Ali Mohsin Ahmed Al-Sabaawi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Ali Mohsin Ahmed Al-Sabaawi has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Ali Adel Dawood
Department of Anatomy College of Medicine University of Mosul Mosul Iraq
1 shared publication - Bassam Ismael Jasim Alzuhairy
MBChB / FIBMS / MRCP specialty certificate / ESMO. Department of Medicine, College of Medicine, Ninevah University. Mosul–Iraq
1 shared publication - Enass Waad Al-Hadidi
University of Mosul, Iraq
1 shared publication Hacer Karacan
Department of Computer Engineering, Faculty of Engineering, Gazi University, Ankara, Turkey
1 shared publication- Karrar Neamah
Faculty of Engineering, School of Computing, University Technology of Malaysia Johor Bahru, Skudai, Johor, Malaysia
1 shared publication - Mohsin Hasan Hussein
University of Kerbala, Iraq
1 shared publication
Latest Publications
- Open Access
- 30 June 2026
- Published by Politechnika Lubelska in Applied Computer Science
- Vol. 22 (2) , 67-84
Improving image retrieval using CNN with PCA and Optimized K-Means clustering
Content-based image retrieval (CBIR) systems play an important role in many applications, including object recognition, digital forensics, and biomedical research. Handling large volumes of digital images in our daily lives requires the development of an efficient CBIR system. In this paper, a comprehensive and unified approach for image retrieval has been introduced. It brings together deep learning, feature reduction, and clustering optimization in a way that hasn't been explored before. The pre-trained VGG16 model has been used to extract rich, high-level features from images. To make the process more efficient and focused, Principal Component Analysis (PCA) is applied to reduce the number of features while retaining the most important information with minimal runtime. What sets our work apart is the use of hybrid K-means clustering with Particle Swarm Optimization (PSO). Instead of relying on random initialization, PSO helps find better starting points and improves the overall clustering by guiding K-means away from local minima. These well-known techniques are combined into a single, coherent framework called VGG16-PCA-PSO-K-means. The effectiveness of the proposed method has been evaluated on the Corel 1K and UC Merced Land Use datasets using mean average precision (mAP), clustering purity, recall, F-score, NDCG, and runtime. The results of the experiments conducted indicate that the proposed system achieves higher precision and clustering purity than state-of-the-art methods. The improvement ranged from 5% to 18% across different feature set sizes, with mAP@10 of 97.5% for the 10-class retrieval of the Corel 1K and mAP@10 of 96.7% for the 21-class retrieval of the UC Merced Land Use datasets, using only 30 features. Furthermore, the results of the Kendall's W and Friedman tests confirmed that the (VGG16-PCA-PSO-K-means) model achieved a higher rank than other methods in the literature, with a significant difference, highlighting its effectiveness and robustness.
- 30 June 2026
- Published by Politechnika Lubelska in Applied Computer Science
- Vol. 22 (2) , 67-84
Open AccessImproving image retrieval using CNN with PCA and Optimized K-Means clustering
Content-based image retrieval (CBIR) systems play an important role in many applications, including object recognition, digital forensics, and biomedical research. Handling large volumes of digital images in our daily lives requires the development of an efficient CBIR system. In this paper, a comprehensive and unified approach for image retrieval has been introduced. It brings together deep learning, feature reduction, and clustering optimization in a way that hasn't been explored before. The pre-trained VGG16 model has been used to extract rich, high-level features from images. To make the process more efficient and focused, Principal Component Analysis (PCA) is applied to reduce the number of features while retaining the most important information with minimal runtime. What sets our work apart is the use of hybrid K-means clustering with Particle Swarm Optimization (PSO). Instead of relying on random initialization, PSO helps find better starting points and improves the overall clustering by guiding K-means away from local minima. These well-known techniques are combined into a single, coherent framework called VGG16-PCA-PSO-K-means. The effectiveness of the proposed method has been evaluated on the Corel 1K and UC Merced Land Use datasets using mean average precision (mAP), clustering purity, recall, F-score, NDCG, and runtime. The results of the experiments conducted indicate that the proposed system achieves higher precision and clustering purity than state-of-the-art methods. The improvement ranged from 5% to 18% across different feature set sizes, with mAP@10 of 97.5% for the 10-class retrieval of the Corel 1K and mAP@10 of 96.7% for the 21-class retrieval of the UC Merced Land Use datasets, using only 30 features. Furthermore, the results of the Kendall's W and Friedman tests confirmed that the (VGG16-PCA-PSO-K-means) model achieved a higher rank than other methods in the literature, with a significant difference, highlighting its effectiveness and robustness.
