Training Set Research Articles

DSIL-DDI: A Domain-Invariant Substructure Interaction Learning for Generalizable Drug-Drug Interaction Prediction.

Drug-drug interactions (DDIs) trigger unexpected pharmacological effects in vivo, often with unknown causal mechanisms. Deep learning methods have been developed to better understand DDI. However, learning domain-invariant representations for DDI remains a challenge. Generalizable DDI predictions are closer to reality than source domain predictions. For existing methods, it is difficult to achieve out-of-distribution (OOD) predictions. In this article, focusing on substructure interaction, we propose DSIL-DDI, a pluggable substructure interaction module that can learn domain-invariant representations of DDIs from source domain. We evaluate DSIL-DDI on three scenarios: the transductive setting (all drugs in test set appear in training set), the inductive setting (test set contains new drugs that were not present in training set), and OOD generalization setting (training set and test set belong to two different datasets). The results demonstrate that DSIL-DDI improve the generalization and interpretability of DDI prediction modeling and provides valuable insights for OOD DDI predictions. DSIL-DDI can help doctors ensuring the safety of drug administration and reducing the harm caused by drug abuse.

Nomograms established for predicting microvascular invasion and early recurrence in patients with small hepatocellular carcinoma

BackgroundIn this study, we aimed to establish nomograms to predict the microvascular invasion (MVI) and early recurrence in patients with small hepatocellular carcinoma (SHCC), thereby guiding individualized treatment strategies for prognosis improvement.MethodsThis study retrospectively analyzed 326 SHCC patients who underwent radical resection at Wuhan Union Hospital between April 2017 and January 2022. They were randomly divided into a training set and a validation set at a 7:3 ratio. The preoperative nomogram for MVI was constructed based on univariate and multivariate logistic regression analysis, and the prognostic nomogram for early recurrence was constructed based on univariate and multivariate Cox regression analysis. We used the receiver operating characteristic (ROC) curves, area under the curves (AUCs), and calibration curves to estimate the predictive accuracy and discriminability of nomograms. Decision curve analysis (DCA) and Kaplan-Meier survival curves were employed to further confirm the clinical effectiveness of nomograms.ResultsThe AUCs of the preoperative nomogram for MVI on the training set and validation set were 0.749 (95%CI: 0.684–0.813) and 0.856 (95%CI: 0.805–0.906), respectively. For the prognostic nomogram, the AUCs of 1-year and 2-year RFS respectively reached 0.839 (95%CI: 0.775–0.903) and 0.856 (95%CI: 0.806–0.905) in the training set, and 0.808 (95%CI: 0.719–0.896) and 0.874 (95%CI: 0.804–0.943) in the validation set. Subsequent calibration curves, DCA analysis and Kaplan-Meier survival curves demonstrated the high accuracy and efficacy of the nomograms for clinical application.ConclusionsThe nomograms we constructed could effectively predict MVI and early recurrence in SHCC patients, providing a basis for clinical decision-making.

A dielectric method for predicting pear firmness combining deep data augmentation and ensemble learning

Firmness is valuable in evaluating pear quality, as it determines ripeness and storability.This research presents a new method that combines deep data augmentation and ensemble learning for predicting firmness in three pear cultivars using dielectric spectra. Firstly, 126 sets of pear dielectric spectra and firmness data were obtained, and the characteristic dielectric frequencies were selected based on the prediction accuracy of partial least squares regression (PLSR) model established using different preprocessing methods. Then the generative adversarial networks (GAN), least squares GAN (LSGAN), and Wasserstein GAN (WGAN) models with different numbers of generated samples in the model's training set on the prediction accuracy of the PLSR model were compared. Finally, the optimal GAN model was combined with support vector regression (SVR), artificial neural networks (ANN), random forest (RF), and AdaBoost decision tree (AdaBoost-DT) to predict pear firmness. Results showed that the selected characteristic dielectric frequencies included two ε' and eight ε" points. GAN, LSGAN, and WGAN all improved the prediction accuracy of the PLSR model, and the model accuracy improvement of GAN and LSGAN was proportional to the generated sample numbers added to the training set. GAN had the highest accuracy improvement for PLSR model, with the mean Rp2 (determination coefficient of prediction set) and RMSEP (root mean square error of prediction set) increasing up to 13.70% and -19.48% during 100-1000 epochs, respectively. Ensemble learning models (RF, AdaBoost-DT) outperformed SVR and ANN. GAN-AdaBoost-DT model added 126 sets of generated data in the training set had the highest prediction accuracy, with average Rp2 and RMSEP of 0.90 and 3.35 N during 300-1000 epochs, respectively. This study proved the feasibility of deep data augmentation and ensemble learning methods to improve the prediction accuracy of pear firmness model without increasing the labor and time costs.

