Higher-Order Neural Additive Models: An Interpretable Machine Learning Model with Feature Interactions

ObjectivePostoperative complications remain a major concern in laryngeal cancer surgery, often requiring invasive interventions or intensive care. This study aimed to develop and validate an interpretable machine learning (ML) model to preoperatively predict Clavien-Dindo Grade ≥ III complications and support risk-informed perioperative decision-making.MethodsWe conducted a retrospective study using a temporally split cohort of laryngeal cancer patients. Postoperative complications were graded using the Clavien-Dindo (CD) classification. Eight ML algorithms were trained and evaluated using receiver operating characteristic (ROC) curves, calibration plots, and decision curve analysis (DCA). Model interpretability was assessed using SHapley Additive exPlanations (SHAP). A web-based calculator was deployed for clinical use.ResultThe random forest (RF) model achieved the best performance, with an area under the curve (AUC) of 0.935 in the training set and 0.842 in the test set. The model demonstrated robust sensitivity and specificity for both surgical and medical complications. Calibration curves indicated strong agreement between predicted and actual outcomes. SHAP analysis identified eight key predictors—such as vocal cord mobility, tumor subsite, and nutritional status—that contributed most to risk estimation. A user-friendly web calculator was developed and is accessible at: https://qilushiny.shinyapps.io/qilupredicate/.ConclusionWe developed a clinically interpretable ML model that accurately predicts major postoperative complications in patients undergoing laryngeal cancer surgery. This tool provides individualized risk assessments that can guide surgical planning, optimize perioperative strategies, and enhance shared decision-making. Prospective multicenter validation is needed to confirm its utility in routine practice.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12893-025-03213-z.

Interpretable vs. noninterpretable machine learning models for data-driven hydro-climatological process modeling

ObjectiveDistant organ metastasis remains the primary factor affecting long-term survival following radical surgery for colorectal cancer (CRC). This study aimed to develop and validate an interpretable machine learning (ML) model to predict the 5-year cumulative risk of distant metastasis after radical CRC surgery.MethodsA retrospective observational cohort study was conducted using clinical and follow-up data from 341 CRC patients who underwent radical surgery. The cohort was randomly divided into a training set n=239 and a validation set n=102 at a 7:3 ratio. Feature selection was performed using least absolute shrinkage and selection operator (LASSO) regression, identifying variables associated with the 5-year cumulative occurrence of metastasis. Prediction models were constructed using seven algorithms. Model performance was evaluated through multiple metrics: area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, F1 score, calibration plots, and decision curve analysis. The SHapley Additive exPlanations (SHAP) method was applied to improve model interpretability.ResultsLASSO combined with tenfold cross-validation selected 11 key features for model development. Among the models tested, the SVM model demonstrated superior performance, achieving a Brier score of 0.144 and an AUC of 0.865 in the validation set. Calibration and clinical decision curves confirmed the SVM model’s strong calibration and clinical applicability. The SHAP dependence plots and force analysis provided explanations at both feature and individual patient levels for the model’s 5-year risk predictions.ConclusionThis study established a high-accuracy and interpretable ML model capable of effectively predicting the 5-year cumulative risk of distant organ metastasis after radical colorectal cancer surgery, while further external validation is necessary to confirm its clinical utility.

Pediatric adrenocortical tumors (pACTs) are rare and clinically heterogeneous. Existing risk stratification systems rely on fixed thresholds and linear assumptions, which may limit their prognostic accuracy-particularly for nonmetastatic, locally advanced cases. We aimed to develop an interpretable machine learning (ML) model for individualized survival prediction using only routine clinical features. We retrospectively analyzed 97 patients with pACT from the German Pediatric Oncology Hematology-Malignant Endocrine Tumors Registry (1997-2024). An Extreme Gradient Boosting Cox proportional hazards model was trained using 4 features-tumor volume, distant metastases, pathologic T stage, and resection status-identified via systematic feature evaluation across 11 737 model combinations. Performance was assessed using a stratified 80/20 train-test split, 500 bootstrap iterations, and Harrell's concordance index (C-index). SHapley Additive exPlanations (SHAP) were used for interpretability. The model achieved strong prognostic performance (test-set C-index: 0.925; bootstrap mean: 0.891, 95% confidence interval: 0.817-0.946). SHAP analysis confirmed the dominant influence of metastatic status, followed by tumor volume, T stage, and resection status. The model uncovered nonlinear and additive effects, including a SHAP- and bootstrap-guided tumor volume cut-off (190 mL, 95% confidence interval 127-910 mL) that only slightly differed from conventional thresholds. Stratification remained robust in subgroups, including nonmetastatic patients with advanced local disease. This interpretable ML model enables individualized survival prediction in pACT using only routine clinical data. It offers a clinically accessible and clinically meaningful complement to existing scoring systems, particularly in patients with ambiguous risk profiles who may benefit from more personalized management.

