Explainable Artificial Intelligence Model Research Articles

The Future of Artificial Intelligence: Trends and Predictions

Artificial Intelligence (AI) has evolved rapidly, transforming diverse industries and societal functions. This paper provides a comprehensive overview of AI's current landscape, examining its advancements, applications, and ethical challenges. Key trends are explored, including innovations in machine learning and deep learning, AI’s expanding role across industries, and its potential for addressing climate change and sustainability. Furthermore, the paper highlights AI's role in enhancing human-machine collaboration, paving the way for systems that augment rather than replace human capabilities. Predictions for AI’s future are discussed, such as the emergence of artificial general intelligence (AGI), advancements in autonomous systems, the impact of quantum computing on AI, and innovations in AI-specific hardware. The paper also examines ethical and societal challenges, such as privacy, algorithmic bias, and the need for global governance, addressing the urgent call for responsible AI. In light of these trends, the paper emphasizes future research directions, encouraging interdisciplinary collaboration and a focus on explainable, robust, and resilient AI models. This work aims to shed light on the transformative potential of AI while advocating for ethical practices to ensure a positive and sustainable impact on society.

Artificial Intelligence-Powered Risk Assessment in Supply Chain Safety

The increasing complexity and globalization of supply chains necessitate robust risk management strategies to ensure safety and resilience. Traditional risk assessment methods often fall short in dynamically adapting to the rapidly changing conditions and voluminous data inherent in modern supply chains. This study explores the potential of Artificial Intelligence (AI)-powered risk assessment to address these limitations in the context of Malaysia's supply chain industry. By employing AI technologies such as machine learning, IoT, and predictive analytics, organizations can significantly enhance their risk management capabilities, improving predictive accuracy, real-time monitoring, and overall operational efficiency. Through a qualitative analysis involving in-depth interviews with supply chain managers, AI experts, and technology vendors, the study identifies the strategies employed for AI integration, the perceived effectiveness of these technologies, and the challenges faced in implementation. The findings highlight the importance of robust data governance, the development of explainable AI models, and continuous skill development to overcome barriers related to data quality, model interpretability, and high implementation costs. The study concludes with recommendations for fostering a safer and more resilient logistics environment in Malaysia, emphasizing the need for comprehensive AI adoption frameworks and scalable solutions for small and medium-sized enterprises.

Exploring AI's Role in Enhancing Risk Assessment Models in Financial Quantitative Trading

The application of Artificial Intelligence (AI) algorithms to financial risk management has fundamentally transformed the nature of risk and the tools being used to control and manage it. This paper examines how AI is deployed to manage market risk, including volatility forecasting, portfolio optimisation, stress testing, and sentiment analysis. While many of these models are well-accepted across the industry and underpin many front-office and risk-management processes, they are often limited in dynamic market environments where traditional linearity doesnt exist and real-time data is king. AI has brought advanced tools that can be applied to large and complex data sets, and that adapt quickly as market conditions change. Reinforcement learning algorithms are now deployed to optimise portfolio weights dynamically. Natural-language processing is also used to augment sentiment analysis, which can be leveraged to assess potential risk. Other forms of AI can be deployed to address challenges that come with more advanced AI models. Explainable AI models provide the opportunity to produce financial outcomes that are more transparent and less opaque while also helping to validate regulatory compliance. Analyzing the applications of AI case studies sheds light on why financial professionals must continue to embrace AI tools, many of which are designed to help them manage risk in an increasingly volatile market environment.