Development of a CT-Based comprehensive model combining clinical, radiomics with deep learning for differentiating pulmonary metastases from noncalcified pulmonary hamartomas: a retrospective cohort study.

Clinical differentiation between pulmonary metastases and noncalcified pulmonary hamartomas (NCPH) often presents challenges, leading to potential misdiagnosis. However, the efficacy of a comprehensive model that integrates clinical features, radiomics, and deep learning (CRDL) for differential diagnosis of these two diseases remains uncertain. This study evaluated the diagnostic efficacy of a CRDL model in differentiating pulmonary metastases from NCPH. The authors retrospectively analyzed the clinical and imaging data of 256 patients from the First Medical Centre of the General Hospital of the People's Liberation Army (PLA) and 85 patients from Shanghai Changhai Hospital, who were pathologically confirmed pulmonary hamartomas or pulmonary metastases after thoracic surgery. Employing Python 3.7 software suites, the authors extracted radiomic features and deep learning (DL) attributes from patient datasets. The cohort was divided into training set, internal validation set, and external validation set. The diagnostic performance of the constructed models was evaluated using receiver operating characteristic (ROC) curve analysis to determine their effectiveness in differentiating between pulmonary metastases and NCPH. Clinical features such as white blood cell count (WBC), platelet count (PLT), history of cancer, carcinoembryonic antigen (CEA) level, tumor marker status, lesion margin characteristics (smooth or blurred), and maximum diameter were found to have diagnostic value in differentiating between the two diseases. In the domains of radiomics and DL. Of the 1130 radiomics features and 512 DL features, 24 and 7, respectively, were selected for model development. The area under the ROC curve (AUC) values for the four groups were 0.980, 0.979, 0.999, and 0.985 in the training set, 0.947, 0.816, 0.934, and 0.952 in the internal validation set, and 0.890, 0.904, 0.923, and 0.938 in the external validation set. This demonstrated that the CRDL model showed the greatest efficacy. The comprehensive model incorporating clinical features, radiomics, and DL shows promise for aiding in the differentiation between pulmonary metastases and hamartomas.

Forecasting rheumatoid arthritis patient arrivals by including meteorological factors and air pollutants

The burden of rheumatoid arthritis (RA) has gradually elevated, increasing the need for medical resource redistribution. Forecasting RA patient arrivals can be helpful in managing medical resources. However, no relevant studies have been conducted yet. This study aims to construct a long short-term memory (LSTM) model, a deep learning model recently developed for novel data processing, to forecast RA patient arrivals considering meteorological factors and air pollutants and compares this model with traditional methods. Data on RA patients, meteorological factors and air pollutants from 2015 to 2022 were collected and normalized to construct moving average (MA)- and autoregressive (AR)-based and LSTM models. After data normalization, the root mean square error (RMSE) was adopted to evaluate models’ forecast ability. A total of 2422 individuals were enrolled. Not using the environmental data, the RMSEs of the MA- and AR-based models’ test sets are 0.131, 0.132, and 0.117 when the training set: test set ratio is 2:1, 3:1, and 7:1, while they are 0.110, 0.130, and 0.112 for the univariate LSTM models. Considering meteorological factors and air pollutants, the RMSEs of the MA- and AR-based model test sets were 0.142, 0.303, and 0.164 when the training set: test set ratio is 2:1, 3:1, and 7:1, while they were 0.108, 0.119, and 0.109 for the multivariable LSTM models. Our study demonstrated that LSTM models can forecast RA patient arrivals more accurately than MA- and AR-based models for datasets of all three sizes. Considering the meteorological factors and air pollutants can further improve the forecasting ability of the LSTM models. This novel method provides valuable information for medical management, the optimization of medical resource redistribution, and the alleviation of resource shortages.

