Seismic fragility analysis of horizontally curved bridge portfolios using a machine-learning-based partial dependence plot approach

A Data-Driven Seismic Fragility Model for Post-Earthquake Repairable Performance of Highway Curved Bridge Portfolios

The ecological and environmental science communities have embraced machine learning (ML) for empirical modelling and prediction. However, going beyond prediction to draw insights into underlying functional relationships between response variables and environmental ‘drivers’ is less straightforward. Deriving ecological insights from fitted ML models requires techniques to extract the ‘learning’ hidden in the ML models.We revisit the theoretical background and effectiveness of four approaches for deriving insights from ML: ranking independent variable importance (Gini importance, GI; permutation importance, PI; split importance, SI; and conditional permutation importance, CPI), and two approaches for inference of bivariate functional relationships (partial dependence plots, PDP; and accumulated local effect plots, ALE). We also explore the use of a surrogate model for visualization and interpretation of complex multi‐variate relationships between response variables and environmental drivers. We examine the challenges and opportunities for extracting ecological insights with these interpretation approaches. Specifically, we aim to improve interpretation of ML models by investigating how effectiveness relates to (a) interpretation algorithm, (b) sample size and (c) the presence of spurious explanatory variables.We base the analysis on simulations with known underlying functional relationships between response and predictor variables, with added white noise and the presence of correlated but non‐influential variables. The results indicate that deriving ecological insight is strongly affected by interpretation algorithm and spurious variables, and moderately impacted by sample size. Removing spurious variables improves interpretation of ML models. Meanwhile, increasing sample size has limited value in the presence of spurious variables, but increasing sample size does improves performance once spurious variables are omitted. Among the four ranking methods, SI is slightly more effective than the other methods in the presence of spurious variables, while GI and SI yield higher accuracy when spurious variables are removed. PDP is more effective in retrieving underlying functional relationships than ALE, but its reliability declines sharply in the presence of spurious variables. Visualization and interpretation of the interactive effects of predictors and the response variable can be enhanced using surrogate models, including three‐dimensional visualizations and use of loess planes to represent independent variable effects and interactions.Machine learning analysts should be aware that including correlated independent variables in ML models with no clear causal relationship to response variables can interfere with ecological inference. When ecological inference is important, ML models should be constructed with independent variables that have clear causal effects on response variables. While interpreting ML models for ecological inference remains challenging, we show that careful choice of interpretation methods, exclusion of spurious variables and adequate sample size can provide more and better opportunities to ‘learn from machine learning’.

To evaluate the long-term prognostic value of coronary CT angiography (cCTA)-derived plaque measures and clinical parameters on major adverse cardiac events (MACE) using machine learning (ML). Datasets of 361 patients (61.9 ± 10.3years, 65% male) with suspected coronary artery disease (CAD) who underwent cCTA were retrospectively analyzed. MACE was recorded. cCTA-derived adverse plaque features and conventional CT risk scores together with cardiovascular risk factors were provided to a ML model to predict MACE. A boosted ensemble algorithm (RUSBoost) utilizing decision trees as weak learners with repeated nested cross-validation to train and validate the model was used. Performance of the ML model was calculated using the area under the curve (AUC). MACE was observed in 31 patients (8.6%) after a median follow-up of 5.4years. Discriminatory power was significantly higher for the ML model (AUC 0.96 [95%CI 0.93-0.98]) compared with conventional CT risk scores including Agatston calcium score (AUC 0.84 [95%CI 0.80-0.87]), segment involvement score (AUC 0.88 [95%CI 0.84-0.91]), and segment stenosis score (AUC 0.89 [95%CI 0.86-0.92], all p < 0.05). Similar results were shown for adverse plaque measures (AUCs 0.72-0.82, all p < 0.05) and clinical parameters including the Framingham risk score (AUCs 0.71-0.76, all p < 0.05). The ML model yielded significantly higher diagnostic performance compared with logistic regression analysis (AUC 0.96 vs. 0.92, p = 0.024). Integration of a ML model improves the long-term prediction of MACE when compared with conventional CT risk scores, adverse plaque measures, and clinical information. ML algorithms may improve the integration of patient's information to enhance risk stratification. • A machine learning (ML) model portends high discriminatory power to predict major adverse cardiac events (MACE). • ML-based risk stratification shows superior diagnostic performance for MACE prediction over coronary CT angiography (cCTA)-derived risk scores or clinical parameters alone. • A ML model outperforms conventional logistic regression analysis for the prediction of MACE.