- Open Access
- 29 June 2026
- Published by Publishing house Uston Ltd in Анали Мечниковського Інституту
- No. 2,p. 52-63
Integrative Transcriptomic Immune Scoring and Network Topology Reveal ME13-Driven Spatial Heterogeneity in Glioblastoma
Glioblastoma (GBM) is one of the most aggressive and heterogeneous variants of brain cancer; it is associated with intricate connections between pathways of proliferative signaling and modulation of the immune system. The paper shall clarify how to explain immune-relevant transcriptional programs in GBM by an integrative gene co-expression network analysis. Powered by normalized expression data of 17,362 genes in the GSE16011 dataset, Weighted Gene Co-expression Network Analysis (WGCNA) was used to build functional coherent modules, and ME13 turned out to be the one most significant to the infiltration of immune cells (r = 0.82, p < 0.001). The gene-gene interaction network identified five crucial central hub genes, including GIMAP4, CD84, KRT23, KLF9, and RCBTB2, which are correlated with FOXP3 and CD8A expression, suggesting their possible regulatory roles in the dynamic processes of cytotoxic T cells and Tregs. To put the pattern of spatial activation into context, overlaying and GSVA scoring of Grad-CAM revealed a wide variety in the immune landscape within glioblastoma tissues. This multimodal combination illustrated the presence of ME13 in IL-17 and the p53 signaling pathway, as well as its value for immune stratification. The results not only provide the molecular framework for interpreting immune heterogeneity in GBM but also reveal the potential of hub genes as therapeutic targets and for future diagnostic profiling. Although some constraints will emerge due to the absence of experimental verification, the study paves the way for translation through multi-omics platforms and spatial immunogenomics.
- 29 June 2026
- Published by Publishing house Uston Ltd in Анали Мечниковського Інституту
- No. 2,p. 52-63
Open AccessIntegrative Transcriptomic Immune Scoring and Network Topology Reveal ME13-Driven Spatial Heterogeneity in Glioblastoma
Glioblastoma (GBM) is one of the most aggressive and heterogeneous variants of brain cancer; it is associated with intricate connections between pathways of proliferative signaling and modulation of the immune system. The paper shall clarify how to explain immune-relevant transcriptional programs in GBM by an integrative gene co-expression network analysis. Powered by normalized expression data of 17,362 genes in the GSE16011 dataset, Weighted Gene Co-expression Network Analysis (WGCNA) was used to build functional coherent modules, and ME13 turned out to be the one most significant to the infiltration of immune cells (r = 0.82, p < 0.001). The gene-gene interaction network identified five crucial central hub genes, including GIMAP4, CD84, KRT23, KLF9, and RCBTB2, which are correlated with FOXP3 and CD8A expression, suggesting their possible regulatory roles in the dynamic processes of cytotoxic T cells and Tregs. To put the pattern of spatial activation into context, overlaying and GSVA scoring of Grad-CAM revealed a wide variety in the immune landscape within glioblastoma tissues. This multimodal combination illustrated the presence of ME13 in IL-17 and the p53 signaling pathway, as well as its value for immune stratification. The results not only provide the molecular framework for interpreting immune heterogeneity in GBM but also reveal the potential of hub genes as therapeutic targets and for future diagnostic profiling. Although some constraints will emerge due to the absence of experimental verification, the study paves the way for translation through multi-omics platforms and spatial immunogenomics.