A nomogram for predicting colorectal cancer liver metastasis using circulating tumor cells from the first drainage vein

PurposeTo use circulating tumor cells (CTC) from the first drainage vein (FDV) of the primary lesion and other clinically relevant parameters to construct a nomogram for predicting liver metastasis in colorectal cancer (CRC) patients, and to provide a theoretical basis for clinical diagnosis and treatment. MethodsInformation from 343 CRC patients was collected and a database was established. Multivariate logistic analysis was used to identify independent factors for colorectal cancer liver metastasis(mCRC) and nomograms were constructed. Receiver operating characteristic curves(ROC), calibration plots, and decision curve analysis (DCA) were used to assess discrimination, agreement with actual risk, and the clinical utility of the prediction model, respectively. ResultCTC levels in FDV were significantly higher in patients with liver metastasis than in those without liver metastasis. Logistic multivariate analysis showed that vascular invasion, T stage, carcinoembryonic antigen (CEA), CA19-9, and CTC could be used as predictors to construct nomograms. The nomograms showed good discriminatory ability in predicting mCRC, with area under the curve (AUC) values of 0.871 [95 % CI: 0.817–0.924) and 0.891 (95 % CI: 0.817–0.964) for the training and validation sets, respectively.] The calibration curves of both the training and validation sets showed that the model was effective in predicting the probability of mCRC. DCA was used to evaluate this predictive model and showed good net clinical benefit. ConclusionWe developed and validated a nomogram model based on the combination of CTC in the FDV with other clinical parameters to better predict the occurrence of mCRC.

Differentiating low-risk thymomas from high-risk thymomas: preoperative radiomics nomogram based on contrast enhanced CT to minimize unnecessary invasive thoracotomy

BackgroundThis study was designed to develop a combined radiomics nomogram to preoperatively predict the risk categorization of thymomas based on contrast-enhanced computed tomography (CE-CT) images.MaterialsThe clinical and CT data of 178 patients with thymoma (100 patients with low-risk thymomas and 78 patients with high-risk thymomas) collected in our hospital from March 2018 to July 2023 were retrospectively analyzed. The patients were randomly divided into a training set (n = 125) and a validation set (n = 53) in a 7:3 ratio. Qualitative radiological features were recorded, including (a) tumor diameter, (b) location, (c) shape, (d) capsule integrity, (e) calcification, (f) necrosis, (g) fatty infiltration, (h) lymphadenopathy, and (i) enhanced CT value. Radiomics features were extracted from each CE-CT volume of interest (VOI), and the least absolute shrinkage and selection operator (LASSO) algorithm was performed to select the optimal discriminative ones. A combined radiomics nomogram was further established based on the clinical factors and radiomics scores. The differentiating efficacy was determined using receiver operating characteristic (ROC) analysis.ResultsOnly one clinical factor (incomplete capsule) and seven radiomics features were found to be independent predictors and were used to establish the radiomics nomogram. In differentiating low-risk thymomas (types A, AB, and B1) from high-risk ones (types B2 and B3), the nomogram demonstrated better diagnostic efficacy than any single model, with the respective area under the curve (AUC), accuracy, sensitivity, and specificity of 0.974, 0.921, 0.962 and 0.900 in the training cohort, 0.960, 0.892, 0923 and 0.897 in the validation cohort, respectively. The calibration curve showed good agreement between the prediction probability and actual clinical findings.ConclusionsThe nomogram incorporating clinical factors and radiomics features provides additional value in differentiating the risk categorization of thymomas, which could potentially be useful in clinical practice for planning personalized treatment strategies.

PreCNet: Next-Frame Video Prediction Based on Predictive Coding.

