جستجو
Trustworthy AI in Healthcare: A Modality-Aware Evaluation Framework for XAI Fidelity and Stability in Clinical Decision Support
seyedehSamira Rabinataj1
Abstract :
Abstract
The clinical translation of artificial intelligence (AI) into decision support workflows remains contingent not only on predictive accuracy but on the reliability of the explanations that accompany model outputs. Methods such as SHAP, LIME, and Integrated Gradients (IG) are widely deployed in healthcare AI, yet their fidelity and stability are typically reported on a single dataset within a single clinical task, leaving open the question of whether ranking relationships transfer across the heterogeneous data structures encountered in clinical decision support. This paper proposes the Modality-Aware Evaluation Framework for XAI (MAEF-XAI), a methodological pipeline that jointly evaluates three normalized properties—fidelity, stability, and clinical plausibility—and aggregates them via a documented Trustworthiness Index (TI). The framework is demonstrated on a fully reproducible micro-experiment that contrasts a tabular electronic-health-record-style classification task with a low-resolution medical-image-style classification task. The experiment shows that ranking relationships between explainers do not transfer between task types: on the tabular task, SHAP achieves the highest stability and ground truth plausibility, whereas on the imaging task IG outperforms both SHAP and LIME on stability and plausibility while LIME achieves the highest composite TI. We also show that a small (±0.1) perturbation of the TI weights can change method rankings in the imaging condition, underscoring that aggregation choices must be reported transparently. The paper is positioned as a methodological contribution: MAEF-XAI provides an audit-ready protocol that frames XAI evaluation as a per-task design problem rather than a universal property of the explainer.
Keywords: explainable AI, trustworthy AI, clinical decision support, fidelity, stability, SHAP, LIME, Integrated Gradients, evaluation methodology
Keywords : explainable ai trustworthy ai clinical decision support fidelity stability shap lime integrated gradients evaluation methodology
زبان مقاله : انگلیسی
محور مقاله : هوش مصنوعی
محل انتشار : چهارمین کنگره بین المللی علوم، مهندسی و فن آوری های نو
آدرس وب سایت کنفرانس : secongress.com
کد IOI مقاله : ACCA-ACAZFF
لینک مستقیم مقاله : https://isnac.ir/ACCA-ACAZFF
نمایش چکیده فارسی
: https://isnac.ir/SA/ACCA-ACAZFF
سایر مقالات محل انتشار
- همگرایی نهادی و کارایی فناورانه در توسعه شهر هوشمند: تحلیل تطبیقی مدلهای آسیایی
- بهینهسازی استوار و چندمرحلهای تصادفی برای کربنزدایی عمیق و استراتژی سبد محصولات در پتروپالایشگاههای یکپارچه تحت عدم قطعیت عمیق
- شناسایی فن آوریهای نوین و اولویتبندی آنها برای استفاده در مدیریت نگهداشت تجهیزات دریایی و بندری؛ مطالعه موردی بندر چابهار
- A Review of Recent Advances in Nanoparticle-Enhanced 1–3 Piezoelectric Composite Sensors for High-Sensitivity and Broadband Medical Imaging
- The dawn of a new generation of CNC on the path to sustainable, precise production (smart machining)
- Optimization of CNC machining parameters with the aim of increasing surface quality and reducing production time
- The role of space and place in the formation of psychological behaviors
- Design and Analysis of Combined Power Amplifiers (Combined PAs) Using Nonlinear Structures: From Circuit Implementation to Spectral-Energy Efficiency Evaluation in Hybrid-Beamforming MIMO Systems
- Power Management Strategies Incorporating Thermal Considerations for Integrated Power Amplifiers in Digital Phased-Array Systems
- Design and Optimization of a Sub-threshold Differential Amplifier for Ultra-low-power Biosignal Processing Applications
- Comparative Analysis and Performance Enhancement of FinFET-Based Differential Amplifiers for High-Speed Telecommunication Systems
- Design of a High-Linearity Power Amplifier Using Feedback Structures at High Frequencies
- Quantum Symmetry and Compactness in Graphs: A Convex-Geometric Approach
- Green Entrepreneurship and the Transition to a Sustainable Economy: A Conceptual Framework for Transformative Transition Capacity
- A Review of Thermoelectric Energy Harvesting Circuits and Methods for Self-Powered Wireless Sensor Nodes and the Internet of Things