Credit Risk Assessment and Financial Decision Support Using Explainable Artificial Intelligence

The greatest technological transformation the world has ever seen was brought about by artificial intelligence (AI). It presents significant opportunities for the financial sector to enhance risk management, democratize financial services, ensure consumer protection, and improve customer experience. Modern machine learning models are more accessible than ever, but it has been challenging to create and implement systems that support real-world financial applications, primarily due to their lack of transparency and explainability—both of which are essential for building trustworthy technology. The novelty of this study lies in the development of an explainable AI (XAI) model that not only addresses these transparency concerns but also serves as a tool for policy development in credit risk management. By offering a clear understanding of the underlying factors influencing AI predictions, the proposed model can assist regulators and financial institutions in shaping data-driven policies, ensuring fairness, and enhancing trust. This study proposes an explainable AI model for credit risk management, specifically aimed at quantifying the risks associated with credit borrowing through peer-to-peer lending platforms. The model leverages Shapley values to generate AI predictions based on key explanatory variables. The decision tree and random forest models achieved the highest accuracy levels of 0.89 and 0.93, respectively. The model’s performance was further tested using a larger dataset, where it maintained stable accuracy levels, with the decision tree and random forest models reaching accuracies of 0.90 and 0.93, respectively. To ensure reliable explainable AI (XAI) modeling, these models were chosen due to the binary classification nature of the problem. LIME and SHAP were employed to present the XAI models as both local and global surrogates.

Explainable artificial intelligence models for predicting pregnancy termination among reproductive-aged women in six east African countries: machine learning approach

Pregnancy termination remains a complex and sensitive issue with approximately 45% of abortions worldwide being unsafe, and 97% of abortions occurring in developing countries. Unsafe pregnancy terminations have implications for women’s reproductive health. This research aims to compare black box models in their prediction of pregnancy termination among reproductive-aged women and identify factors associated with pregnancy termination using explainable artificial intelligence (XAI) methods. We used comprehensive secondary data on reproductive-aged women’s demographic and socioeconomic data from the Demographic Health Survey (DHS) from six countries in East Africa in the analysis. This study implemented five black box ML models, Bagging classifier, Random Forest, Extreme Gradient Boosting (XGB) Classifier, CatBoost Classifier, and Extra Trees Classifier on a dataset with 338,904 instances and 18 features. Additionally, SHAP, Eli5, and LIME XAI techniques were used to determine features associated with pregnancy termination and Statistical analysis were employed to understand the distribution of pregnancy termination. The results demonstrated that machine learning algorithms were able to predict pregnancy termination on DHS data with an overall accuracy ranging from 79.4 to 85.6%. The ML classifier random forest achieved the highest result, with an accuracy of 85.6%. Based on the results of the XAI tool, the most contributing factors for pregnancy termination are wealth index, current working experience, and source of drinking water, sex of household, education level, and marital status. The outcomes of this study using random forest is expected to significantly contribute to the field of reproductive healthcare in East Africa and can assist healthcare providers in identifying individuals’ countries at greater risk of pregnancy termination, allowing for targeted interventions and support.

Navigating the Integration of Machine Learning in Healthcare: Challenges, Strategies, and Ethical Considerations

The amalgamation of artificial intelligence (AI) and machine learning (ML) in healthcare offers a revolutionary prospect to improve patient outcomes, optimize workflow, and curtail expenses. Robust computational resources, competent interdisciplinary teams, and careful management of high-quality, diverse datasets are necessary for the successful deployment of ML models. To preserve public confidence and guarantee adherence to legal requirements, ethical issues, specifically those pertaining to patient consent, data privacy, and confidentiality are crucial. Regulatory bodies, legislators, and healthcare providers must all be involved in stakeholder engagement to establish a supportive environment for ML integration. The research highlights the significance of transparency and interpretability, promoting explainable AI models that augment clinician trust and enable wider adoption in the medical community. This thorough analysis finds important gaps in the current state of ML applications in healthcare and highlights new trends. It discusses ethical ramifications, realistic implementation guidelines, and the need for cooperative stakeholder engagement. The goal of the paper is to close the “translation gap” between model development and clinical application by examining real-world applications and offering a framework for the methodical integration of ML in clinical settings. The study promotes models that are not only technically sound but also user-friendly and in line with the practical needs of healthcare professionals by emphasizing the human-centric design of ML systems. By facilitating accurate and timely diagnoses, optimizing treatment plans, and eventually raising the standard of healthcare services as a whole, the strategic application of ML in the medical field holds the potential to completely transform patient care. This study opens the door to a future in which AI-driven healthcare solutions are smoothly integrated into routine clinical practice by providing useful insights and guidelines to enhance trustworthy and moral integration.

