Author

Jannatul Ferdaus Disha

View on Semantic Scholar

2 papers indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Open access Jul 2026

Integrating Reinforcement Learning With Explainable Artificial Intelligence for Real‐Time Clinical Decision Support in Dynamic Healthcare Environments

In critical care settings such as intensive care units (ICUs) and emergency departments, clinical decision support must adapt rapidly to changing patient states while remaining transparent enough to earn clinician trust. This article presents an integrated Reinforcement Learning and Explainable Artificial Intelligence (RL‐XAI) framework for real‐time critical‐care decision support. A Deep Q‐Network (DQN) is trained offline on 18 142 mechanically ventilated ICU stays from the eICU Collaborative Research Database, learning adaptive policies for ventilator settings and the titration and timing of vasopressors and fluids. An explainability layer embedded within the DQN inference loop generates dual‐modal explanations in real time: global, population‐level feature‐importance summaries and local, patient‐specific rationales. Running on a single Intel Xeon Gold 6154 CPU server (3.0 GHz, 18 cores, no GPU), end‐to‐end latency—comprising the DQN forward pass, TreeSHAP attribution (background n = 100), and LIME surrogate fitting (50 perturbations)—ranged from 210 to 300 ms across 500 held‐out test cases (median 247 ms; 95th percentile 291 ms), well below the 500 ms threshold for clinical actionability. An initial clinician assessment confirmed the explanations’ usefulness and interpretability. The framework advances real‐time, explainable, adaptive AI for safety‐critical healthcare.

Jannatul Ferdaus Disha, Faiaz Zaman Dehan, Kazi Md. Tanvir Anzum · Advanced Intelligent Systems · 0 citations

Conference Open access Aug 2026

Exploring The Barriers to Artificial Intelligence Integration in Green Supply Chain Using Interpretive Structural Modeling Approach: Insights from Bangladesh's Multi-Stakeholder Perspective

In recent years, Artificial Intelligence (AI) integration to Green Supply Chains (GSC) has been identified as a vital organizational practice for achieving profitability by reducing both environmental and social risks. So far, no study has explored the barriers to implementing AI integrate GSC. The current research will analyze the barriers to adopting the concept of AI in GSC through multi-stakeholder aspect with regard to opinions provided by industry practitioners, government, and academics. The study utilized a thorough literature analysis, and consultations with experts resulted in a list of 11 barriers to AI implementation in GSC. These are the low top management commitment, the high cost of investing in AI, low digital infrastructure etc. This study used Interpretive Structural Modeling (ISM) to represent the contextual relationships between the barriers, resulting in a hierarchical model. The results of the study show that obstacles are categorized into six hierarchic levels in the model with the top management commitment and Government support are identified as the leading drivers. Addressing the main cause of the problem, which is improving infrastructure, raising awareness, and developing professional ability, is critical to encouraging AI adoption. One limitation is that the study is based on the opinions of experts limited to Bangladesh, which may reduce generalizability. The implications include the recommendation that policymakers and supply chain managers focus on initiatives to address the driving barriers and develop supporting policies to assist AI-enabled practices in the sustainable process of resource-constrained situations.

Jannatul Ferdaus Disha, Kazi Md. Tanvir Anzum, M. Masum et al. · Engineering · 0 citations