Predictive coding, currently a highly influential theory in neuroscience, has not been widely adopted in machine learning yet. In this work, we transform the seminal model of Rao and Ballard (1999) into a modern deep learning framework while remaining maximally faithful to the original schema. The resulting network we propose (PreCNet) is tested on a widely used next-frame video prediction benchmark, which consists of images from an urban environment recorded from a car-mounted camera, and achieves state-of-the-art performance. Performance on all measures (MSE, PSNR, and SSIM) was further improved when a larger training set (2M images from BDD100k) pointed to the limitations of the KITTI training set. This work demonstrates that an architecture carefully based on a neuroscience model, without being explicitly tailored to the task at hand, can exhibit exceptional performance.

Predicting Sprint Performance From the Vertical and Horizontal Jumps in National Football League Combine Athletes.

Agar-Newman, DJ, MacRae, F, Tsai, M-C, and Klimstra, M. Predicting sprint performance from the vertical and horizontal jumps in National Football League Combine athletes. J Strength Cond Res 38(8): 1433-1439, 2024-Identifying fast athletes is an important part of the National Football League (NFL) Combine. However, not all athletes partake in the 36.58-m sprint, and relying on this single test may miss potentially fast athletes. Therefore, the purpose of this study was to determine whether sprinting times can be predicted using simple anthropometric and jumping measures. Data from the NFL Combine between the years 1999-2020 inclusive were used (n = 4,149). Subjects had a mean (±SD) height = 1.87 ± 0.07 m and body mass = 111.96 ± 20.78 kg. The cross-validation technique was used, partitioning the data into a training set (n = 2,071) to develop regression models to predict time over the 9.14-, 9.14- to 18.29-, 18.29- to 36.58-m, and 36.58-m segments using vertical jump, broad jump, height, and mass as the independent variables. The models were then evaluated against a test set (n = 2,070) for agreement. Statistically significant (p < 0.01) models were determined for 9.14-m time (adjusted R2 = 0.76, SEE = 0.05 seconds), 9.14- to 18.29-m time (adjusted R2 = 0.74, SEE = 0.04 seconds), 18.29- to 36.59-m time (adjusted R2 = 0.79, SEE = 0.07 seconds), and 36.58-m time (adjusted R2 = 0.84, SEE = 0.12 seconds). When evaluated against the test set, the models showed biases of -0.05, -0.04, -0.02, and -0.02 seconds and root-mean-square error of 0.07, 0.05, 0.07, and 0.12 seconds for the 9.14-, 9.14- to 18.29-, 18.29- to 36.58-m, and 36.58-m segments, respectively. However, 5-6% of the predictions lay outside of the limits of agreement. This study provides 4 formulae that can be used to predict sprint performance when the 36.58-m sprint test is not performed, and practitioners can use these equations to determine training areas of opportunity when working with athletes preparing for the NFL Combine.

Identification of Biomarkers for Lung Adenocarcinoma With Qi Deficiency and Phlegm Dampness.

Qi deficiency and phlegm dampness (QPD) is one of the most common traditional Chinese medicine (TCM) syndromes in lung adenocarcinoma (LUAD). This study aimed to identify syndrome-specific biomarkers for LUAD with QPD syndrome. Peripheral blood mononuclear cells (PBMCs) from LUAD patients with QPD, LUAD patients with non-QPD (N-QPD), and healthy control (H) were collected and analyzed with RNA-seq to identify differentially expressed genes (DEGs). The area under the receiver operator characteristic curve (AUC) of each DEG was calculated, and the top 10 highest AUC DEGs were validated by qRT-PCR. Logistic regression analysis was used to develop a diagnostic model evaluated with AUC. A total of 135 individuals were enrolled in this study (training set: 15 QPD, 15 N-QPD, 15 H; validation set: 30 QPD, 30 N-QPD, 30 H). A total of 1480 DEGs were identified between QPD and N-QPD. The qRT-PCR results showed that the expression of DDR2 was downregulated, and PPARG was upregulated, which was in line with the finding of the training set. We developed a diagnostic model with these two genes. The AUC of the diagnostic model in the training cohort and validation cohort was 0.891 and 0.777, respectively. We identified the two genes (DDR2 and PPARG) as syndrome-specific biomarkers for LUAD with QPD syndrome and developed a novel diagnostic model, which may help to improve the accuracy and sensibility of clinical diagnosis and provide a new target for natural drug treatment of LUAD.