- Open Access
- 22 April 2020
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 15 (4) , e0231457
Exploiting implicit social relationships via dimension reduction to improve recommendation system performance
The development of Web 2.0 and the rapid growth of available data have led to the development of systems, such as recommendation systems (RSs), that can handle the information overload. However, RS performance is severely limited by sparsity and cold-start problems. Thus, this paper aims to alleviate these problems. To realize this objective, a new model is proposed by integrating three sources of information: a user-item matrix, explicit and implicit relationships. The core strategy of this study is to use the multi-step resource allocation (MSRA) method to identify hidden relations in social information. First, explicit social information is used to compute the similarity between each pair of users. Second, for each non-friend pair of users, the MSRA method is applied to determine the probability of their relation. If the probability exceeds a threshold, a new relationship will be established. Then, all sources are incorporated into the Singular Value Decomposition (SVD) method to compute the missing prediction values. Furthermore, the stochastic gradient descent technique is applied to optimize the training process. Additionally, two real datasets, namely, Last.Fm and Ciao, are utilized to evaluate the proposed method. In terms of accuracy, the experiment results demonstrate that the proposed method outperforms eight state-of-the-art approaches: Heats, PMF, SVD, SR, EISR-JC, EISR-CN, EISR-PA and EISR-RAI.
- 22 April 2020
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 15 (4) , e0231457
Open AccessExploiting implicit social relationships via dimension reduction to improve recommendation system performance
The development of Web 2.0 and the rapid growth of available data have led to the development of systems, such as recommendation systems (RSs), that can handle the information overload. However, RS performance is severely limited by sparsity and cold-start problems. Thus, this paper aims to alleviate these problems. To realize this objective, a new model is proposed by integrating three sources of information: a user-item matrix, explicit and implicit relationships. The core strategy of this study is to use the multi-step resource allocation (MSRA) method to identify hidden relations in social information. First, explicit social information is used to compute the similarity between each pair of users. Second, for each non-friend pair of users, the MSRA method is applied to determine the probability of their relation. If the probability exceeds a threshold, a new relationship will be established. Then, all sources are incorporated into the Singular Value Decomposition (SVD) method to compute the missing prediction values. Furthermore, the stochastic gradient descent technique is applied to optimize the training process. Additionally, two real datasets, namely, Last.Fm and Ciao, are utilized to evaluate the proposed method. In terms of accuracy, the experiment results demonstrate that the proposed method outperforms eight state-of-the-art approaches: Heats, PMF, SVD, SR, EISR-JC, EISR-CN, EISR-PA and EISR-RAI.
- Open Access
- 30 June 2017
- Published by University of Kufa in Journal of Kufa for Mathematics and Computer
- Vol. 4 (2) , 18-26
Image Retrieval Based on Chain Code Algorithm Using Color and Texture Features
The rapid growth of image retrieval has provided an efficient Content-Based Image Retrieval CBIR system to retrieve image accurately. In this paper, a precise retrieval result by exploiting color, texture and shape features is proposed. First, extract the features by color moment and (Hue, Saturation, Value HSV color space as a color feature, and then get the co-occurrence matrix as well as Discrete Wavelet Transform DWT for a texture feature. Chain codes algorithm, specifically chain code histogram, is then applied to obtain the codes of the shape feature. Second, collect all these features and store it in the database, where each record represents one image of the dataset. Similarity process is executed to find the images that are more similar to the query image, retrieved images ranked. The dataset applied in this study is WANG that includes 10 classes with each class containing 100 images. Experimental results have revealed that the proposed method outperformed the previous studies with an average of 0.824 in term of precision
- 30 June 2017
- Published by University of Kufa in Journal of Kufa for Mathematics and Computer
- Vol. 4 (2) , 18-26
Open AccessImage Retrieval Based on Chain Code Algorithm Using Color and Texture Features
The rapid growth of image retrieval has provided an efficient Content-Based Image Retrieval CBIR system to retrieve image accurately. In this paper, a precise retrieval result by exploiting color, texture and shape features is proposed. First, extract the features by color moment and (Hue, Saturation, Value HSV color space as a color feature, and then get the co-occurrence matrix as well as Discrete Wavelet Transform DWT for a texture feature. Chain codes algorithm, specifically chain code histogram, is then applied to obtain the codes of the shape feature. Second, collect all these features and store it in the database, where each record represents one image of the dataset. Similarity process is executed to find the images that are more similar to the query image, retrieved images ranked. The dataset applied in this study is WANG that includes 10 classes with each class containing 100 images. Experimental results have revealed that the proposed method outperformed the previous studies with an average of 0.824 in term of precision