This Special Issue of RIST : “Revue d’Information Scientifique et Technique” (in English: Journal of Scientific and Technical Information), within the CERIST Collection “Information Processing at the Digital Age Review”, brings together six rigorously selected contributions presented in the framework of the Multi-Conference OCTA’2025, held on 6–7 November 2025 at HIDE University of Tunis. The event gathered five parallel international conferences: SIIE, ISKO-Maghreb, CITED, TBMS, and OCTA, establishing an interdisciplinary platform dedicated to Artificial Intelligence, knowledge organization, information engineering, and advanced digital technologies.
Centered on the thematic focus “Artificial Intelligence in Medical Imaging and Healthcare: Intelligent and Explainable Diagnostic Systems”, this issue reflects the growing convergence between cognitive computing, data-centric medical analysis, and responsible clinical innovation. These six collected chapters collectively, in a set of best papers, address foundational and applied challenges in AI-driven healthcare systems, including structured knowledge modeling, adaptive learning architectures, scalable distributed intelligence, domain specific clinical deployment, explainable AI mechanisms, and governance frameworks for sustainable and ethical medical AI eco-systems.
These contributions advance the state of the art in intelligent diagnostic systems by integrating semantic interoperability, hybrid symbolic-statistical reasoning, robustness under data variability, and transparency in high-stakes clinical environments. Experimental validations across heterogeneous medical datasets demonstrate improvements in predictive reliability, interpretability, computational efficiency, and cross-domain generalization. Beyond algorithmic performance, the issue emphasizes methodological rigor, reproducibility, ethical compliance, and alignment with emerging international regulatory standards.
By bridging theoretical foundations, computational innovation, and translational medical applications, this Special Issue on “Artificial Intelligence in Medical Imaging and Healthcare: Intelligent and Explainable Diagnostic Systems”, contributes to the structuring of AI as a mature scientific discipline within healthcare. It positions intelligent and explainable diagnostics as a cornerstone of next-generation clinical decision support systems and sustainable digital health infrastructures. The interdisciplinary synergy fostered by the Int. Multi Conference OCTA’2025 reinforces the global ambition of RIST to promote responsible, scalable, and knowledge driven Artificial Intelligence research at the international level.

Keywords: Artificial Intelligence; Medical Imaging; Intelligent Diagnostic Systems; Explainable AI (XAI); Clinical Decision Support; Cognitive Computing; Knowledge Representation; Hybrid AI Systems; Data-Centric Healthcare; Distributed AI Architectures; Responsible AI; Digital Health; Semantic Interoperability; Machine Learning in Medicine; AI Governance.

Téléchargement : PDF

Medical X-ray imaging is a fundamental pillar of musculoskeletal assessment, providing useful information on bone structure, alignment, and potential anomalies. However, such images are generally accompanied by noise, poor contrast, and a number of different types of artifacts, which can mask important anatomical detail and render accurate analysis impossible. These limitations not only affect diagnostic interpretation but also pose challenges to the design and optimization of ergonomic medical devices that rely on precise anatomical data to deliver personalized support. In this article, we review classic image-enhancement techniques alongside modern deep-learning approaches for improving the quality and interpretability of X-ray images. We propose an experimental design workflow that evaluates the performance of several augmentation methods towards the application in intelligent chair design. Further, we present an integrated system architecture that incorporates augmented imaging and AI-driven decision-making modules.
Finally, we consider both objective metrics and observer-rated evaluation criteria to evaluate the performance of the proposed strategies and their capacity to optimize patient care and device function.
Index Terms—X-ray enhancement, CLAHE, Bilateral Filter,Histogram Equalization, Log Correction, Adaptive Threshold,Gaussian Blur, AI, Smart Chair, IoT.