A Computational Study of Green Tea Extracts and their Derivatives as Potential Inhibitors for Squalene Monooxygenase.

According to the World Health Organisation, cardiovascular complications have been recognized as the leading course of death between 2000 and 2019. Cardiovascular complications are caused by excess LDL cholesterol in the body or arteries that can build up to form a plaque. There are drugs currently in clinical use called statins that target HMGCoA reductase. However, these drugs result in several side effects. This work investigated using computational approaches to lower cholesterol by investigating green tea extracts as an inhibitors for squalene monooxygenase (the second-rate-controlling step in cholesterol synthesis). Pharmacophore modeling was done to identify possible pharmacophoric sites based on the pIC50 values. The best hypothesis generated by pharmacophore modeling was further validated by atom-based 3D QSAR, where 70% of the data set was treated as the training set. Prior molecular docking ADMET studies were done to investigate the physiochemical properties of these molecules. Glide docking was performed, followed by molecular dynamics to evaluate the protein conformational changes. Pharmacophore results suggest that the best molecules to interact with the biological target should have at least one hydrogen acceptor (A5), two hydrogen donors (D9 and D10), and two benzene rings (R14 and R15) for green tea polyphenols and theasinensin A. ADMET result shows that all molecules in this class have low oral adsorption. Molecular docking results showed that some green tea polyphenols have good binding affinities, with most of these structures having a docking score of less than -10 kcal/mol. Molecular dynamics further illustrated that the best-docked ligands perfectly stay within the active site over a 100 ns simulation. The results obtained from this study suggest that green tea polyphenols have the potential for inhibition of squalene monooxygenase, except for theasinensin A.

A Deep Learning-Based Framework for Predicting Intracerebral Hemorrhage Hematoma Expansion Using Head Non-contrast CT Scan

Rationale and ObjectivesHematoma expansion (HE) in intracerebral hemorrhage (ICH) is a critical factor affecting patient outcomes, yet effective clinical tools for predicting HE are currently lacking. We aim to develop a fully automated framework based on deep learning for predicting HE using only clinical non-contrast CT (NCCT) scans. Materials and MethodsA large retrospective dataset (n=2,484) was collected from 84 centers, while a prospective dataset (n=500) was obtained from 26 additional centers. Baseline NCCT scans and follow-up NCCT scans were conducted within 6 hours and 48 hours from symptom onset, respectively. HE was defined as a volume increase of more than 6 mL on the follow-up NCCT. The retrospective dataset was divided into a training set (n=1,876) and a validation set (n=608) by patient inclusion time. A two-stage framework was trained to predict HE, and its performance was evaluated on both the validation and prospective sets. Receiver operating characteristics area under the curve (AUC), sensitivity, and specificity were leveraged. ResultsOur two-stage framework achieved an AUC of 0.760 (95% CI 0.724-0.799) on the retrospective validation set and 0.806 (95% CI 0.750-0.859) on the prospective set, outperforming the commonly used BAT score, which had AUCs of 0.582 and 0.699, respectively. ConclusionOur framework can automatically and robustly identify ICH patients at high risk of HE using admission head NCCT scans, providing more accurate predictions than the BAT score.

MultiModRLBP: A Deep Learning Approach for Multi-Modal RNA-Small Molecule Ligand Binding Sites Prediction.

