Zaied Alhaj

Dr. Zaied Alhaj

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
21 February 2026 in Scientific Reports
Verified
Journal Article
Open access
PDF file
AAAzzam Altairi
Azzam Altairi
ZAZaied Alhaj
Zaied Alhaj
MAMohammed Alsharafi
Mohammed Alsharafi
YZYusuf Zeren
Yusuf Zeren
Graph–theoretic degree–based descriptors play a central role in chemoinformatics and QSPR/QSAR modelling, yet most classical indices either focus purely on vertex degrees or treat bond contributions in a purely multiplicative way. In this work we introduce and systematically study a new family of modified bond-based indices in which each edge \(uv\in E(G)\) is weighted by a local bond factor \(\mathcal {T}(uv)={\phi _G(\textrm{u})}+{\phi _G(\textrm{v})}-2\) in the denominator, coupled with a vertex kernel in the numerator. This construction yields modified versions of the first and second Zagreb indices, the Forgotten and Yemen indices, several connectivity-type descriptors (product, sum, Nirmala, ABC, CAB, GA, harmonic, and misbalance prodeg), as well as Sombor- and Dharwad-type bond indices. We first present a unified edge–partition representation for any symmetric kernel, expressing each modified index as a finite sum over degree classes \(E_{(a,b)}(G)\). This framework allows us to derive closed-form expressions for all sixteen modified bond-based indices on a broad collection of benchmark families: paths \(P_{\textrm{n}}\), cycles \(C_{\textrm{n}}\), complete graphs \(K_{\textrm{n}}\), complete bipartite graphs \(K_{m,\textrm{n}}\), stars \(S_{\textrm{n}}\), friendship graphs \(F_{\textrm{n}}\), wheels \(W_{\textrm{n}}\), book graphs \(B_{\textrm{n}}\), Dutch windmill graphs \(D_{\textrm{n}}^{(m)}\), and hypercubes \(Q_d\). The resulting tables reveal clear asymptotic growth patterns and highlight which structures are extremal for the modified descriptors. Moreover, we obtain sharp degree–extreme bounds for a representative subset of the indices in terms of the order \(\textrm{n}\), size m, and the minimum and maximum degrees \(\delta\) and \(\Delta\), with equality characterizing regular graphs. The proposed modified bond-based indices thus provide a flexible and analytically tractable family of descriptors that couple vertex and bond information in a novel way, and are well suited as structured features for modern chemoinformatics and graph-based machine-learning models on molecular graphs. Finally, to demonstrate predictive utility in a hypothesis-driven setting, we further benchmark these \({}^{m}\textrm{BI}\) descriptors within a large multi-task QSAR/QSPR pipeline on 3,219 ChEMBL antibacterial molecules across ten continuous properties using a heterogeneous model zoo under three descriptor scenarios, where the combined descriptors scenario achieves the best overall generalisation (Macro Test \(R^2 = 0.861\); Global zRMSE \(= 0.373\)), improving upon the Physicochemical descriptors scenario (Macro Test \(R^2 = 0.852\); Global zRMSE \(= 0.385\)).

31 March 2026 in IEEE Access
Verified
Journal Article
Open access
ZAZaied Alhaj
Zaied Alhaj
AAAzzam Altairi
Azzam Altairi
Endoscopic image classification is a critical component of computer-aided diagnosis but remains hindered by significant challenges, including inter-patient domain shifts, data scarcity, and a conventional reliance on spatial features that often fail to capture subtle, high-frequency textural anomalies inherent to early-stage pathologies. To mitigate these limitations, we propose SSOMT (Spatial-Spectral Operator Mixture Transformer), a novel dual-stream architecture that explicitly integrates learnable multi-scale frequency-domain representations with hierarchical spatial features through a principled, contentadaptive fusion mechanism. The spatial stream leverages a pretrained DaViT backbone to extract rich local semantic information, while a parallel spectral encoder utilizes learnable Fourier neural operators to perform decomposition at multiple patch scales, isolating amplitude and phase cues that are robust to illumination variations. To effectively synthesize these complementary modalities, we introduce a heterogeneous Mixture-of-Experts (MoE) module that dynamically routes spatial and spectral tokens through learned gating networks. This allows the model to adaptively weigh the importance of texture (spectral) versus shape (spatial) for each specific input, reinforced by MixStyle regularization to improve generalization across diverse imaging domains. Extensive evaluation on two benchmark endoscopic datasets demonstrates the strong performance of the proposed approach. SSOMT achieves 99.43% ± 0.57% accuracy on bladder tissue classification, corresponding to a +0.40 percentage-point gain over the strongest listed baseline, and 94.08% ± 0.57% accuracy on gastrointestinal classification, showing competitive performance under the adopted five-fold protocol. Crucially, the model achieves 100% recall for benign tissue and 99.79% recall for high-grade cancer, reducing missed diagnoses. These results show that integrating spectral operators with spatial transformers yields a robust, clinically reliable framework for precise endoscopic tissue characterization, with stable performance across two benchmark endoscopic datasets overall.