An Overview of Explainable AI Studies in the Prediction of Sepsis Onset and Sepsis Mortality.

Explainable artificial intelligence (AI) focuses on developing models and algorithms that provide transparent and interpretable insights into decision-making processes. By elucidating the reasoning behind AI-driven diagnoses and treatment recommendations, explainability can gain the trust of healthcare experts and assist them in difficult diagnostic tasks. Sepsis is characterized as a serious condition that happens when the immune system of the body has an extreme response to an infection, causing tissue and organ damage and leading to death. Physicians face challenges in diagnosing and treating sepsis due to its complex pathogenesis. This work aims to provide an overview of the recent studies that propose explainable AI models in the prediction of sepsis onset and sepsis mortality using intensive care data. The general findings showed that explainable AI can provide the most significant features guiding the decision-making process of the model. Future research will investigate explainability through argumentation theory using intensive care data focused on sepsis patients.

An explainable artificial intelligence model for predictive maintenance and spare parts optimization

Maintenance strategies are vital for industrial and manufacturing systems. This study considers a proactive maintenance strategy and emphasizes using analytics and data science. We propose an Explainable Artificial Intelligence (XAI) methodology for predictive maintenance. The proposed method utilizes a machine learning project cycle and Python libraries to interpret the results using the Local Interpretable Model-agnostic Explanations (LIME) method. We also introduce an early concept of spare parts management, presenting insights from predictive maintenance outcomes and providing explanations for decision-makers to enhance their understanding of the influential factors behind predictions. This study demonstrates that utilizing machine learning models in predictive maintenance is highly beneficial; however, the binary outcomes of these models can be misunderstood by decision-makers. Detailed explanations provided to decision-makers will directly impact maintenance decisions and improve spare part management.

Comparison of Explainable Artificial Intelligence Model and Radiologist Review Performances to Detect Breast Cancer in 752 Patients.

Breast cancer is a type of cancer caused by the uncontrolled growth of cells in the breast tissue. In a few cases, erroneous diagnosis of breast cancer by specialists and unnecessary biopsies can lead to various negative consequences. In some cases, radiologic examinations or clinical findings may raise the suspicion of breast cancer, but subsequent detailed evaluations may not confirm cancer. In addition to causing unnecessary anxiety and stress to patients, such diagnosis can also lead to unnecessary biopsy procedures, which are painful, expensive, and prone to misdiagnosis. Therefore, there is a need for the development of more accurate and reliable methods for breast cancer diagnosis. In this study, we proposed an artificial intelligence (AI)-based method for automatically classifying breast solid mass lesions as benign vs malignant. In this study, a new breast cancer dataset (Breast-XD) was created with 791 solid mass lesions belonging to 752 different patients aged 18 to 85 years, which were examined by experienced radiologists between 2017 and 2022. Six classifiers, support vector machine (SVM), K-nearest neighbor (K-NN), random forest (RF), decision tree (DT), logistic regression (LR), and XGBoost, were trained on the training samples of the Breast-XD dataset. Then, each classifier made predictions on 159 test data that it had not seen before. The highest classification result was obtained using the explainable XGBoost model (X2GAI) with an accuracy of 94.34%. An explainable structure is also implemented to build the reliability of the developed model. The results obtained by radiologists and the X2GAI model were compared according to the diagnosis obtained from the biopsy. It was observed that our developed model performed well in cases where experienced radiologists gave false positive results.