This study aims to tackle the intricate challenge of predicting RNA-small molecule binding sites to explore the potential value in the field of RNA drug targets. To address this challenge, we propose the MultiModRLBP method, which integrates multi-modal features using deep learning algorithms. These features include 3D structural properties at the nucleotide base level of the RNA molecule, relational graphs based on overall RNA structure, and rich RNA semantic information. In our investigation, we gathered 851 interactions between RNA and small molecule ligand from the RNAglib dataset and RLBind training set. Unlike conventional training sets, this collection broadened its scope by including RNA complexes that have the same RNA sequence but change their respective binding sites due to structural differences or the presence of different ligands. This enhancement enables the MultiModRLBP model to more accurately capture subtle changes at the structural level, ultimately improving its ability to discern nuances among similar RNA conformations. Furthermore, we evaluated MultiModRLBP on two classic test sets, Test18 and Test3, highlighting its performance disparities on small molecules based on metal and non-metal ions. Additionally, we conducted a structural sensitivity analysis on specific complex categories, considering RNA instances with varying degrees of structural changes and whether they share the same ligands. The research results indicate that MultiModRLBP outperforms the current state-of-the-art methods on multiple classic test sets, particularly excelling in predicting binding sites for non-metal ions and instances where the binding sites are widely distributed along the sequence. MultiModRLBP also can be used as a potential tool when the RNA structure is perturbed or the RNA experimental tertiary structure is not available. Most importantly, MultiModRLBP exhibits the capability to distinguish binding characteristics of RNA that are structurally diverse yet exhibit sequence similarity. These advancements hold promise in reducing the costs associated with the development of RNA-targeted drugs.

Identification of histopathological classification and establishment of prognostic indicators of gastric adenocarcinoma based on deep learning algorithm.

The aim of this study is to establish a deep learning (DL) model to predict the pathological type of gastric adenocarcinoma cancer based on whole-slide images(WSIs). We downloaded 356 histopathological images of gastric adenocarcinoma (STAD) patients from The Cancer Genome Atlas database and randomly divided them into the training set, validation set and test set (8:1:1). Additionally, 80 H&E-stained WSIs of STAD were collected for external validation. The CLAM tool was used to cut the WSIs and further construct the model by DL algorithm, achieving an accuracy of over 90% in identifying and predicting histopathological subtypes. External validation results demonstrated the model had a certain generalization ability. Moreover, DL features were extracted from the model to further investigate the differences in immune infiltration and patient prognosis between the two subtypes. The DL model can accurately predict the pathological classification of STAD patients, and provide certain reference value for clinical diagnosis. The nomogram combining DL-signature, gene-signature and clinical features can be used as a prognostic classifier for clinical decision-making and treatment.

Preoperative prediction of microvascular invasion risk in hepatocellular carcinoma with MRI: peritumoral versus tumor region

ObjectivesTo explore the predictive performance of tumor and multiple peritumoral regions on dynamic contrast-enhanced magnetic resonance imaging (MRI), to identify optimal regions of interest for developing a preoperative predictive model for the grade of microvascular invasion (MVI).MethodsA total of 147 patients who were surgically diagnosed with hepatocellular carcinoma, and had a maximum tumor diameter ≤ 5 cm were recruited and subsequently divided into a training set (n = 117) and a testing set (n = 30) based on the date of surgery. We utilized a pre-trained AlexNet to extract deep learning features from seven different regions of the maximum transverse cross-section of tumors in various MRI sequence images. Subsequently, an extreme gradient boosting (XGBoost) classifier was employed to construct the MVI grade prediction model, with evaluation based on the area under the curve (AUC).ResultsThe XGBoost classifier trained with data from the 20-mm peritumoral region showed superior AUC compared to the tumor region alone. AUC values consistently increased when utilizing data from 5-mm, 10-mm, and 20-mm peritumoral regions. Combining arterial and delayed-phase data yielded the highest predictive performance, with micro- and macro-average AUCs of 0.78 and 0.74, respectively. Integration of clinical data further improved AUCs values to 0.83 and 0.80.ConclusionCompared with those of the tumor region, the deep learning features of the peritumoral region provide more important information for predicting the grade of MVI. Combining the tumor region and the 20-mm peritumoral region resulted in a relatively ideal and accurate region within which the grade of MVI can be predicted.Clinical relevance statementThe 20-mm peritumoral region holds more significance than the tumor region in predicting MVI grade. Deep learning features can indirectly predict MVI by extracting information from the tumor region and directly capturing MVI information from the peritumoral region.Key PointsWe investigated tumor and different peritumoral regions, as well as their fusion.MVI predominantly occurs in the peritumoral region, a superior predictor compared to the tumor region.The peritumoral 20 mm region is reasonable for accurately predicting the three-grade MVI.Graphical

