Ahmed Al Marouf

Dr. Ahmed Al Marouf

Department of Medicine, University of Alberta, Edmonton, Alberta, Canada
orcid

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

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

Representative Publications

View All Publications
19 February 2020 in IEEE Transactions on Computational Social Systems
Verified
Journal Article
Closed access
Ahmed Al MaroufAhmed Al Marouf
Ahmed Al Marouf
KHKamrul Hasan
Kamrul Hasan
HMHasan Mahmud
Hasan Mahmud
With the rapid growth of social media, users are getting involved in virtual socialism, generating a huge volume of textual and image contents. Considering the contents such as status updates/tweets and shared posts/retweets, liking other posts is reflecting the online behavior of the users. Predicting personality of a user from these digital footprints has become a computationally challenging problem. In a profile-based approach, utilizing the user-generated textual contents could be useful to reflect the personality in social media. Using huge number of features of different categories, such as traditional linguistic features (character-level, word-level, structural, and so on), psycholinguistic features (emotional affects, perceptions, social relationships, and so on) or social network features (network size, betweenness, and so on) could be useful to predict personality traits from social media. According to a widely popular personality model, namely, big-five-factor model (BFFM), the five factors are openness-to-experience, conscientiousness, extraversion, agreeableness, and neuroticism. Predicting personality is redefined as predicting each of these traits separately from the extracted features. Traditionally, it takes huge number of features to get better accuracy on any prediction task although applying feature selection algorithms may improve the performance of the model. In this article, we have compared the performance of five feature selection algorithms, namely the Pearson correlation coefficient (PCC), correlation-based feature subset (CFS), information gain (IG), symmetric uncertainly (SU) evaluator, and chi-squared (CHI) method. The performance is evaluated using the classic metrics, namely, precision, recall, f-measure, and accuracy as evaluation matrices.

29 August 2023 in IEEE Access
Verified
Journal Article
Open access
KHKhan Md Hasib
Khan Md Hasib
Sami AzamSami Azam
Sami Azam
AKAsif Karim
Asif Karim
Ahmed Al MaroufAhmed Al Marouf
Ahmed Al Marouf
F M Javed Mehedi ShamratF M Javed Mehedi Shamrat
F M Javed Mehedi Shamrat
JRJon Rokne
Jon Rokne
Searching, retrieving, and arranging text in ever-larger document collections necessitate more efficient information processing algorithms. Document categorization is a crucial component of various information processing systems for supervised learning. As the quantity of documents grows, the performance of classic supervised classifiers has deteriorated because of the number of document categories. Assigning documents to a predetermined set of classes is called text classification. It is utilized extensively in a wide range of data-intensive applications. However, the fact that real-world implementations of these models are plagued with shortcomings begs for more investigation. Imbalanced datasets hinder the most prevalent high-performance algorithms. In this paper, we propose an approach name multi-class Convolutional Neural Network (MCNN)-Long Short-Time Memory (LSTM), which combines two deep learning techniques, Convolutional Neural Network (CNN) and Long Short-Time Memory, for text classification in news data. CNN’s are used as feature extractors for the LSTMs on text input data and have the spatial structure of words in a sentence, paragraph, or document. The dataset is also imbalanced, and we use the Tomek-Link algorithm to balance the dataset and then apply our model, which shows better performance in terms of F1-score (98%) and Accuracy (99.71%) than the existing works. The combination of deep learning techniques used in our approach is ideal for the classification of imbalanced datasets with underrepresented categories. Hence, our method outperformed other machine learning algorithms in text classification by a large margin. We also compare our results with traditional machine learning algorithms in terms of imbalanced and balanced datasets.

6 January 2025 in MethodsX
Verified
Journal Article
Open access
ASAssaduzzaman
Assaduzzaman
PBPrayma Bishshash
Prayma Bishshash
ANAsraful Sharker Nirob
Asraful Sharker Nirob
Ahmed Al MaroufAhmed Al Marouf
Ahmed Al Marouf
JRJon G. Rokne
Jon G. Rokne
RAReda Alhajj
Reda Alhajj
Tomatoes are globally valued for their nutritional benefits and unique taste, playing a crucial role in agricultural productivity. Accurate diagnosis of tomato leaf diseases is vital to avoid ineffective treatments that can harm plants and ecosystems. While deep learning models excel in classifying these diseases, distinguishing subtle variations remains challenging. This study introduces XSE-TomatoNet, an enhanced version of EfficientNetB0, incorporating Squeeze-and-Excitation (SE) blocks and multi-scale feature fusion to boost classification performance. XSE-TomatoNet extracts multi-scale features, refines them with SE blocks, and merges them through Global Average Pooling, providing detailed and broad insights for precise disease classification. Our approach achieves an impressive accuracy of 99.11%, with 99% precision and recall, outperforming models like MobileNet and VGG19, especially when combined with data augmentation and ablation studies. The model achieved an average training accuracy of 99.41% and a validation accuracy of 98.88% in 10-fold cross-validation, showing strong generalization to unseen data. We also used LIME and SHAP for model interpretability, offering insights into the decision-making process, and employed Grad-CAM and Grad-CAM++ to visually highlight key areas in leaf images. Finally, the best model was integrated into a web-based system for practical use by tomato cultivators. • XSE-TomatoNet is an enhanced version of EfficientNetB0 which incorporates Squeeze-and-Excitation (SE) blocks and multi-scale feature fusion. • XSE-TomatoNet outperformed MobileNet (87.44%) and VGG-19 (95.50%), in terms of accuracy, achieving 99.41%. • Integration of interpretation using LIME and SHAP models gives higher level understanding of the diseases and employment of Grad-CAM and Grad-CAM++ shows visual representation of the diseased leaves.

