes: a study based on the China health and retirement longitudinal study (CHARLS)

PurposeSarcopenia is a common complication of diabetes. Nevertheless, precise evaluation of sarcopenia risk among patients with diabetes is still a big challenge. The objective of this study was to develop a nomogram model which could serve as a practical tool to diagnose sarcopenia in patients with diabetes.MethodsA total of 783 participants with diabetes from China Health and Retirement Longitudinal Study (CHARLS) 2015 were included in this study. After oversampling process, 1,000 samples were randomly divided into the training set and internal validation set. To mitigate the overfitting effect caused by oversampling, data of CHARLS 2011 were utilized as the external validation set. Least absolute shrinkage and selection operator (LASSO) regression analysis and multivariate logistic regression analysis were employed to explore predictors. Subsequently, a nomogram was developed based on the 9 selected predictors. The model was assessed by area under receiver operating characteristic (ROC) curves (AUC) for discrimination, calibration curves for calibration, and decision curve analysis (DCA) for clinical efficacy. In addition, machine learning models were constructed to enhance the robustness of our findings and evaluate the importance of the predictors.Results9 factors were selected as predictors of sarcopenia for patients with diabetes. The nomogram model exhibited good discrimination in training, internal validation and external validation sets, with AUC of 0.808, 0.811 and 0.794. machine learning models revealed that age and hemoglobin were the most significant predictors. Calibration curves and DCA illustrated excellent calibration and clinical applicability of this model.ConclusionThis comprehensive nomogram presented high clinical predictability, which was a promising tool to evaluate the risk of sarcopenia in patients with diabetes.

Rapid transformation of nanobodies affinity based on AlphaFold2's high-accuracy predictions and interaction analysis for enrofloxacin detection in coastal fish

High-affinity antibodies are crucial in biosensors, disease diagnostics, therapeutic drug development, and immunological analysis, making the enhancement of antibody affinity a key research focus within the field. Computer-aided design is recognized as a time-saving and labor-efficient method for nanobodies in vitro affinity maturation. Compared to experimental mutagenesis techniques, it is advantageous due to the elimination of the need for laborious library construction and screening processes. However, these approaches are constrained by structural prediction since inaccuracy in structure could readily result in maturation failures. Herein, a novel nanobodies modification method for in vitro affinity maturation, utilizing the high accuracy prediction of AlphaFold2, was employed to rapidly transform a low affinity nanobody against enrofloxacin (ENR) into one with high affinity. The molecular docking results revealed a 1.5- to 2.5-fold increase in the number of noncovalent interactions of modified nanobodies, accompanied by a reduction in binding free energy ranging from 14.1 to 62.6%. The evaluation results from ELISA and BLI indicated that the affinity of the modified nanobodies had been enhanced by 6.2–91.6 times compared to the template nanobody. Furthermore, the modified nanobodies were employed for the detection of ENR-spiked coastal fish samples. In summary, this research proposed a nanobodies modification method from a new perspective, endowing its great application potential in biosensors, food safety, and environmental monitoring.

Few-shot learning for screening 2D Ga2CoS4−x supported single-atom catalysts for hydrogen production

Hydrogen generation and related energy applications heavily rely on the hydrogen evolution reaction (HER), which faces challenges of slow kinetics and high overpotential. Efficient electrocatalysts, particularly single-atom catalysts (SACs) on two-dimensional (2D) materials, are essential. This study presents a few-shot machine learning (ML) assisted high-throughput screening of 2D septuple-atomic-layer Ga2CoS4−x supported SACs to predict HER catalytic activity. Initially, density functional theory (DFT) calculations showed that 2D Ga2CoS4 is inactive for HER. However, defective Ga2CoS4−x (x = 0–0.25) monolayers exhibit excellent HER activity due to surface sulfur vacancies (SVs), with predicted overpotentials (0–60 mV) comparable to or lower than commercial Pt/C, which typically exhibits an overpotential of around 50 mV in the acidic electrolyte, when the concentration of surface SV is lower than 8.3%. SVs generate spin-polarized states near the Fermi level, making them effective HER sites. We demonstrate ML-accelerated HER overpotential predictions for all transition metal SACs on 2D Ga2CoS4−x. Using DFT data from 18 SACs, an ML model with high prediction accuracy and reduced computation time was developed. An intrinsic descriptor linking SAC atomic properties to HER overpotential was identified. This study thus provides a framework for screening SACs on 2D materials, enhancing catalyst design.

Energy management optimization of hybrid electric vehicles based on deep learning model predictive control

In this paper, the hybrid electric vehicle (HEV) energy management optimization method is proposed based on deep learning (DL) model predictive control. Through empirical research combined with the questionnaire survey, this article not only provides a new perspective and practical basis but also improves the efficiency and accuracy of the model by improving the relevant algorithms. The study first analyzes the importance of HEV energy management and reviews the existing literature. Then, the optimization method of HEV energy management based on the deep learning model is introduced in detail, including the composition of energy management for hybrid electric vehicles, the structure and working principle of the deep learning model, especially the backpropagation neural network (BPNN) and the convolutional neural network (CNN), and the steps of application of deep learning in energy management. In the experimental part, questionnaire data from 1,500 consumers were used to design the HEV energy management optimization scheme, and consumers’ attitudes and preferences towards HEV energy optimization were discussed. The experimental results show that the proposed model can predict HEV energy consumption under different road conditions (urban roads, highways, mountain areas, suburban areas, and construction sites), and the difference between the average predicted energy consumption and the actual energy consumption is between 0.1KWH and 0.3KWH, showing high prediction accuracy. In addition, the deep learning-based energy management strategy outperforms traditional control strategies in terms of fuel consumption (6.2 L/100 km), battery charge and discharge times (814), battery life, and CO2 emissions, significantly improving the efficiency of HEV energy. These results demonstrate the great potential and practical application value of deep learning models in the optimization of energy management of HEVs, helping to drive the development of more sustainable and efficient transportation systems.