To develop an interpretable machine learning (ML) model for predicting surgical outcomes in renal hilar tumors and propose a hilar-specific anatomical nephrometry scoring system. A total of 414 patients with renal hilar tumors who underwent robot-assisted partial nephrectomy (RAPN) were included in this study, comprising 304 patients from the First Affiliated Hospital of Nanchang University and 110 patients from the Second Affiliated Hospital of Nanchang University, which served as the external validation cohort. To identify predictors of trifecta achievement, we used least absolute shrinkage and selection operator regression and the Boruta algorithm, followed by multivariate logistic regression to identify independent factors. Five ML models were developed and evaluated using receiver operating characteristic curves, calibration plots, decision curve analysis, and precision-recall curves. The generalizability of the model was further validated in external cohort. Finally, we used SHapley Additive exPlanations (SHAP) to interpret the contribution of each predictor and enhance the model's explainability. Furthermore, based on anatomical features identified through logistic regression, we developed a modified nephrometry scoring system and compared its risk stratification performance with the traditional R.E.N.A.L. (i.e., Radius, Exophytic or endophytic, Nearness, Anterior or posterior, and Location) scoring system. Among the 304 patients in the primary cohort, 168 achieved trifecta outcomes. Eight variables were incorporated into the predictive model, with logistic regression model ultimately being selected as the optimal predictive model. It showed robust predictive performance in internal and external validation. SHAP methods identified surgeon, classification of hilar tumor, and radius as the three most significant predictive variables. Compared with the traditional R.E.N.A.L. score, the modified R.E.N.A.L. score demonstrated superior stratification ability for operation time, change in serum creatinine, change in estimated glomerular filtration rate, and trifecta achievement. The interpretable ML model accurately predicts trifecta in RAPN for hilar tumors. The modified R.E.N.A.L. score provides refined anatomical stratification and facilitates individualized surgical planning.

Optimisation and interpretation of machine and deep learning models for improved water quality management in Lake Loktak

Ovarian cancer (OC) is a deadly disease that affects a large number of women worldwide. Machine Learning (ML) models can help in the early detection of this disease, however, the use of these models may be limited by their lack of interpretability and the difficulty to evaluate their performance. In this work, five types of datasets were used, employing clinical features, different types of coding genomic features, and combining both. The use of interpretable ML (IML) models (one linear and one nonlinear model) provided us with better interpretability of the five feature sets. Following this study, nine binary classification models were compared, and the Accuracy, Recall, and Area Under the Curve were analyzed. The results showed that ML models employed the combination of clinical features and genomes with the coding of the position of genes in patients significantly improves the prediction. We demonstrated that the inclusion of different preprocessed patient data and especially through the information provided by IML models, can help clinicians to understand the disease better and make informed treatment decisions.

Machine learning is a promising method to solve the stability assessment problems of modern complex power systems. The interpretability of machine learning models is fundamental for trusting the prediction and deploying the models on engineering application. In this paper, a model-agnostic interpretation method for machine learning power system stability assessment models is proposed based on Shapley value. Shapley values represent the contribution of each state variable to the stability evaluation result. By constructing kernel function and simplifying the form of training data, Shapley values are efficiently calculated based on the weighted linear regression. The resulting additive feature contribution model can be used to explain the internal logic of the prediction and verify the credibility of the model. The experiment results on voltage margin estimation model in the New England 39-bus system demonstrate the correctness and comprehensibility of the interpretations generated by the proposed method.