The unconfined compressive strength (Qu) is one of the most important criteria of stabilized soil to design in order to evaluate the effective of soft soil improvement. The unconfined compressive strength of stabilized soil is strongly affected by numerous factors such as the soil properties, the binder content, etc. Machine Learning (ML) approach can take into account these factors to predict the unconfined compressive strength (Qu) with high performance and reliability. The aim of this paper is to select a single ML model to design Qu of stabilized soil containing some chemical stabilizer agents such as lime, cement and bitumen. In order to build the single ML model, a database is created based on the literature investigation. The database contains 200 data samples, 12 input variables (Liquid limit, Plastic limit, Plasticity index, Linear shrinkage, Clay content, Sand content, Gravel content, Optimum water content, Density of stabilized soil, Lime content, Cement content, Bitumen content) and the output variable Qu. The performance and reliability of ML model are evaluated by the popular validation technique Monte Carlo simulation with aided of three criteria metrics including coefficient of determination R2, Root Mean Square Error (RMSE) and Mean Square Error (MAE). ML model based on Gradient Boosting algorithm is selected as highest performance and highest reliability ML model for designing Qu of stabilized soil. Explanation of feature effects on the unconfined compressive strength Qu of stabilized soil is carried out by Permutation importance, Partial Dependence Plot (PDP 2D) in two dimensions and SHapley Additive exPlanations (SHAP) local value. The ML model proposed in this investigation is single and useful for professional engineers with using the mapping Maximal dry density-Linear shrinkage created by PDP 2D.

To develop and externally validate a multiphase computed tomography (CT)-based machine learning (ML) model for staging liver fibrosis (LF) by using whole liver slices. The development dataset comprised 232 patients with pathological analysis for LF, and the test dataset comprised 100 patients from an independent outside institution. Feature extraction was performed based on the precontrast (PCP), arterial (AP), portal vein (PVP) phase, and three-phase CT images. CatBoost was utilized for ML model investigation by using the features with good reproducibility. The diagnostic performance of ML models based on each single- and three-phase CT image was compared with that of radiologists' interpretations, the aminotransferase-to-platelet ratio index, and the fibrosis index based on four factors (FIB-4) by using the receiver operating characteristic curve with the area under the curve (AUC) value. Although the ML model based on three-phase CT image (AUC = 0.65-0.80) achieved higher AUC value than that based on PCP (AUC = 0.56-0.69) and PVP (AUC = 0.51-0.74) in predicting various stage of LF, significant difference was not found. The best CT-based ML model (AUC = 0.65-0.80) outperformed the FIB-4 in differentiating advanced LF and cirrhosis and radiologists' interpretation (AUC = 0.50-0.76) in the diagnosis of significant and advanced LF. All PCP, PVP, and three-phase CT-based ML models can be an acceptable in assessing LF, and the performance of the PCP-based ML model is comparable to that of the enhanced CT image-based ML model.