Deeper insights into flame retardancy of polymers by interpretable, quantifiable, yet accurate machine-learning model

Fire-safety polymer materials are essential in modern society such as electronics, aerospace, new energy. The quantification and prediction of flame retardancy, determined by the chemical composition and the burning process, has always been a bottleneck challenge. Previous empirical design rules and the existing models show large deviations for predicting flame retardancy and are often unexplainable. Here, this study proposes an interpretable model that can quantify the groups contribution of flame-retardancy and predict the flame retardance of intrinsically flame-retardant polymers. The machine learning model simultaneously considers the group structures and their flame-retardant mechanism in both the gas phase and condensed phase, achieving high prediction accuracy (89.8% for the training set and 83.8% for the testing set). It also quantifies the contribution values of various flame-retardant groups (halogen-containing structures, phosphorus-containing structures, phosphorus-nitrogen-containing structures, aromatic ring-containing structures, etc.) in both phases for the first time. The running script that integrates the model has also been open-sourced, providing an emerging strategy for transitioning flame-retardant research from empirical methods to scientific design.

Exploring clinical factors to predict the survival of patients with resectable non-small cell lung cancer with neoadjuvant immunotherapy.

The goal was to explore clinical factors and build a predictive model for the disease-free and overall survival of patients with non-small cell lung cancer (NSCLC) receiving neoadjuvant chemotherapy combined with immune checkpoint inhibitors. Inclusion criteria for patients in this multicentre study were as follows: (i) Patients who were diagnosed with stages I-III NSCLC after a bronchoscopy biopsy or puncture; (ii) patients who were examined with computed tomography/positron emission tomography-computed tomography before treatment and surgery; (iii) patients who received neoadjuvant chemotherapy combined with immune checkpoint inhibitors for 2 to 6 cycles preoperatively; (iv) patients whose peripheral blood indicators and tumour markers were assessed before treatment and preoperatively; (v) patients who underwent radical lung cancer surgery after neoadjuvant therapy. Cases were divided into high- and low-risk groups according to 78 clinical indicators based on a 10-fold Least Absolute Shrinkage and Selection Operator selection. We used Cox proportional hazards models to predict disease-free and overall survival. Then, we used time-dependent area under the curve and decision curve analyses to examine the accuracy of the results. Data were collected continuously, and 212 and 85 cases were randomly assigned to training and testing sets, respectively. The area under the curve for the prediction of disease-free survival (training: 1 year, 0.83; 2 years, 0.81; 3 years, 0.83 versus testing: 1 year, 0.65; 2 years, 0.66; 3 years, 0.70), overall survival (training: 1 year, 0.86; 2 years, 0.85; 3 years, 0.86 versus testing: 1 year, 0.66; 2 years, 0.57; 3 years, 0.70) were determined. The coefficient factors including pathological response; preoperative tumour maximum diameter; preoperative lymph shorter diameter; preoperative tumour and lymph maximum standardized uptake value; change in tumour standardized uptake value preoperatively; and blood-related risk factors were favourably associated with prognosis (P < 0.001). Our prediction model, which integrated data from preoperative positron emission tomography-CT, preoperative blood parameters and pathological response, was able to make highly accurate predictions for disease-free and overall survival in patients with NSCLC receiving neoadjuvant immunity with chemical therapy.

Innovative molecular descriptors in QSPR modeling: Integrating Carnahan-Starling EoS for predicting diffusion coefficients in hydrocarbons and mixtures

The rapid and accurate prediction of self-diffusion coefficients for pure fluids and mutual diffusion coefficients for binary mixtures remains a focal point of current research. In this work, molecular descriptors are innovatively defined by introducing molecular repulsive and attractive pressures based on the Carnahan-Starling equation of state. This approach aims to improve the quantitative structure–property relationship (QSPR) models for predicting thermophysical properties that vary with thermodynamic conditions. The shapely additive explanation method is employed for a comprehensive analysis of the model mechanism. Among the six molecular descriptors selected, our defined attractive pressure (patt), MATS2e and SpMin3_Bh(v) contribute significantly to the self-diffusion coefficient QSPR model. Based on this approach, QSPR models are constructed for 15 hydrocarbons and five of their binary mixtures. The results demonstrate that the GA-BPNN-based QSPR model for pure fluids achieves high prediction accuracy, with an external Rext2 of 0.978 and an external RMSEext of 5.349 × 10−9·m2·s−1. The optimal mixture descriptor type is x1d1+x2d2, and the GS-SVM-based mixture QSPR model predicts 98.24 % of values within a 5 % error margin, with Rext2 of 0.991 and RMSEext of 0.042 × 10−9·m2·s−1.