Explainable artificial intelligence models for mineral prospectivity mapping

Explainable artificial intelligence models for mineral prospectivity mapping

Evaluating cooling effect of blue-green infrastructure on urban thermal environment in a metropolitan city: Using geospatial and machine learning techniques

Urban blue-green infrastructures (BGIs) have a considerable cooling effect, reducing the urban heat island effect. Therefore, this study aims to evaluate the cooling effects of BGIs and role of the influencing factors on urban thermal environment in Kolkata city, which is one of the largest mega cities of India. An integrated radius and non-linear regression approach was used to determine the maximum distance over which cooling is effective. Further, the relationship between cooling effects of BGIs and influencing factors was analysed using Random Forest based Explainable artificial intelligence model. The result shows that for most of the BGIs, the range of maximum cooling distance varies from 210 to 240 m from the boundary of BGIs. The average PCA value ranges from 6.12 to 16.77 hectares and PCE value ranges from 1.62 to 2.87 depending of BGIs. The results of PCA, PCE and maximum cooling distance indicate that the spatial configurations of BGIs and landscape pattern within and surrounding of BGIs sites strongly influence the cooling effects. The outcomes of this study provide valuable insights for the policymakers and urban planner about the intervention of blue-green infrastructures for the better urban environmental conditions and heat mitigation in mega cities and other urban areas.

An interpretable artificial intelligence model based on CT for prognosis of intracerebral hemorrhage: a multicenter study

ObjectivesTo develop and validate a novel interpretable artificial intelligence (AI) model that integrates radiomic features, deep learning features, and imaging features at multiple semantic levels to predict the prognosis of intracerebral hemorrhage (ICH) patients at 6 months post-onset.Materials and methodsRetrospectively enrolled 222 patients with ICH for Non-contrast Computed Tomography (NCCT) images and clinical data, who were divided into a training cohort (n = 186, medical center 1) and an external testing cohort (n = 36, medical center 2). Following image preprocessing, the entire hematoma region was segmented by two radiologists as the volume of interest (VOI). Pyradiomics algorithm library was utilized to extract 1762 radiomics features, while a deep convolutional neural network (EfficientnetV2-L) was employed to extract 1000 deep learning features. Additionally, radiologists evaluated imaging features. Based on the three different modalities of features mentioned above, the Random Forest (RF) model was trained, resulting in three models (Radiomics Model, Radiomics-Clinical Model, and DL-Radiomics-Clinical Model). The performance and clinical utility of the models were assessed using the Area Under the Receiver Operating Characteristic Curve (AUC), calibration curve, and Decision Curve Analysis (DCA), with AUC compared using the DeLong test. Furthermore, this study employs three methods, Shapley Additive Explanations (SHAP), Grad-CAM, and Guided Grad-CAM, to conduct a multidimensional interpretability analysis of model decisions.ResultsThe Radiomics-Clinical Model and DL-Radiomics-Clinical Model exhibited relatively good predictive performance, with an AUC of 0.86 [95% Confidence Intervals (CI): 0.71, 0.95; P < 0.01] and 0.89 (95% CI: 0.74, 0.97; P < 0.01), respectively, in the external testing cohort.ConclusionThe multimodal explainable AI model proposed in this study can accurately predict the prognosis of ICH. Interpretability methods such as SHAP, Grad-CAM, and Guided Grad-Cam partially address the interpretability limitations of AI models. Integrating multimodal imaging features can effectively improve the performance of the model.Clinical relevance statementPredicting the prognosis of patients with ICH is a key objective in emergency care. Accurate and efficient prognostic tools can effectively prevent, manage, and monitor adverse events in ICH patients, maximizing treatment outcomes.

Explainable Artificial Intelligence Methods to Enhance Transparency and Trust in Digital Deliberation Settings

Digital deliberation has been steadily growing in recent years, enabling citizens from different geographical locations and diverse opinions and expertise to participate in policy-making processes. Software platforms aiming to support digital deliberation usually suffer from information overload, due to the large amount of feedback that is often provided. While Machine Learning and Natural Language Processing techniques can alleviate this drawback, their complex structure discourages users from trusting their results. This paper proposes two Explainable Artificial Intelligence models to enhance transparency and trust in the modus operandi of the above techniques, which concern the processes of clustering and summarization of citizens’ feedback that has been uploaded on a digital deliberation platform.