Auteurs : Ali Hamdi , Sawsan Selmi, Hédi Sakli

Téléchargement : PDF

Colorectal cancer remains a major public health challenge, and accurate quantification of T lymphocytes (CD3+,CD8+) within the tumor microenvironment holds important prognostic and predictive value. We present MS-HAF-Net, a multiscale hybrid dual-encoder architecture (a convolutional branch based on ResNet34 and a Swin Transformer branch) designed for both semantic segmentation and instance segmentation of lymphocytes in histopathological images. MS-HAF-Net fuses fine
local features with global contextual representations through multi-scale fusion, and is enhanced by a Residual Feature Augmentation (RFA) module and ECA attention blocks to improve detection of very small objects.
The network outputs multi-task heads (masks, contours, and center heatmaps) that enable robust instance separation via watershed post-processing. Comparative experiments against several baselines (YOLOv8-Seg, YOLOv11-Seg, SwinUNet, TransUNet, U-Net+EfficientNetB4) demonstrate that MS-HAF-Net reduces instance merging in dense regions while maintaining high localization accuracy. These results position MS-HAF-Net as a promising solution for integrating automated diagnostic assistance tools in oncology.
Index Terms—instance segmentation, histopathology, lymphocytes, CNN+Transformer, Swin, ResNet, MS-HAF-Net, ECA,RFA.

Auteurs : Feriel Lamirem , Wyssem Fathallah, Imen Helal, Nizar Sakli, Hedia Bellali, Hedi Sakli

Téléchargement : PDF

We conduct a comparative evaluation of chest X-ray and computed tomography (CT) scan imaging modalities for machine learning-based automatic COVID-19 diagnosis. Using a public corpus of 17,599 medical images, we applied transfer learning with MobileNetV2 to assess diagnostic accuracy, sensitivity, and specificity and established that CT scans are more accurate than X-rays, yet X-rays are more sensitive to non-COVID cases, and therefore have complementary roles in clinical practice.
Keywords— COVID-19 screening, Medical imaging, Chest X-ray, Computed tomography (CT) scan, Machine learning, Deep learning, Transfer learning, Image reprocessing, Artifact removal, Deblurring, AI-aided diagnosis, Data augmentation,Diagnostic accuracy, Multimodal imaging.

Auteurs : Jawhara Lazhar , Boutheina Ben-Ismail, Wyssem Fathallah, Hedi Sakli

Téléchargement : PDF

Pneumonia remains a critical public health issue, requiring accurate and efficient diagnostic support. This study presents a deep learning–based method for the automatic classification of chest X-ray images to assist clinicians in pneumonia detection. A convolutional neural network (CNN) was trained and validated on the NIH Chest X-ray dataset,incorporating tailored preprocessing and optimization techniques.
The proposed model achieved an accuracy exceeding 91% and an F1-score above 93%, demonstrating high robustness and outperforming conventional approaches. These results confirm the potential of deep learning to enhance diagnostic precision while reducing the annotation workload. Despite ongoing challenges—such as data quality, interpretability, and generalization—this work underlines the value of artificial intelligence as a promising decision-support tool in medical imaging.
Keywords— medical imaging, artificial intelligence, deep learning, thoracic imaging, pneumonia.

Auteurs : Bourkache Noureddine , Laghrouche Mourad, Sahbi Sidhom

Téléchargement : PDF

A major obstacle in fields including sports, clinical rehabilitation, and workplace safety is the timely detection and prevention of physical injuries. The majority of conventional monitoring systems are reactive, depending on post-event analysis or unimodal data sources, which restricts their ability to provide
proactive actions and early warnings. Furthermore, current AI-driven health systems lack rigorous validation procedures, which compromises their suitability for practical implementation in safety-critical settings.
In this work, we present MHIDS (Multimodal Hybrid Injury Detection System), an integrated, AI-based diagnostic framework that combines wearable physiological sensors, computer vision,and personalized physiological modeling for real-time injury forecasting. A continuously updated digital twin is employed
to capture each user’s biomechanical and physiological profile, allowing adaptive, individualized risk assessment. Unlike conventional approaches, MHIDS incorporates a closed-loop feedback mechanism that dynamically reconfigures sensing parameters and provides actionable recommendations (e.g., posture correc-
tion, intensity adjustment, or rest scheduling), thereby shifting the paradigm from passive detection to proactive prevention.
To guarantee correctness and operational trustworthiness,MHIDS is formally modeled in UPPAAL as a network of timed automata, ensuring critical properties such as bounded response times (<100 ms), safety, liveness, and deadlock freedom. Experimental validation using the publicly available MHEALTH dataset demonstrates superior predictive performance, achieving an accuracy of 99.21%, precision of 98.94%, recall of 99.07%,and F1-score of 99.00%, significantly outperforming state-of-the art baselines. Index Terms—Multimodal, Hybrid, Injury, Detection, AI, Healthcare Monitoring Auteurs : Imen Chebbi , Sarra Abidi, Leila Ben Ayed

