Mubashir Ahmad
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Mubashir Ahmad has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
Abdul Hye
Rasuta Energy LTD
1 shared publication- Afif Abdul Latiff
Faculty of Medicine, University Kebangsaan Malaysia, Kuala Lumpur 56000, Malaysia
1 shared publication - Anith Khairunnisa Ghazali
Faculty of Engineering and Technology, Multimedia University, Ayer Keroh 75450, Melaka, Malaysia
1 shared publication - Farshid Amirabdollahian
Robotics Research Group, University of Hertfordshire, Hatfield AL10 9AB, UK
1 shared publication Hafiz Alaka
1 shared publication- Heng Siong Lim
Faculty of Engineering and Technology, Multimedia University, Melaka 75450, Malaysia
1 shared publication
Latest Publications
- Open Access
- 12 March 2026
- Published by MDPI AG in Biomimetics
- Vol. 11 (3) , 206
A Trustable Spine Abnormalities Classification System Using ResNet50 and VGG16 Supported by Explainable Artificial Intelligence
Deep learning has been applied in various fields and has been proven to provide good results for classification tasks. However, there is limited understanding of a deep learning model’s decisions, so deep learning is commonly described as a black box. Applying deep learning for critical applications such as medical diagnostic process introduces trust issues. For the deep learning model to be trusted by the medical practitioners, the methods employed by the deep learning model must be seen to be aligned with the diagnostic process employed by the medical practitioners. Explainable methods such as Grad-CAM can be applied to improve the explainability of the deep learning models by providing an visual interpretation of the deep learning classification result decision process. In this study, two deep learning models, VGG16 and ResNet50 are trained using three training methods, one with randomly initialized weights, and two transfer learning methods, which are feature extraction and fine-tuning, to classify the spinal abnormalities based on X-ray images. The classification metrics results are compared and further analyses using Grad-CAM heatmaps are included. The models also evaluated using a stratified five-fold cross-validation, results revealed some disparity between the model’s accuracy and clinical relevance. The randomly initialized VGG16 obtained a classification accuracy of 93.79% but does not focus on clinically relevant regions. On the other hand, not only do the fine-tuned ResNet50 and VGG16 obtain high accuracies of 98.22% and 99.12%, but the heatmaps show that the models focus on more relevant regions. A comparison of the two models shows that the heatmaps produced by the fine-tuned ResNet50 are in more agreement with the clinical view than the fine-tuned VGG16. This study provides a useful reference for interpreting a deep learning-based classification result using explainable method particularly in spine abnormalities analysis with Grad-CAM.
- 12 March 2026
- Published by MDPI AG in Biomimetics
- Vol. 11 (3) , 206
Open AccessA Trustable Spine Abnormalities Classification System Using ResNet50 and VGG16 Supported by Explainable Artificial Intelligence
Deep learning has been applied in various fields and has been proven to provide good results for classification tasks. However, there is limited understanding of a deep learning model’s decisions, so deep learning is commonly described as a black box. Applying deep learning for critical applications such as medical diagnostic process introduces trust issues. For the deep learning model to be trusted by the medical practitioners, the methods employed by the deep learning model must be seen to be aligned with the diagnostic process employed by the medical practitioners. Explainable methods such as Grad-CAM can be applied to improve the explainability of the deep learning models by providing an visual interpretation of the deep learning classification result decision process. In this study, two deep learning models, VGG16 and ResNet50 are trained using three training methods, one with randomly initialized weights, and two transfer learning methods, which are feature extraction and fine-tuning, to classify the spinal abnormalities based on X-ray images. The classification metrics results are compared and further analyses using Grad-CAM heatmaps are included. The models also evaluated using a stratified five-fold cross-validation, results revealed some disparity between the model’s accuracy and clinical relevance. The randomly initialized VGG16 obtained a classification accuracy of 93.79% but does not focus on clinically relevant regions. On the other hand, not only do the fine-tuned ResNet50 and VGG16 obtain high accuracies of 98.22% and 99.12%, but the heatmaps show that the models focus on more relevant regions. A comparison of the two models shows that the heatmaps produced by the fine-tuned ResNet50 are in more agreement with the clinical view than the fine-tuned VGG16. This study provides a useful reference for interpreting a deep learning-based classification result using explainable method particularly in spine abnormalities analysis with Grad-CAM.
- Open Access
- 1 January 2022
- Published by Elsevier in SSRN Electronic Journal
A Novel Approach to Using Geospatial Data with Big-Data and Machine-Learning: Predicting Air Quality for Locations Without Air Quality Measuring Equipment
The air quality dataset derived through air quality sensors is a major dataset used in the development of air quality prediction models, but these datasets are not available for every location of interest in the world partly due to the cost of setting up the air quality sensors. As critical as the negative effect of air pollution is, many of the air quality prediction models are limited to the location whose air quality dataset is used to develop them. Thereby leaving locations without air quality datasets to be presented with speculative air quality forecasts. This situation increases the pollution exposure risk of vulnerable groups as their decision is based on speculative and inaccurate air quality prediction. This research is to investigate alternative methodology and datasets for the prediction of air quality in locations without air quality datasets or air quality sensors installed. To achieve this, we incorporate the geospatial dataset with other datasets such as traffic dataset, weather dataset and air quality dataset of other locations in the development of an air quality prediction model capable of predicting the air quality for locations without recorded air quality datasets or air quality sensor installed. This study presents a novel approach to using geospatial datasets in the development of an air quality prediction system.
- 1 January 2022
- Published by Elsevier in SSRN Electronic Journal
Open AccessA Novel Approach to Using Geospatial Data with Big-Data and Machine-Learning: Predicting Air Quality for Locations Without Air Quality Measuring Equipment
The air quality dataset derived through air quality sensors is a major dataset used in the development of air quality prediction models, but these datasets are not available for every location of interest in the world partly due to the cost of setting up the air quality sensors. As critical as the negative effect of air pollution is, many of the air quality prediction models are limited to the location whose air quality dataset is used to develop them. Thereby leaving locations without air quality datasets to be presented with speculative air quality forecasts. This situation increases the pollution exposure risk of vulnerable groups as their decision is based on speculative and inaccurate air quality prediction. This research is to investigate alternative methodology and datasets for the prediction of air quality in locations without air quality datasets or air quality sensors installed. To achieve this, we incorporate the geospatial dataset with other datasets such as traffic dataset, weather dataset and air quality dataset of other locations in the development of an air quality prediction model capable of predicting the air quality for locations without recorded air quality datasets or air quality sensor installed. This study presents a novel approach to using geospatial datasets in the development of an air quality prediction system.