Ali Nadhim Kamber

Facial expression recognition (FER) is a highly active field with applications in computer vision, human-computer interaction, security, and computer graphics animation. Recent advancements in deep learning and machine learning have increased interest in utilizing these techniques for accurate facial expression classification. This paper presents a comparative study that evaluates the performance of deep learning and machine learning as classifiers in FER systems, specifically after data fusion. Data fusion techniques combine and integrate multiple sources of information, aiming to enhance the overall classification accuracy by extracting two types of features using geometrical and appearance features trained using two types of convolutional neural networks. The feature outputs of these networks are fused to create a final feature vector for the classification process. The study evaluates the performance of deep learning on two benchmark datasets, the extended Cohn-Kanade (CK+) and Oulu-CASIA datasets, to assess the performance of deep learning. As a point of comparison, the traditional machine learning approach based on the support vector machine (SVM) is also evaluated on the same datasets. Performance metrics such as classification accuracy, precision, recall, and F1-score are utilized. The results obtained from the study highlight the strengths and limitations of both deep learning and machine learning techniques when employed as classifiers in FER systems. Notably, the experimental results demonstrate that the deep learning approach significantly outperforms the baseline methods, achieving an increase in recognition accuracy of 5.22% for the CK+ and 3.07% for the Oulu-CASIA dataset.

Explainable Artificial Intelligence (XAI) has emerged as a critical enabler for deploying AI-driven medical imaging systems where transparency, trust, and accountability are paramount. However, most current taxonomies of XAI methods categorize techniques based on algorithmic families (e.g., saliency maps, attribution methods), which often fail to reflect the practical requirements of clinical tasks. This paper proposes a novel task-centric taxonomy of XAI in medical imaging that aligns explanation techniques with four key clinical tasks: classification, detection, segmentation, and prognostic assessment. For each task, we analyze how different XAI methods enhance model interpretability, their suitability for clinical decision-making, and their limitations in real-world applications. Our taxonomy aims to provide a practical framework for researchers and practitioners to select appropriate XAI strategies tailored to the specific demands of medical imaging workflows. Furthermore, we highlight the current gaps in task-specific explainability and propose future research directions towards clinically meaningful, task-driven XAI solutions. This work serves as a step towards bridging the gap between technical XAI developments and the functional needs of clinical practice.