13 November 2025 in Cancers
Verified
Journal Article
Open access
PDF file
Ahmed Al MaroufAhmed Al Marouf
Ahmed Al Marouf
JRJon Rokne
Jon Rokne
RAReda Alhajj
Reda Alhajj
Background: The combination of multi-omics data, including genomics, transcriptomics, and epigenomics, with medical imaging modalities (PET, CT, MRI, histopathology) has emerged in recent years as a promising direction for the advancement of precision oncology. Many researchers have contributed to this domain, exploring the multi-modality aspect of using both multi-omics and image data for better cancer identification, subtype classifications, cancer prognosis, etc. Methods: We present an umbrella review summarizing the state of the art in fusing imaging modalities with omics and artificial intelligence, focusing on existing reviews and meta-analyses. The analysis highlights early, late, and hybrid fusion strategies and their advantages and disadvantages, mainly in tumor classification, prognosis, and treatment prediction. We searched review articles until 25 May 2025 across multiple databases following PRISMA guidelines, with registration on PROSPERO (CRD420251062147). Results: After identifying 56 articles from different databases (i.e., PubMed, Scopus, Web of Science and Dimensions.ai), 35 articles were screened out based on the inclusion and exclusion criteria, keeping 21 studies for the umbrella review. Discussion: We investigated prominent fusion techniques in various contexts of cancer types and the role of machine learning in model performance enhancement. We address the problems of model generalizability versus interpretability within the clinical context and argue how these multi-modal issues can facilitate translating research into actual clinical scenarios. Conclusions: Lastly, we recommend future work to define clearer and more reliable validation criteria, address the need for integration of human clinicians with the AI system, and describe the trust issue with AI in cancer care, which requires more standardized approaches.

24 December 2022 in Bioengineering
Verified
Journal Article
Open access
PDF file
Ahmed Al MaroufAhmed Al Marouf
Ahmed Al Marouf
MMMozaharul Mottalib
Mozaharul Mottalib
RAReda Alhajj
Reda Alhajj
JRJon Rokne
Jon Rokne
OJOmar Jafarullah
Omar Jafarullah
The eye is generally considered to be the most important sensory organ of humans. Diseases and other degenerative conditions of the eye are therefore of great concern as they affect the function of this vital organ. With proper early diagnosis by experts and with optimal use of medicines and surgical techniques, these diseases or conditions can in many cases be either cured or greatly mitigated. Experts that perform the diagnosis are in high demand and their services are expensive, hence the appropriate identification of the cause of vision problems is either postponed or not done at all such that corrective measures are either not done or done too late. An efficient model to predict eye diseases using machine learning (ML) and ranker-based feature selection (r-FS) methods is therefore proposed which will aid in obtaining a correct diagnosis. The aim of this model is to automatically predict one or more of five common eye diseases namely, Cataracts (CT), Acute Angle-Closure Glaucoma (AACG), Primary Congenital Glaucoma (PCG), Exophthalmos or Bulging Eyes (BE) and Ocular Hypertension (OH). We have used efficient data collection methods, data annotations by professional ophthalmologists, applied five different feature selection methods, two types of data splitting techniques (train-test and stratified k-fold cross validation), and applied nine ML methods for the overall prediction approach. While applying ML methods, we have chosen suitable classic ML methods, such as Decision Tree (DT), Random Forest (RF), Naive Bayes (NB), AdaBoost (AB), Logistic Regression (LR), k-Nearest Neighbour (k-NN), Bagging (Bg), Boosting (BS) and Support Vector Machine (SVM). We have performed a symptomatic analysis of the prominent symptoms of each of the five eye diseases. The results of the analysis and comparison between methods are shown separately. While comparing the methods, we have adopted traditional performance indices, such as accuracy, precision, sensitivity, F1-Score, etc. Finally, SVM outperformed other models obtaining the highest accuracy of 99.11% for 10-fold cross-validation and LR obtained 98.58% for the split ratio of 80:20.

29 October 2024 in Technologies
Verified
Journal Article
Open access
PDF file
MDMonoronjon Dutta
Monoronjon Dutta
MSMD RASHEDUL ISLAM SUJAN
MD RASHEDUL ISLAM SUJAN
MMMayen Uddin Mojumdar
Mayen Uddin Mojumdar
NCNarayan Ranjan Chakraborty
Narayan Ranjan Chakraborty
Ahmed Al MaroufAhmed Al Marouf
Ahmed Al Marouf
RAReda Alhajj
Reda Alhajj
Classifying rice leaf diseases in agricultural technology helps to maintain crop health and to ensure a good yield. In this work, deep learning algorithms were, therefore, employed for the identification and classification of rice leaf diseases from images of crops in the field. The initial algorithmic phase involved image pre-processing of the crop images, using a bilateral filter to improve image quality. The effectiveness of this step was measured by using metrics like the Structural Similarity Index (SSIM) and the Peak Signal-to-Noise Ratio (PSNR). Following this, this work employed advanced neural network architectures for classification, including Cascading Autoencoder with Attention Residual U-Net (CAAR-U-Net), MobileNetV2, and Convolutional Neural Network (CNN). The proposed CNN model stood out, since it demonstrated exceptional performance in identifying rice leaf diseases, with test Accuracy of 98% and high Precision, Recall, and F1 scores. This result highlights that the proposed model is particularly well suited for rice leaf disease classification. The robustness of the proposed model was validated through k-fold cross-validation, confirming its generalizability and minimizing the risk of overfitting. This study not only focused on classifying rice leaf diseases but also has the potential to benefit farmers and the agricultural community greatly. This work highlights the advantages of custom CNN models for efficient and accurate rice leaf disease classification, paving the way for technology-driven advancements in farming practices.