Annual burned areas in the United States have increased twofold during the past decades. With more large fires resulting in more emissions of fine particulate matter, an accurate prediction of fire emissions is critical for quantifying the impacts of fires on air quality, human health, and climate. This study aims to construct a machine learning (ML) model with game-theory interpretation to predict monthly fire emissions over the contiguous US and to understand the controlling factors of fire emissions. By comparing the predicted fire PM2.5 emissions from the interpretable ML model with the Global Fire Emissions Database (GFED) observations and predictions from process-based models in the Fire Modeling Intercomparison Project (FireMIP), the ML model is also used to diagnose the process-based models to inform future development. Results show promising performance for the ML model, Community Land Model (CLM), and Joint UK Land Environment Simulator-Interactive Fire And Emission Algorithm For Natural Environments (JULES-INFERNO) in reproducing the spatial distributions, seasonality, and interannual variability of fire emissions over CONUS. Regional analysis shows that only the ML model and CLM simulate the realistic interannual variability of fire emissions for most of the subregions (r > 0.95 for ML and r = 0.14 ~ 0.70 for CLM), except for Mediterranean California, where all the models perform poorly (r = 0.74 for ML and r < 0.30 for the FireMIP models). Regarding seasonality, most models capture the peak emission in July over western US. However, all models except for the ML model fail to reproduce the bimodal peaks in July and October over Mediterranean California, which may be explained by the coarse spatial resolutions of the processed-based models or atmospheric forcing data or limitations in model parameterizations for capturing the effects of Santa Ana winds on fire activity. Furthermore, most models struggle to capture the spring peak in emissions in southeastern US, probably due to underrepresentation of human effects and the influences of winter dryness on fires in the models. As for extreme events, both the ML model and CLM successfully reproduce the frequency map of extreme emission occurrence but overestimate the number of months with extremely large fire emissions. Comparing the fire PM2.5 emissions from the interpretable ML model with process-based fire models highlights their strengths and uncertainties for regional analysis and prediction and provides useful insights on future directions for model improvements.

Developing Machine Learning (ML) algorithms for heterogeneous/mixed data is a longstanding problem. Many ML algorithms are not applicable to mixed data, which include numeric and non-numeric data, text, graphs and so on to generate interpretable models. Another longstanding problem is developing algorithms for lossless visualization of multidimensional mixed data. The further progress in ML heavily depends on success interpretable ML algorithms for mixed data and lossless interpretable visualization of multidimensional data. The later allows developing interpretable ML models using visual knowledge discovery by end-users, who can bring valuable domain knowledge which is absent in the training data. The challenges for mixed data include: (1) generating numeric coding schemes for non-numeric attributes for numeric ML algorithms to provide accurate and interpretable ML models, (2) generating methods for lossless visualization of n-D non-numeric data and visual rule discovery in these visualizations. This paper presents a classification of mixed data types, analyzes their importance for ML and present the developed experimental toolkit to deal with mixed data. It combines the Data Types Editor, VisCanvas data visualization and rule discovery system which is available on GitHub.

PurposeDiabetes mellitus causes various problems in our life. With the big data boom in our society, some risk factors for Diabetes must still exist. To identify new risk factors for diabetes in the big data society and explore further efficient use of big data, the non-objective-oriented census data about the Japanese Citizen’s Survey of Living Conditions were analyzed using interpretable machine learning methods.MethodsSeven interpretable machine learning methods were used to analysis Japan citizens’ census data. Firstly, logistic analysis was used to analyze the risk factors of diabetes from 19 selected initial elements. Then, the linear analysis, linear discriminate analysis, Hayashi’s quantification analysis method 2, random forest, XGBoost, and SHAP methods were used to re-check and find the different factor contributions. Finally, the relationship among the factors was analyzed to understand the relationship among factors.ResultsFour new risk factors: the number of family members, insurance type, public pension type, and health awareness level, were found as risk factors for diabetes mellitus for the first time, while another 11 risk factors were reconfirmed in this analysis. Especially the insurance type factor and health awareness level factor make more contributions to diabetes than factors: hypertension, hyperlipidemia, and stress in some interpretable models. We also found that work years were identified as a risk factor for diabetes because it has a high coefficient with the risk factor of age.ConclusionsNew risk factors for diabetes mellitus were identified based on Japan's non-objective-oriented anonymous census data using interpretable machine learning models. The newly identified risk factors inspire new possible policies for preventing diabetes. Moreover, our analysis certifies that big data can help us find helpful knowledge in today's prosperous society. Our study also paves the way for identifying more risk factors and promoting the efficiency of using big data.

Development and external validation of interpretable machine learning models for personalized multiple treatment recommendations in non-small cell lung cancer.