Téléchargement : PDF

Heart failure (HF) is still a major cause of morbidity and death worldwide, hence prompt intervention requires early
detection. In this work, we suggest a unique hybrid deep learning architecture that uses clinical data to predict the early development of heart failure by combining supervised and unsupervised learning. Our design uses a deep neural network (DNN) for risk categorization after a deep autoencoder for denoised feature
extraction.
Unlike prior works that focus solely on predictive accuracy,we incorporate formal logic-based safety constraints using the Marabou verification framework, enabling our model to operate with mathematical safety guarantees an essential step for real world deployment in clinical settings. Specifically, we verify two
critical properties: output stability (resistance to small input perturbations) and bounded input robustness (reliable behavior
within clinically valid input ranges).
The model achieves a classification accuracy of 99.45%,validated through stratified 10-fold cross-validation and repeated trials to ensure statistical robustness. Our results demonstrate that integrating formal methods with deep learning significantly improves both performance and safety, highlighting the framework’s potential for deployment in trustworthy clinical AI systems.
Index Terms—Safety, Formally, Verified, Heart, Failure, DNN

Auteurs : Sarra Abidi , Imen Chebbi, Leila Ben Ayed

Téléchargement : PDF

Volume 28 Numéro 01

• AI-Enhanced Framework for Automated Brain Tumor Detection and Segmentation Using MRI
• Classification de texte : Approches et avancées en apprentissage profond
• Harmonizing Industry: Techniques, Challenges, and Case Studies in Industrial Data Integration
• Modèles de Processus Administratifs : Analyse de Conformité
• نحو براديقم جديد لتصميم ‫الصورة

Accurate and rapid detection of brain tumors is vital for timely diagnosis and treatment planning. This paper proposes an improved computer-aided diagnosis (CAD) system enhanced with artificial intelligence (AI) techniques for automated brain tumor segmentation from magnetic resonance imaging (MRI). The proposed framework consists of three main stages: (1)
image preprocessing using spatial filters and contrast enhancement, (2) segmentation using a deep learning-based thresholding and region proposal network, and (3) post-processing based on morphological operations and adaptive masking. Experimental results on a public dataset show an average accuracy of 97% and a Dice similarity coefficient above 0.86. Compared to traditional segmentation techniques, the integration of AI significantly improves robustness, reduces
false positives, and enhances boundary precision, making this system suitable for clinical implementation.

Auteurs : Abdellatif BOUZID-DAHO

Téléchargement : PDF

Ces dernières années, le nombre de documents et de textes complexes a connu une croissance exponentielle, nécessitant une compréhension approfondie des méthodes d’apprentissage automatique pour classer les textes avec précision dans de nombreuses applications. De nombreuses approches d’apprentissage automatique ont obtenu des résultats remarquables dans le domaine du traitement du langage naturel.
La classification de texte consiste à classer automatiquement un ensemble de documents dans plusieurs catégories prédéfinies en fonction de leur contenu et de leur sujet. L’objectif principal de la classification de texte est de permettre aux utilisateurs d’extraire des informations des ressources textuelles et de traiter conjointement des processus tels que la recherche, la classification et les techniques d’apprentissage automatique afin de classifier différentes catégories.
Le succès de ces algorithmes d’apprentissage repose sur leur capacité à comprendre des modèles complexes et des relations non linéaires dans les données. Cependant, identifier des structures, des architectures et des techniques adaptées pour la classification de texte constitue un défi pour les chercheurs.
Dans cet article, un aperçu des algorithmes de classification de texte est présenté. Cet aperçu couvre différentes méthodes
d’extraction des caractéristiques textuelles, des techniques de réduction de la dimensionnalité, des algorithmes et
techniques existants. Enfin, les différentes méthodes de classification de texte par apprentissage profond sont comparées et
résumées.

Auteurs : Hadjer MEZIANI , Sabah KIRAT

Téléchargement : PDF