AI-Powered Customs Clearance: Optimizing Trade Compliance and Border Management

This research explores the integration of artificial intelligence (AI) in customs clearance systems to enhance trade compliance and border management. By leveraging machine learning (ML), natural language processing (NLP), and predictive analytics, the study demonstrates how AI-driven systems can optimize risk assessment, automate document processing, and detect trade anomalies. Case studies illustrate the successful implementation of AI-based solutions in enhancing customs efficiency, reducing processing times, and improving decision-making accuracy. The research identifies key challenges, including data quality, interoperability, and ethical considerations, and proposes future directions, such as integrating blockchain with AI and developing explainable AI models.

Pediatric Posterior Fossa Tumors Classification and Explanation-Driven with Explainable Artificial Intelligence Models

The use of deep learning for identifying defects in medical images has rapidly emerged as a significant area of interest across various medical diagnostic applications. The automated recognition of Posterior Fossa Tumors (PFT) in Magnetic Resonance Imaging (MRI) plays a vital role, as it furnishes essential data about irregular tissue, essential for treatment planning. Human examination has traditionally been the standard approach for identifying defects in brain MRI. This technique is unsuitable for a massive quantity of data. Therefore, automated PFT detection techniques are being established to minimize radiologist's time. In this paper, the posterior fossa tumor is detected and classified in brain MRI using Convolutional Neural Network (CNN) algorithms, and the model result and accuracy obtained from each algorithm are explained. A dataset collection made up of 3,00,000 images with an average of 500 patients from the Children's Cancer Hospital Egypt (CCHE) was used. The CNN algorithms investigated to classify the PFT were VGG19, VGG16, and ResNet50. Moreover, explanations for the behavior of networks were investigated using three different techniques: LIME, SHAP, and ICE. Overall, the results showed that the best model was VGG16 compared with other CNN-used models with accuracy rate values of 95.33%, 93.25%, and 87.4%, respectively.

Hybrid Explainable Artificial Intelligence Models for Targeted Metabolomics Analysis of Diabetic Retinopathy.

Diabetic retinopathy (DR) is a prevalent microvascular complication of diabetes mellitus, and early detection is crucial for effective management. Metabolomics profiling has emerged as a promising approach for identifying potential biomarkers associated with DR progression. This study aimed to develop a hybrid explainable artificial intelligence (XAI) model for targeted metabolomics analysis of patients with DR, utilizing a focused approach to identify specific metabolites exhibiting varying concentrations among individuals without DR (NDR), those with non-proliferative DR (NPDR), and individuals with proliferative DR (PDR) who have type 2 diabetes mellitus (T2DM). A total of 317 T2DM patients, including 143 NDR, 123 NPDR, and 51 PDR cases, were included in the study. Serum samples underwent targeted metabolomics analysis using liquid chromatography and mass spectrometry. Several machine learning models, including Support Vector Machines (SVC), Random Forest (RF), Decision Tree (DT), Logistic Regression (LR), and Multilayer Perceptrons (MLP), were implemented as solo models and in a two-stage ensemble hybrid approach. The models were trained and validated using 10-fold cross-validation. SHapley Additive exPlanations (SHAP) were employed to interpret the contributions of each feature to the model predictions. Statistical analyses were conducted using the Shapiro-Wilk test for normality, the Kruskal-Wallis H test for group differences, and the Mann-Whitney U test with Bonferroni correction for post-hoc comparisons. The hybrid SVC + MLP model achieved the highest performance, with an accuracy of 89.58%, a precision of 87.18%, an F1-score of 88.20%, and an F-beta score of 87.55%. SHAP analysis revealed that glucose, glycine, and age were consistently important features across all DR classes, while creatinine and various phosphatidylcholines exhibited higher importance in the PDR class, suggesting their potential as biomarkers for severe DR. The hybrid XAI models, particularly the SVC + MLP ensemble, demonstrated superior performance in predicting DR progression compared to solo models. The application of SHAP facilitates the interpretation of feature importance, providing valuable insights into the metabolic and physiological markers associated with different stages of DR. These findings highlight the potential of hybrid XAI models combined with explainable techniques for early detection, targeted interventions, and personalized treatment strategies in DR management.