29 June 2026 in Journal of King Saud University Computer and Information Sciences
Verified
Journal Article
Open access
PDF file
RARazan Almnawer
Razan Almnawer
ZAZaied Alhaj
Zaied Alhaj
MOMahmut Ozturk
Mahmut Ozturk
Accurate brain tumor MRI decision support requires both reliable lesion localization and consistent tumor-type prediction, yet many deep learning pipelines optimize segmentation and classification separately or with limited task interaction. We propose CrossTaskFormer, a unified multi-task framework that coordinates these objectives via learnable bidirectional cross-task attention, enabling structured exchange of localization cues and subtype-discriminative semantics across decoder stages. The model uses transfer learning for multi-scale feature extraction, incorporates an attention-based bottleneck to capture global context, and applies progressive cross-task fusion to improve boundary stability and diagnostic consistency. We evaluate CrossTaskFormer under a deployment-oriented inference protocol that includes ensembling, test-time augmentation, and morphological mask refinement. On the BRISC contrast-enhanced MRI dataset spanning four diagnostic categories, CrossTaskFormer achieves 99.30% classification accuracy while maintaining strong segmentation performance (Dice = 0.8925, IoU = 0.8122) under the full inference pipeline, while the raw base model already attains 99.23% classification accuracy with Dice = 0.8866 and IoU = 0.8032. Error analysis indicates that residual misclassifications are infrequent and largely confined to clinically plausible inter-class confusions, whereas segmentation degradations occur primarily in small-lesion cases where overlap metrics are inherently sensitive to minor boundary deviations. These results suggest that explicit cross-task coupling can simplify integrated pipelines while producing coherent outputs promising for integrated neuro-oncology decision support under controlled benchmarking conditions.

18 April 2026 in Scientific Reports
Verified
Journal Article
Open access
ZAZaied Alhaj
Zaied Alhaj
MOMahmut Ozturk
Mahmut Ozturk
MAMohammed Alsharafi
Mohammed Alsharafi
Transfer learning with convolutional neural networks (CNNs) for breast ultrasound tumor classification can be susceptible to overfitting and unstable generalization when training data are limited, noisy, and heterogeneous. Sharpness-Aware Minimization (SAM) is an optimizer-level training strategy that encourages convergence to flatter regions of the loss landscape and may improve held-out performance without modifying model architecture. In this study, we systematically evaluated SAM for breast ultrasound image classification across seventeen CNN architectures spanning five families–VGG/AlexNet, ResNet, DenseNet, MobileNet, and EfficientNet–using four experiments on two datasets (BUSI and BUS-UCLM). We compared standard Adam versus SAM+Adam and standard stochastic gradient descent (SGD) versus SAM+SGD under matched experimental settings. On BUSI, SAM improved mean validation accuracy by +2.30 percentage points with Adam and by +2.51 percentage points with SGD. On BUS-UCLM, the corresponding mean gains were +2.27 and +2.55 percentage points. The largest individual improvement was observed for MobileNetV3 Small under SAM+SGD (+4.86 percentage points on BUSI), while the highest validation accuracy was achieved by VGG19 with SAM+SGD (96.20% on BUSI). Improvements were also observed in precision, recall, and F1-score, and paired statistical testing across architectures showed consistently positive effects with large effect sizes. Computational analysis showed moderate training-time overhead but no architectural inference overhead, supporting the practical feasibility of SAM. Overall, these findings indicate that SAM is a useful and broadly effective training enhancement for transfer-learned breast ultrasound classification across multiple CNN families and two independent datasets.