Development and Validation of a CT-Radiomics Nomogram for the Diagnosis of Small Prevascular Mediastinal Nodules: Reducing Nontherapeutic Surgeries

Rationale and ObjectivesThe preoperative diagnosis of small prevascular mediastinal nodules (SPMNs) presents a challenge, often leading to unnecessary surgical interventions. Our objective was to develop a nomogram based on preoperative CT radiomics features, serving as a non-invasive diagnostic tool for SPMNs. Materials and MethodsPatients with surgically resected SPMNs from two medical centers between January 2018 and December 2022 were retrospectively reviewed. Radiomics features were extracted and screened from preoperative CT images. Logistic regression was employed to establish clinical, radiomics, and hybrid models for differentiating thymic epithelial tumors (TETs) from cysts. The performance of these models was validated in both internal and external test sets by area under the receiver operating characteristic curve (AUC), while also comparing their diagnostic capability with human experts. ResultsThe study enrolled a total of 363 patients (median age, 53 years [IQR:45-59 years]; 175 [48.2%] males) for model development and validation, including 136 TETs and 227 cysts. Lesions' enhancement status, shape, calcification, and rad-score were identified as independent factors for distinction. The hybrid model demonstrated superior diagnostic performance compared to other models and human experts, with an AUC of 0.95 (95% CI:0.92-0.98), 0.94 (95% CI:0.89-0.99), and 0.93 (95% CI:0.83-1.00) in the training set, internal test set, and external test set respectively. The calibration curve of the model demonstrated excellent fit, while decision curve analysis underscored its clinical value. ConclusionThe radiomics-based nomogram effectively discriminates between the most prevalent types of SPMNs, namely TETs and cysts, thus presenting a promising tool for treatment guidance.

A deep learning approach based on domain generalization for mooring tension prediction in floating structures

Generalization remains a significant challenge in the application of deep learning methods, particularly in predicting mooring line tension for moored floating structures. The dynamic and complex nature of marine environments makes accurate prediction of mooring line tension a formidable task. Existing deep learning models, constrained by the data distribution of their training sets, typically perform well within the scope of these datasets. However, their limited generalization capability becomes apparent when confronted with unknown and changing marine conditions, resulting in a decline in prediction accuracy. To address this issue, this paper proposes a Domain Generalization (DG) model approach, which incorporates a distribution matching regularization term to the model's loss function, aimed at reducing distributional disparities between data from different domains. The proposed method was compared with existing Deep Neural Network (DNN) model method through validation with engineering case studies. The results indicate that the DG model method significantly surpasses the DNN model in terms of generalization capabilities. The mooring line tension prediction model developed herein holds potential for constructing digital twins, offering near real-time data support for moored floating engineering projects. This facilitates timely and accurate decision-making in response to environmental changes.

A Six-gene Prognostic Model Based on Neutrophil Extracellular Traps (NETs)-related Gene Signature for Lung Adenocarcinoma.