Systematic Review of Machine Learning Models for Myelodysplastic Syndrome Diagnosis

BackgroundAccurate prediction of survival in oncology can guide targeted interventions. The traditional regression-based Cox proportional hazards (CPH) model has statistical assumptions and may have limited predictive accuracy. With the capability to model large datasets, machine learning (ML) holds the potential to improve the prediction of time-to-event outcomes, such as cancer survival outcomes. The present study aimed to systematically summarize the use of ML models for cancer survival outcomes in observational studies and to compare the performance of ML models with CPH models.MethodsWe systematically searched PubMed, MEDLINE (via EBSCO), and Embase for studies that evaluated ML models vs. CPH models for cancer survival outcomes. The use of ML algorithms was summarized, and either the area under the curve (AUC) or the concordance index (C-index) for the ML and CPH models were presented descriptively. Only studies that provided a measure of discrimination, i.e., AUC or C-index, and 95% confidence interval (CI) were included in the final meta-analysis. A random-effects model was used to compare the predictive performance in the pooled AUC or C-index estimates between ML and CPH models using R. The quality of the studies was evaluated using available checklists. Multiple sensitivity analyses were performed.ResultsA total of 21 studies were included for systematic review and 7 for meta-analysis. Across the 21 articles, diverse ML models were used, including random survival forest (N=16, 76.19%), gradient boosting (N=5, 23.81%), and deep learning (N=8, 38.09%). In predicting cancer survival outcomes, ML models showed no superior performance over CPH regression. The standardized mean difference in AUC or C-index was 0.01 (95% CI: -0.01 to 0.03). Results from the sensitivity analyses confirmed the robustness of the main findings.ConclusionsML models had similar performance compared with CPH models in predicting cancer survival outcomes. Although this systematic review highlights the promising use of ML to improve the quality of care in oncology, findings from this review also suggest opportunities to improve ML reporting transparency. Future systematic reviews should focus on the comparative performance between specific ML models and CPH regression in time-to-event outcomes in specific type of cancer or other disease areas.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12874-025-02694-z.

BackgroundThe goal of this study was to assess the effectiveness of machine learning models and create an interpretable machine learning model that adequately explained 3-year all-cause mortality in patients with chronic heart failure.MethodsThe data in this paper were selected from patients with chronic heart failure who were hospitalized at the First Affiliated Hospital of Kunming Medical University, from 2017 to 2019 with cardiac function class III-IV. The dataset was explored using six different machine learning models, including logistic regression, naive Bayes, random forest classifier, extreme gradient boost, K-nearest neighbor, and decision tree. Finally, interpretable methods based on machine learning, such as SHAP value, permutation importance, and partial dependence plots, were used to estimate the 3-year all-cause mortality risk and produce individual interpretations of the model's conclusions.ResultIn this paper, random forest was identified as the optimal aools lgorithm for this dataset. We also incorporated relevant machine learning interpretable tand techniques to improve disease prognosis, including permutation importance, PDP plots and SHAP values for analysis. From this study, we can see that the number of hospitalizations, age, glomerular filtration rate, BNP, NYHA cardiac function classification, lymphocyte absolute value, serum albumin, hemoglobin, total cholesterol, pulmonary artery systolic pressure and so on were important for providing an optimal risk assessment and were important predictive factors of chronic heart failure.ConclusionThe machine learning-based cardiovascular risk models could be used to accurately assess and stratify the 3-year risk of all-cause mortality among CHF patients. Machine learning in combination with permutation importance, PDP plots, and the SHAP value could offer a clear explanation of individual risk prediction and give doctors an intuitive knowledge of the functions of important model components.

Prediction of post-stroke functional outcomes is crucial for allocating medical resources. In this study, a total of 577 patients were enrolled in the Post-Acute Care-Cerebrovascular Disease (PAC-CVD) program, and 77 predictors were collected at admission. The outcome was whether a patient could achieve a Barthel Index (BI) score of >60 upon discharge. Eight machine-learning (ML) methods were applied, and their results were integrated by stacking method. The area under the curve (AUC) of the eight ML models ranged from 0.83 to 0.887, with random forest, stacking, logistic regression, and support vector machine demonstrating superior performance. The feature importance analysis indicated that the initial Berg Balance Test (BBS-I), initial BI (BI-I), and initial Concise Chinese Aphasia Test (CCAT-I) were the top three predictors of BI scores at discharge. The partial dependence plot (PDP) and individual conditional expectation (ICE) plot indicated that the predictors’ ability to predict outcomes was the most pronounced within a specific value range (e.g., BBS-I < 40 and BI-I < 60). BI at discharge could be predicted by information collected at admission with the aid of various ML models, and the PDP and ICE plots indicated that the predictors could predict outcomes at a certain value range.