Machine learning (ML) models have recently shown important advantages in predicting nanomaterial properties, which avoids many trial-and-error explorations. However, complex variables that control the formation of nanomaterials exhibiting the desired properties still need to be better understood owing to the low interpretability of ML models and the lack of detailed mechanism information on nanomaterial properties. In this study, we developed a methodology for accurately predicting multiple synthesis parameter-property relationships of nanomaterials to improve the interpretability of the nanomaterial property mechanism. As a proof-of-concept, we designed glutathione-gold nanoclusters (GSH-AuNCs) exhibiting an appropriate fluorescence quantum yield (QY). First, we conducted 189 experiments and synthesized different GSH-AuNCs by varying the thiol-to-metal molar ratio and reaction temperature and time in reasonable ranges. The fluorescence QY of GSH-AuNCs could be systematically and independently programmed using different experimental parameters. We used limited GSH-AuNC synthesis parameter data to train an extreme gradient boosting regressor model. Moreover, we improved the interpretability of the ML model by combining individual conditional expectation, double-variable partial dependence, and feature interaction network analyses. The interpretability analyses established the relationship between multiple synthesis parameters and fluorescence QYs of GSH-AuNCs. The results represent an essential step towards revealing the complex fluorescence mechanism of thiolated AuNCs. Finally, we constructed a synthesis phase diagram exceeding 6.0 × 104 prediction variables for accurately predicting the fluorescence QY of GSH-AuNCs. A multidimensional synthesis phase diagram was obtained for the fluorescence QY of GSH-AuNCs by searching the synthesis parameter space in the trained ML model. Our methodology is a general and powerful complementary strategy for application in material informatics.

Repeat transurethral resection (ReTUR) is essential for reducing residual and recurrent non-muscle-invasive bladder cancer (NMIBC). Pathological upstaging after ReTUR significantly influences prognosis. This study aimed to develop an interpretable machine learning model using CT radiomics to predict the risk of pathological upstaging following ReTUR in NMIBC patients. We retrospectively analyzed 104 NMIBC patients who underwent ReTUR at the Second Affiliated Hospital of Nanchang University from March 2019 to July 2022. Data were split 7:3 into training and internal validation sets. An external validation set included 40 patients from two other hospitals. Radiomic features were extracted from preoperative CT scans. Least Absolute Shrinkage and Selection Operator (LASSO) and multivariate logistic regression were used to identify predictors of pathological upstaging. Four machine learning models, including Extreme Gradient Boosting (XGBoost), Gradient Boosting Decision Tree (GBDT), Random Forest (RF), and Linear Discriminant Analysis (LDA), were constructed and evaluated using AUC, accuracy, precision, F1 score, calibration curves, and decision curve analysis (DCA). The best model was interpreted via SHapley Additive exPlanations (SHAP) to identify key predictive features. umor grade (OR = 7.02, 95% CI: 1.17-42.21), tumor size (OR = 5.83, 95% CI: 1.21-28.15), and tumor number (OR = 6.83, 95% CI: 1.18-39.52) were independent risk factors. From 4,738 radiomic features, nine were selected. The XGBoost model outperformed others, with an AUC of 0.804 (95% CI: 0.756-0.862), accuracy of 77.4%, precision of 82.7%, and F1 score of 0.701 in internal validation. External validation confirmed its robustness. SHAP analysis highlighted Wavelet_LLH_firstorder_Maximum.1, Gradient_ngtdm_Complexity, and tumor grade as top predictors. The model showed good calibration and clinical utility on DCA. An interpretable CT radiomics-based machine learning model integrating clinical and imaging features was developed to accurately predict pathological upstaging risk after ReTUR in NMIBC patients. This tool may support clinical decision-making for individualized treatment after multicenter validation.

This chapter examines the domain of interpretable machine learning with a focus on methods and tools that enhance the transparency and interpretability of machine learning models, specifically within the context of official statistics. We explore various post hoc interpretation methods that provide insights into model behavior, such as loss-based feature importance methods, where we give guidance on how to choose between different such methods. Additionally, we discuss counterfactual and semi-factual explanations (SFEs) and explicitly describe interpretable regional descriptors as an example of a method generating SFEs. Furthermore, we highlight innovative contributions such as the R package mlr3summary, which facilitates model diagnostics and performance evaluation for black-box machine learning models through resampling strategies. This package provides a unified, model-agnostic summary of machine learning models, which facilitates model selection and comparison. Finally, we discuss the importance of ethical considerations and fairness in machine learning applications, emphasizing the need for methods that can address historical biases in data. The objective of this chapter is to provide practitioners and researchers with the practical tools and comprehensive knowledge required to implement interpretable and fair machine learning models, in a variety of domains—but especially in official statistics.