A5069102475- Advanced Steganography and Watermarking Techniques
- Digital Media Forensic Detection
- Medical Imaging Techniques and Applications
- Radiomics and Machine Learning in Medical Imaging
- Digital and Cyber Forensics
- AI in cancer detection
- Medical Image Segmentation Techniques
- Brain Tumor Detection and Classification
- Advanced X-ray and CT Imaging
- Structural Behavior of Reinforced Concrete
- Advanced MRI Techniques and Applications
- Seismic Performance and Analysis
- Masonry and Concrete Structural Analysis
- Internet Traffic Analysis and Secure E-voting
- Image Retrieval and Classification Techniques
- Face and Expression Recognition
- Law, AI, and Intellectual Property
- Video Analysis and Summarization
- Imbalanced Data Classification Techniques
- ECG Monitoring and Analysis
- Epilepsy research and treatment
- Atrial Fibrillation Management and Outcomes
- Neonatal and fetal brain pathology
- Atomic and Subatomic Physics Research
- Face recognition and analysis
Institut Gustave Roussy
2023-2024
Inserm
2023-2024
Université Paris-Saclay
2023-2024
Children's Hospital of Los Angeles
2023-2024
Radiothérapie Moléculaire et Innovation Thérapeutique
2023-2024
Laboratoire Lorrain de Recherche en Informatique et ses Applications
2020-2021
Centre National de la Recherche Scientifique
2020-2021
Université de Lorraine
2020-2021
University of Bologna
2018-2020
Islamic Azad University Kerman
2014-2015
Atrial Fibrillation (AF) is the most common type of cardiac arrhythmia. Early diagnosis AF helps to improve therapy and prognosis. Machine Learning (ML) has been successfully applied effectiveness Computer-Aided Diagnosis (CADx) systems for detection. Presenting an explanation decision made by ML model considerable from cardiologists' point view, which decreases complexity can provide tangible information in their diagnosis. In this paper, a range techniques hand-crafted features based...
Background and PurposeAutomatic segmentation methods have greatly changed the radiotherapy workflow, but still need to be extended target volumes. In this paper, Deep Learning (DL) models were compared for Gross Tumor Volume (GTV) in locally advanced cervical cancer, a novel investigation into failure detection was introduced by utilizing radiomic features.Methods MaterialsWe trained eight DL (UNet, VNet, SegResNet, SegResNetVAE) 2D 3D segmentation. Ensembling individually during...
Clustering the images shared through social network (SN) platforms according to acquisition cameras embedded in smartphones is regarded as a significant task forensic investigations of cybercrimes. The sensor pattern noise (SPN) caused by camera imperfections during manufacturing process can be extracted from and used fingerprint smartphones. content compression performed SNs causes loss image details weakens SPN, making clustering even more challenging. In this paper, we present hybrid...
Due to a growing number of the computer networks in recent years, there has been an increasing interest intrusion detection systems (IDSs). In this paper we have proposed method applied instance selection from KDD CUP 99 dataset which is used for evaluating anomaly techniques. order determine performance reduction, feed forward neural network was trained by reduced classify normal or attack records dataset. The most obvious finding resulted study considerable increase accuracy rate obtained network.
The fast growth of Social Networks (SNs), amplified by the ever-increasing use smartphones, has intensified online cybercrimes. This trend accelerated digital investigations through SNs. In particular, camera Sensor Pattern Noise (SPN) uniquely characterizing each smartphone attracted a lot attention. this paper, we propose clustering and classification approach to achieve Smartphone Identification (SI) User Profiles Linking (UPL) across SNs provide investigators with significant findings in...
The popularity of social networks (SNs), amplified by the ever-increasing use smartphones, has intensified online cybercrimes. This trend accelerated digital forensics through SNs. One areas that received lots attention is camera fingerprinting, which each smartphone uniquely characterized. Hence, in this paper, we compare classification-based methods to achieve identification (SI) and user profile linking (UPL) within same or across different SNs, can provide investigators with significant...
In the last decades, Social Networks (SNs) have deeply changed interactions and habits of users that are also prone to create more than one profile on same SN. On flip side, fake profiles (i.e., impersonating profiles), become a considerable problem in digital investigations. this paper, we propose method for user resolution through cluster-based approach smartphone fingerprints extracted from images being posted SNs. The proposed is thus able detect profiles. To evaluate our approach, use...
Many existing structures require seismic retrofitting based on the latest findings and standards.Most of works presented in literature have considered effect different solutions behavior structures.Therefore, a comprehensive comparison these techniques is needed to select most effective one.Hence, this paper, we present comparative study useful RC structures, using recent updated versions standards design codes.The merits weaknesses retrofit available approaches are demonstrated, which...
Deep-learning-based automatic segmentation is widely used in radiation oncology to delineate organs-at-risk. Dual-energy CT (DECT) allows the reconstruction of enhanced contrast images that could help with manual and auto-delineation. This paper presents a performance evaluation commercial auto-segmentation software on image series generated by DECT. Different types DECT from seventy four head-and-neck (HN) patients were retrieved, including polyenergetic at different voltages [80 kV...
Response Spectrum Analysis (RSA) and modal combination techniques are widely used to estimate the peak response of structures subjected earthquake vibrations.This paper presents numerical modeling seismic analysis structures.The spectra based on Eurocode 8 is implemented for a five storey moment-resisting 3D structure earthquake-induced vibration.Eigenvector carried out determine undamped free vibration mode shapes frequencies.The lumped mass matrix, stiffness matrix full derived natural...
The infant hippocampus plays a pivotal role in early brain development and is linked to cognitive memory functions. Accurate delineation of the essential for studying normal detecting abnormalities associated with various neurodevelopmental disorders. In this paper, different deep neural network models were trained 3D-automatic segmentation based on cross-validation cohort T1-Weighted (T1W) images acquired from 100 subjects ground truth. tested another image 86 without Ensembling...