Feasibility study and optimization of limestone calcined clay composites for compressive strength using multi-layered explainable artificial intelligence models

Interpretable white box machine learning models to predict the compressive strength of Limestone Calcined Clay (LC3) cementitious composite have been developed and analyzed. For this, a large dataset consisting of more than 4000 numerical instances has been collected from the literature, which has been tuned and normalized. Mainly based on the proportion of the chemical composition of its main ingredients – Limestone (0–31 by %mass of LC3), Calcined Clay (0–70 by %mass of LC3), and Cement (30–100 by %mass of LC3) with a water-to-binder ratio ranging from 0.40 to 0.60, predictive models have been developed for the 28 days compressive strength (fc) of LC3. The performance of various predictive models is then compared across the dataset, and suggestions are made for their reliability and explainability. The models are further interpretated using multi-layered explanations invoking model-agnostic explainable artificial intelligence (XAI) technique. Optimization and interaction plots have also been generated and shown as three-dimensional (3D) visualization. Percent of calcined clay, cement and the water to binder ratio were found to be the most influencing parameters for the compressive strength of LC3. Results of this study indicate that the optimized blends of LC3 can be expressed with up to 50 % replacement of OPC without any significant compromise in fc, contributing to the development of sustainable cementitious composites. The major advantage of using this framework lies in the model agnostic approach, so different predictive models can be compared in a same manner and thus, this approach is applicable to interpretable forecasting models for binary, ternary, quaternary sustainable cementitious composites.

Oil Market Dynamics and US Monetary Policy Uncertainty: Evidence from Explainable Artificial Intelligence Models

We utilize explainable artificial intelligence techniques to examine the main link between oil market dynamics and US monetary policy uncertainty. Our analysis unveils key findings. First, higher oil prices and rising oil market uncertainty are significantly associated with elevated levels of monetary policy uncertainty. Second, option-implied oil market volatility (an indicator of oil market uncertainty) emerges as the most important predictor of monetary policy uncertainty. Third, option-implied oil market volatility frequently interacts with option-implied gold market volatility while oil prices mostly interact with fluctuations in the US dollar. The findings offer valuable insights into the understanding of monetary policy uncertainty.

A tree-based explainable AI model for early detection of Covid-19 using physiological data

With the outbreak of COVID-19 in 2020, countries worldwide faced significant concerns and challenges. Various studies have emerged utilizing Artificial Intelligence (AI) and Data Science techniques for disease detection. Although COVID-19 cases have declined, there are still cases and deaths around the world. Therefore, early detection of COVID-19 before the onset of symptoms has become crucial in reducing its extensive impact. Fortunately, wearable devices such as smartwatches have proven to be valuable sources of physiological data, including Heart Rate (HR) and sleep quality, enabling the detection of inflammatory diseases. In this study, we utilize an already-existing dataset that includes individual step counts and heart rate data to predict the probability of COVID-19 infection before the onset of symptoms. We train three main model architectures: the Gradient Boosting classifier (GB), CatBoost trees, and TabNet classifier to analyze the physiological data and compare their respective performances. We also add an interpretability layer to our best-performing model, which clarifies prediction results and allows a detailed assessment of effectiveness. Moreover, we created a private dataset by gathering physiological data from Fitbit devices to guarantee reliability and avoid bias.The identical set of models was then applied to this private dataset using the same pre-trained models, and the results were documented. Using the CatBoost tree-based method, our best-performing model outperformed previous studies with an accuracy rate of 85% on the publicly available dataset. Furthermore, this identical pre-trained CatBoost model produced an accuracy of 81% when applied to the private dataset. You will find the source code in the link: https://github.com/OpenUAE-LAB/Covid-19-detection-using-Wearable-data.git.

An extensive review on significance of Explainable Artificial Intelligence models in discrete domains for informed decisions making

An extensive review on significance of Explainable Artificial Intelligence models in discrete domains for informed decisions making