Nassima Dif

• Ph.D
• Teacher and researcher at Ecole Superieure en Informatique, Sidi Bel-Abbes, Algeria

Predictive medicine aims to improve the stratification and management of individuals according to their probability of developing a pathology in the future. The risk of breast cancer is influenced by multiple factors, such as lifestyle, age, breast density, and family history. Nevertheless, radiologists face challenges in accurately classifying bre...

Breast density classification presents a crucial risk factor for breast cancer. The breast density evaluation via the breast imaging reporting and data system (BI-RADS) presents several challenges for medical professionals due to their variable diagnostics and workload. To address these challenges, researchers have developed computer-aided diagnost...

This chapter compares 4 variants of metaheuristics (RFA, EMVO, RPSO, and RBAT). The purpose is to test the impact of refinement on different types of metaheuristics (FA, MVO, PSO, and BAT). The refinement helps to enhance exploitation and to speed up the search process in multidimensional spaces. Moreover, it presents a powerful tool to solve diffe...

Chest x-rays (CXRs) are broadly used in clinical practice to diagnose pulmonary diseases. Developing reliable computer-aided diagnosis (CAD) systems to automate the interpretation of CXRs can save medical practitioners time and improve diagnosis accuracy. Besides, segmenting lungs accurately plays a vital role in CAD systems. However, several chall...

The advances of deep learning in histopathology show the ability to assist pathologists in reducing workload and avoiding subjective decisions. Such algorithms lead to a more reliable diagnosis because they give computer-based second opinions to the clinician. However, in histopathology cancer image analysis, pathologists mostly diagnose the pathol...

This study introduces a new strategy that combines unsupervised learning (clustering) and transfer learning. Clustering methods are employed to generate synthetic labels for the source dataset (ICAR-2018). The generated dataset is then used for transfer learning to other histopathological datasets (KimiaPath960, CRC, Biomaging−2015, Breakhis, and L...

Hybrid metaheuristics have received a lot of attention lately to solve combinatorial optimization problems. The purpose of hybridization is to create a cooperation between metaheuristics for better solutions. Most proposed works were interested in static hybridization. The objective of this work is to propose a novel dynamic hybridization method (G...

La vision par ordinateur est un champ d’étude qui permet aux systèmes automatiques à reconnaitre les entrées visuelles pour les exploiter dans des tâches de recommandation. Dans ces dernières années, la quantité des images et des vidéos a largement augmenté. L’exploitation des systèmes de vision par ordinateur pour l’analyse de cette quantité d’inf...

One of the important processes in the data quality field is record linkage (RL). RL (also known as entity resolution) is the process of detecting duplicates that refer to the same real-world entity in one or more datasets. The most critical step during the RL process is blocking, which reduces the quadratic complexity of the process by dividing the...

In breast cancer histology, there are three important features for tumor grading, where the proliferation score presents a key component. The mitotic count strategy is among the used methods to predict this score. However, this task is tedious and time consuming for pathologists. To simplify their work, there is a recognized need for computer-aided...

Deep learning is one of the most commonly used techniques in computer-aided diagnosis systems. Their exploitation for histopathological image analysis is important because of the complex morphology of whole slide images. However, the main limitation of these methods is the restricted number of available medical images, which can lead to an overfitt...

This chapter compares 4 variants of metaheuristics (RFA, EMVO, RPSO, and RBAT). The purpose is to test the impact of refinement on different types of metaheuristics (FA, MVO, PSO, and BAT). The refinement helps to enhance exploitation and to speed up the search process in multidimensional spaces. Moreover, it presents a powerful tool to solve diffe...

This article presents a new fine-tuning framework for histopathological images analysis. Despite the most common solutions where the ImageNet models are reused for image classification, thisresearch sets out to perform an intra-domain fine tuning between the trained models on the histopathological images. The purpose is to take advantage of the hyp...

Deep learning methods are characterized by their capacity to learn data representation compared to the traditional machine learning algorithms. However, these methods are prone to overfitting on small volumes of data. The objective of this research is to overcome this limitation by improving the generalization in the proposed deep learning framewor...

Feature selection is the process of identifying good performing combinations of significant features among many possibilities. This preprocess improves the classification accuracy and facilitates the learning task. For this optimization problem, the authors have used a metaheuristics approach. Their main objective is to propose an enhanced version...

The hybridization of metaheuristics got a lot of interest lately. The crucial step lies in the choice of the hybrid methods. The major purpose is to make a tradeoff between exploitation and exploration concepts to create a more robust method. Hybrid metaheuristics are used as a solution to many optimization problems such as feature selection. In th...

Deep learning in computer aided diagnostic systems has received a lot of interest lately. The motivation of deep learning compared to machine learning applications is their power to learn data representation. The purpose of this study is to propose an ensemble learning application based on MobileNet convolutional neural network. The proposed framew...

Instance selection and feature selection are important steps in the data mining process. They help reduce the excessive number of instances and features. The purpose of this reduction is to eliminate the noisy and redundant instances and features in order to improve the classifiers performance. Various related works in the literature proves that me...

The Multi-verse optimizer (MVO) is a recently proposed algorithm, inspired by the multi-verse theories, to solve optimization problems. Despite the efficiency of MVO, and its ability to create a balance between exploration and exploitation, like all other metaheuristics, it needs to be adapted to the field of the application. In this paper, we prop...