Unveiling the effect of urinary xenoestrogens on chronic kidney disease in adults: A machine learning model.

As more machine learning and deep learning models are applied in studying the quantitative relationship between the climate and terrestrial vegetation growth, the uncertainty of these advanced models requires clarification. Partial dependence plots (PDPs) are one of the most widely used methods to estimate the marginal effect of independent variables on the predicted outcome of a machine learning model, and it is regarded as the main basis for conclusions in relevant research. As more controversies regarding the reliability of the results of the PDPs emerge, the uncertainty of the PDPs remains unclear. In this paper, we experiment with real, remote sensing data to systematically analyze the uncertainty of partial dependence relationships between four climate variables (temperature, rainfall, radiation, and windspeed) and vegetation growth, with one conventional linear model and six machine learning models. We tested the uncertainty of the PDP curves across different machine learning models from three aspects: variation, whole linear trends, and the trait of change points. Results show that the PDP of the dominant climate factor (mean air temperature) and vegetation growth parameter (indicated by the normalized difference vegetation index, NDVI) has the smallest relative variation and the whole linear trend of the PDP was comparatively stable across the different models. The mean relative variation of change points across the partial dependence curves of the non-dominant climate factors (i.e., radiation, windspeed, and rainfall) and vegetation growth ranged from 8.96% to 23.8%, respectively, which was much higher than those of the dominant climate factor and vegetation growth. Lastly, the model used for creating the PDP, rather than the relative importance of these climate factors, determines the fluctuation of the PDP output of these climate variables and vegetation growth. These findings have significant implications for using remote sensing data and machine learning models to investigate the quantitative relationships between the climate and terrestrial vegetation.

To evaluate the predictive performance of the triple-D and quadruple-D scores integrated with machine learning (ML) models in determining stone-free outcomes after extracorporeal shock wave lithotripsy (ESWL), and to compare ML model performance and identify its key predictors influencing ESWL success. An observational study. Place and Duration of the Study: Department of Urology, Gaziosmanpasa Training and Research Hospital, Istanbul, Turkiye, from October 2020 to November 2024. A total of 309 patients who underwent ESWL were analysed. The patients were categorised into stone-free and non-stone- free groups based on post-treatment imaging. Clinical parameters, including quadruple-D score (stone volume, density, skin-to-stone distance [SSD], and location), were recorded. Three ML models‒random forest (RF), logistic regression (LR), and neural network (NN)‒were trained on 80% of the dataset and tested on 20%. Model performance was assessed using accuracy, area under the curve (AUC), precision, recall, and F1 score. The quadruple-D score (AUC: 0.724) demonstrated superior predictive power compared to the Triple-D score (AUC: 0.700). Among ML models, RF achieved the highest accuracy (82.9%, AUC: 0.91), followed by NN (80.9%, AUC: 0.87) and LR (79.6%, AUC: 0.85). Significant predictors of ESWL success were stone density, volume, SSD, and the quadruple-D score, while age and body mass index (BMI) were not significant. Integrating the quadruple-D score with ML models, particularly RF, enhances the prediction of ESWL outcomes. Combining clinical expertise with computational intelligence can refine patient selection and optimise treatment strategies. However, prospective studies are needed to validate these findings. Extracorporeal shock wave lithotripsy, Quadruple-D score, Machine learning, Random forest, Stone-free prediction.

