A5018353705- EEG and Brain-Computer Interfaces
- Advanced Steganography and Watermarking Techniques
- ECG Monitoring and Analysis
- Blind Source Separation Techniques
- Chaos-based Image/Signal Encryption
- Speech and Audio Processing
- Digital Media Forensic Detection
- Speech Recognition and Synthesis
- Non-Invasive Vital Sign Monitoring
- Image and Signal Denoising Methods
- Music and Audio Processing
- Advanced Data Compression Techniques
- Emotion and Mood Recognition
- Neural Networks and Applications
- Phonocardiography and Auscultation Techniques
- Brain Tumor Detection and Classification
- Neuroscience and Neural Engineering
- Advanced Memory and Neural Computing
- Advanced Wireless Communication Techniques
- Vehicle License Plate Recognition
- Nutrition and Health in Aging
- Power Line Communications and Noise
- Advanced Adaptive Filtering Techniques
- Advanced Computing and Algorithms
- Machine Fault Diagnosis Techniques
Aberdeen Royal Infirmary
2020-2025
NHS Grampian
2023-2025
Umm al-Qura University
2025
University of Duhok
2023
Université Hassan II Mohammedia
2013-2022
University of Hassan II Casablanca
2018-2022
École Normale Supérieure de l'Enseignement Technique de Mohammedia
2021
University of Electronic Science and Technology of China
2014
Abstract Circulating tumor DNA (ctDNA) detection can predict clinical risk in early-stage tumors. However, applications are constrained by the sensitivity of clinically validated ctDNA approaches. NeXT Personal is a whole-genome-based, tumor-informed platform that has been analytically for ultrasensitive at 1–3 ppm with 99.9% specificity. Through an analysis 171 patients lung cancer from TRACERx study, we detected pre-operatively within 81% adenocarcinoma (LUAD), including 53% those...
Abstract The introduction of the International Association for Study Lung Cancer grading system has furthered interest in histopathological risk stratification lung adenocarcinoma. Complex morphology and high intratumoral heterogeneity present challenges to pathologists, prompting development artificial intelligence (AI) methods. Here we developed ANORAK (pyrAmid pooliNg crOss stReam Attention networK), encoding multiresolution inputs with an attention mechanism, delineate growth patterns...
Highly-pathogenic avian influenza virus (HPAIV) and Newcastle disease (NDV) are the two most important poultry viruses in world. Natural low-virulence NDV strains have been used as vaccines over past 70 years with proven track records. We previously developed a reverse genetics system to produce low-virulent vaccine strain LaSota from cloned cDNA. This allows us use vector for other pathogens.Here, we constructed recombinant NDVs (rNDVs) each of which expresses hemagglutinin (HA) gene HPAIV...
In this paper, we present a MR-Brain image classification system to classify given MR-brain as normal or abnormal. This first employs three feature extraction techniques namely, Gray-Level Co-Occurrence Matrix (GLCM), Local Binary Pattern (LBP) and Histogram of Oriented Gradient (HOG). The obtained vector each technique is passed through k-Nearest Neighbor (k-NN) classifier. resulting dissimilarity measure values the classifiers are combined then by fusion operator in order increase...
This paper proposes a novel Deep Learning technique for ECG beats classification. Unlike the traditional models, new Multi-Scale Wavelet Convolutional Neural Networks (MS-WCNN) is proposed to recognize automatically various cardiac arrhythmias. The MS-WCNN model incorporates one dimensional CNN and Stationary Transform (SWT) extract discriminative features from signal its wavelet sub-bands simultaneously. extracted are then merged using concatenation strategy. improves greatly learning...
The parasite Toxoplasma gondii is the source of toxoplasmosis. This infection can spread from mother to fetus during pregnancy, through contaminated food or water, contact with infected cat feces. study aims investigate whether chronic toxoplasmosis associated an increased risk type 2 diabetes mellitus (T2DM) in women. A total 274 samples were collected women reproductive age (18-55 years) as they relevant for both and T2DM. Serological testing was performed detect presence T. antibodies...
IGF-1 plays a role in the growth of multiple tumor types, including pancreatic cancer. also serves as factor for muscle. The impact therapeutic targeting on muscle mass is unknown.We evaluated at L3 patients enrolled randomized phase II study MK-0646 (M), monoclonal antibody directed against protein, with metastatic cancer (MPC). Two different doses M were tested, 5 and 10 mg/kg. We used Slice-o-matic (ver 4.3) software to segregate CT images into fat components measured area (cm(2))...
Because one of the key issues in improving performance Speech Emotion Recognition (SER) systems is choice an effective feature representation, most research has focused on developing a level fusion using large set features. In our study, we propose relatively low-dimensional that combines three features: baseline Mel Frequency Cepstral Coefficients (MFCCs), MFCCs derived from Discrete Wavelet Transform (DWT) sub-band coefficients are denoted as DMFCC, and pitch based Moreover, proposed...
Electrocardiogram (ECG) signal is a non-invasive technique that currently used to diagnose various types of cardiovascular diseases. However, ECG recording vulnerable different noises and artifacts make it very difficult obtain an accurate diagnosis. In this context, we propose novel denoising algorithm based on the deep Convolutional Denoising Auto-Encoder (CDAE) which requires minimal preprocessing steps, conserves important features. study, proposed CDAE specifically implemented remove...
In this paper, we propose a new ECG signal enhancement based on Ensemble Empirical Mode Decomposition (EEMD) and Higher Order Statistics (HOS). our scheme, the EEMD is used to decompose adaptively noisy into Intrinsic Functions (IMFs), novel criterion kurtosis proposed determine IMFs that contain sufficient information about QRS complex in which need be filtered. After that, two interval thresholding methods have been applied each selected IMF order reduce noise preserve complex. The final...
This paper presents a new semi-blind image watermarking system for medical applications.The scheme utilizes Singular Value Decomposition (SVD) and Discrete Wavelet Transform (DWT) to embed textual data into original images.In particular, text characters are encoded by Quadrature Amplitude Modulation (QAM-16).In order increase the security of protect then watermark from several attacks, embedded is submitted Arnold before inserting it host image.To evaluate performances scheme, images have...
In this paper, a novel method for removing white gaussian noise from raw ECG signals based on Stationary Wavelet Transform and Interval Thresholding is proposed. Unlike the classical existing denoising methods in wavelet domains, that apply simple thresholding to details coefficients, we considers extremums of each zero-crossings interval domain whole perform function, which can highlight preserve most clinical information about region QRS complex reconstructed signal. The performance...
Medical Imaging is currently a hot area of biomedical engineers, researchers and medical doctors as it extensively used in diagnosing human health by care institutes. The imaging equipment the device, which for better image processing highlighting important features. These images are affected with random noise during acquisition, analyzing transmission process. This results blurry visible low contrast. Image De-noising System (IDs) tool removing preserving data. de-noising one most...
The objective of this paper is to propose a new method for increasing the capacity watermark information in proposed audio watermarking system. basic idea exploit CDMA (Code Division Multiple Access) technique while considering multiple embedded watermarks as different users that share same channel which host signal. Once are embedded, we aim recover them from only received signal with low BER (Binary Error Rate) by applying UISA (Underdetermined Independent Subspaces Analysis) technique....