30 December 2024 in University of Science and Technology Journal for Engineering and Technology
Verified
Journal Article
Open access
PDF file
HAHusam Y. Al-Hetari
Husam Y. Al-Hetari
NANoman Q. Al-Naggar
Noman Q. Al-Naggar
Abdu AlameriAbdu Alameri
Abdu Alameri
MAMahmoud A. Al-Rumaima
Mahmoud A. Al-Rumaima
MKMuhammad Nomani Kabir
Muhammad Nomani Kabir
ZAZaied Alhaj
Zaied Alhaj
Undesirable noise and interference present a substantial obstacle in the process of obtaining biomedical signals for analysis and diagnosis. It is necessary to reduce the noise and enhance signal quality for many designs of medical devices such as Mechanical Ventilators (MV), patient monitors and overall biosignal processing. To contribute for addressing this problem, and for medical applications such as MV which are used for critical diseases such as pneumonia, COVID 19 etc. This paper, conducted comparative study to evaluate the performance of a 3-point Moving Average Filter (MAF) and a 5-point MAF in order to reducing noise and improving signal quality in MVs which are used in medical applications. MATLAB was utilized to perform filter approximation, realization and analysis. LabVIEW was employed for pressure signal data acquisition, implementation, and Signal-to-Noise Ratio (SNR) calculation. Experimental data, which needed for filters implementation, were obtained by running a MV connected to an artificial lung and a reference device which is used to measure pressure signal. In general, 3 point presented better SNR against 5 points MAF for pressure signal required in MVs. This approach can potentially enhance healthcare quality by providing better quality of biomedical signals which are used in medical devices.

7 August 2026 in Scientific Reports
Verified
Journal Article
Open access
PDF file
ZAZaied Alhaj
Zaied Alhaj
MOMahmut Ozturk
Mahmut Ozturk
Progress in computational pathology is limited not only by model architecture, but also by evaluation design. In breast histopathology benchmarks such as BreaKHis, multiple correlated image tiles often originate from the same patient; consequently, image-wise splits can inflate performance by allowing patient-specific staining, preparation, or morphology signatures to appear in both training and test sets. At the same time, many classifiers emphasize local texture patterns, whereas diagnostic interpretation often depends on spatial relations among glands, stroma, tumor nests, and tissue compartments. We address these two issues with a patient-level topology-aware graph transformer that preserves spatial backbone features, converts each image into a grid of regional tissue tokens, and applies dense multi-head attention as a learned weighted graph over tissue regions. Evaluation was organized into three acts. Act I reproduced the commonly used image-wise protocol and achieved high tile-level performance, with 99.6% binary accuracy and 97.1% eight-class accuracy. Act II enforced patient-wise splitting while retaining tile-level scoring; eight-class accuracy decreased to 58.6% and binary accuracy to 92.1%, demonstrating the effect of patient-disjoint evaluation. Act III preserved patient-wise folds and aggregated tile evidence into calibrated patient-level predictions. Across magnifications, patient-level multiclass accuracy ranged from 71.5% to 76.3%, macro-F1 from 50.8% to 59.8%, Top-3 accuracy from 83.5% to 87.0%, and binary patient-level accuracy from 93.8% to 97.5%. Comparisons with plain Swin aggregation, graph-removal variants, explicit k-nearest-neighbor and grid-graph baselines, multiple-instance learning baselines, RetCCL, and CTransPath showed that the full graph model provided the strongest primary patient-level classification performance among the tested configurations. These findings support a shift from image-level leaderboards toward leakage-aware, patient-level reporting on BreaKHis, while external validation and pathologist assessment of attention maps remain necessary before broader deployment claims.