To explore the value of machine learning (ML) models based on contrast-enhanced cone-beam breast computed tomography (CE-CBBCT) radiomics features for the preoperative prediction of human epidermal growth factor receptor 2 (HER2)-low expression breast cancer (BC). Fifty-six patients with HER2-negative invasive BC who underwent preoperative CE-CBBCT were prospectively analyzed. Patients were randomly divided into training and validation cohorts at approximately 7:3. A total of 1046 quantitative radiomic features were extracted from CE-CBBCT images and normalized using z-scores. The Pearson correlation coefficient and recursive feature elimination were used to identify the optimal features. Six ML models were constructed based on the selected features: linear discriminant analysis (LDA), random forest (RF), support vector machine (SVM), logistic regression (LR), AdaBoost (AB), and decision tree (DT). To evaluate the performance of these models, receiver operating characteristic curves and area under the curve (AUC) were used. Seven features were selected as the optimal features for constructing the ML models. In the training cohort, the AUC values for SVM, LDA, RF, LR, AB, and DT were 0.984, 0.981, 1.000, 0.970, 1.000, and 1.000, respectively. In the validation cohort, the AUC values for the SVM, LDA, RF, LR, AB, and DT were 0.859, 0.880, 0.781, 0.880, 0.750, and 0.713, respectively. Among all ML models, the LDA and LR models demonstrated the best performance. The DeLong test showed that there were no significant differences among the receiver operating characteristic curves in all ML models in the training cohort (P > .05); however, in the validation cohort, the DeLong test showed that the differences between the AUCs of LDA and RF, AB, and DT were statistically significant (P = .037, .003, .046). The AUCs of LR and RF, AB, and DT were statistically significant (P = .023, .005, .030). Nevertheless, no statistically significant differences were observed when compared to the other ML models. ML models based on CE-CBBCT radiomics features achieved excellent performance in the preoperative prediction of HER2-low BC and could potentially serve as an effective tool to assist in precise and personalized targeted therapy.

Development and Internal Validation of Machine Learning Algorithms for Predicting Subsequent Contralateral Slipped Capital Femoral Epiphysis in Patients With Unilateral Slips.

Multiple studies have attempted to generate visual field (VF) mean deviation (MD) estimates using cross-sectional optical coherence tomography (OCT) data. However, whether such models offer any value in detecting longitudinal VF progression is unclear. We address this by developing a machine learning (ML) model to convert OCT data to MD and assessing its ability to detect longitudinal worsening. Retrospective, longitudinal study. A model dataset of 70,575 paired OCT/VFs to train an ML model converting OCT to VF-MD. A separate progression dataset of 4,044 eyes with ≥ 5 paired OCT/VFs to assess the ability of OCT-derived MD to detect worsening. Progression dataset eyes had two additional unpaired VFs (≥ 7 total) to establish a "ground truth" rate of progression defined by MD slope. We trained an ML model using paired VF/OCT data to estimate MD measurements for each OCT scan (OCT-MD). We used this ML model to generate longitudinal OCT-MD estimates for progression dataset eyes. We calculated MD slopes after substituting/supplementing VF-MD with OCT-MD and measured the ability to detect progression. We labeled true progressors using a ground truth MD slope <0.5 dB/year calculated from ≥ 7 VF-MD measurements. We compared the area under the curve (AUC) of MD slopes calculated using both VF-MD (with <7 measurements) and OCT-MD. Because we found OCT-MD substitution had a statistically inferior AUC to VF-MD, we simulated the effect of reducing OCT-MD mean absolute error (MAE) on the ability to detect worsening. AUC. OCT-MD estimates had an MAE of 1.62 dB. AUC of MD slopes with partial OCT-MD substitution was significantly worse than the VF-MD slope. Supplementing VF-MD with OCT-MD also did not improve AUC, regardless of MAE. OCT-MD estimates needed an MAE ≤ 1.00 dB before AUC was statistically similar to VF-MD alone. ML models converting OCT data to VF-MD with error levels lower than published in prior work (MAE: 1.62 dB) were inferior to VF-MD data for detecting trend-based VF progression. Models converting OCT data to VF-MD must achieve better prediction errors (MAE ≤ 1 dB) to be clinically valuable at detecting VF worsening.