Lung adenocarcinoma (LUAD) is one of the most common malignant cancers. Neutrophil extracellular traps (NETs) have been discovered to play a crucial role in the pathogenesis of LUAD. We aimed to establish an innovative prognostic model for LUAD based on the distinct expression patterns of NETs-related genes. The TCGA LUAD dataset was utilized as the training set, while GSE31210, GSE37745, and GSE50081 were undertaken as the verification sets. The patients were grouped into clusters based on the expression signature of NETs-related genes. Differentially expressed genes between clusters were identified through the utilization of the random forest and LASSO algorithms. The NETs score model for LUAD prognosis was developed by multiplying the expression levels of specific genes with their corresponding LASSO coefficients and then summing them. The validity of the model was confirmed by analysis of the survival curves and ROC curves. Additionally, immune infiltration, GSEA, mutation analysis, and drug analysis were conducted. Silencing ABCC2 in A549 cells was achieved to investigate its effect. We identified six novel NETs-related genes, namely UPK1B, SFTA3, GGTLC1, SCGB3A1, ABCC2, and NTS, and developed a NETs score signature, which exhibited a significant correlation with the clinicopathological and immune traits of the LUAD patients. High-risk patients showed inhibition of immune-related processes. Mutation patterns exhibited variability among the different groups. AZD3759, lapatinib, and dasatinib have been identified as potential candidates for LUAD treatment. Moreover, the downregulation of ABCC2 resulted in the induction of apoptosis and suppression of migration and invasion in A549 cells. Altogether, this study has identified a novel NET-score signature based on six novel NET-related genes to predict the prognosis of LUAD and ABCC2 and has also explored a new method for personalized chemo-/immuno-therapy of LUAD.

Ensemble Algorithm for Risk Prediction of Clinical Failure After Anterior Cruciate Ligament Reconstruction.

Patient-specific risk profiles of clinical failure after anterior cruciate ligament reconstruction (ACLR) are meaningful for preoperative surgical planning and postoperative rehabilitation guidance. To create an ensemble algorithm machine learning (ML) model and ML-based web-based tool that can predict the patient-specific risk of clinical failure after ACLR. Cohort study; Level of evidence, 3. Included were 432 patients (mean age, 26.8 ± 8.4 years; 74.1% male) who underwent anatomic double-bundle ACLR with hamstring tendon autograft between January 2010 and February 2019. The primary outcome was the probability of clinical failure at a minimum 2-year follow-up. The authors included 24 independent variables for feature selection and model development. The data set was split randomly into training sets (75%) and test sets (25%). Models were built using 4 ML algorithms: extreme gradient boosting, random forest, light gradient boosting machine, and adaptive boosting. In addition, a weighted-average voting (WAV) ensemble model was constructed using the ensemble-voting technique to predict clinical failure after ACLR. Concordance (area under the receiver operating characteristic curve [AUC]), calibration, and decision curve analysis were used to evaluate predictive performances of the 5 models. Clinical failure occurred in 73 of the 432 patients (16.9%). The 8 most important predictors for clinical failure were follow-up period, high-grade preoperative knee laxity, time from injury to ACLR, participation in competitive sports, posterior tibial slope, graft diameter, age at surgery, and medial meniscus resection. The WAV ensemble algorithm achieved the best predictive performance based on concordance (AUC, 0.9139), calibration (calibration intercept, -0.1806; calibration slope, 1.2794; Brier score, 0.0888), and decision curve analysis (greatest net benefits) and was used to develop an web-based application to predict a patient's clinical failure risk of ACLR. The WAV ensemble algorithm was able to accurately predict patient-specific risk of clinical failure after ACLR. Clinicians and patients can use the web-based application during preoperative consultation to understand individual prediction outcomes.

Prediction of glass transition temperature of oxide glasses based on interpretable machine learning and sparse data sets

The glass transition temperature (Tg) is a crucial characteristic of oxide glasses, exerting significant influence on their properties and applications. In this study, we utilized a sparse dataset and machine learning techniques to establish a predictive model for the relationship between the composition of oxide glasses and Tg. Among four machine learning algorithms compared, the Extreme Tree Regressor (ETR) algorithm demonstrated superior performance, exhibiting robust generalization capabilities on the validation set. We performed feature selection on the original features using Pearson Correlation Coefficient (PCC) and Maximal Information Coefficient (MIC), resulting in a subset of 38 dimensions. Subsequently, we employed Sparse Principal Component Analysis (PCA) and Recursive Feature Elimination (RFE) for dimensionality reduction, yielding a final subset of 12 dimensions. Introducing SHapley Additive exPlanations (SHAP) values for interpretability analysis of the predictive model, we obtained important feature rankings and analyzed how variations in features affect the target variable. Finally, using the CaO-SiO2-Al2O3 system as an example, we demonstrated how the model predicts Tg based on composition, facilitating rational design of oxide glass compositions.