Muhammad Zafar Iqbal
Research Subjects
Muhammad Zafar Iqbal has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Ghazanfar Farooq Siddiqui
Department of Computer Sciences, Quaid-i-Azam University, Islamabad, Pakistan
2 shared publications - Ibrahim A. Hameed
Department of Mechanical Engineering and Technology Management, Norwegian University of Life Sciences, Ås, Akershus, 1433, Norway
2 shared publications - Abdul Qayyum
Imperial College London
1 shared publication - Adeel Ahmed
Department of Information Technology, The University of Haripur, Haripur, Pakistan
1 shared publication - Arcot Sowmya
University of New South Wales
1 shared publication Charles Kelly
Newcastle upon Tyne Hospitals NHS Foundation Trust
1 shared publication
Latest Publications
- 28 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 441-447
Hybrid Unsupervised Deformable Image Registration for 4D CT Lung Image Intra-modalities
Deformable image registration is inherently challenging, but when applied to lung images, it becomes even more challenging due to the significant deformations because of extreme inhalation and extreme exhalation phases of breathing. The greatest challenge arises when dealing with the complex, non-linear deformations that occur between the extreme inhale and exhale phases. Recent advancements in deep learning techniques show promise in handling these complex deformation patterns observed in 4D CT lung image modalities. In this study, we proposed an unsupervised hybrid image registration (HIR) framework designed to handle the non-linear deformations in 4DCT lung images without relying on ground truth data. This framework effectively learns the multi-resolution and multi-scale features from 4D CT image volumes to compute more precise displacement vector fields (DVFs). Our evaluation of the proposed HIR framework, conducted on the DIRLAB 4D CT lung image dataset, yielded an average Target Registration Error (TRE) value of $2.05 \pm 1.43$. We compared this performance with state-of-the-art unsupervised image registration methods, demonstrating the effectiveness and accuracy of our approach.- 28 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 441-447
Hybrid Unsupervised Deformable Image Registration for 4D CT Lung Image Intra-modalities
Deformable image registration is inherently challenging, but when applied to lung images, it becomes even more challenging due to the significant deformations because of extreme inhalation and extreme exhalation phases of breathing. The greatest challenge arises when dealing with the complex, non-linear deformations that occur between the extreme inhale and exhale phases. Recent advancements in deep learning techniques show promise in handling these complex deformation patterns observed in 4D CT lung image modalities. In this study, we proposed an unsupervised hybrid image registration (HIR) framework designed to handle the non-linear deformations in 4DCT lung images without relying on ground truth data. This framework effectively learns the multi-resolution and multi-scale features from 4D CT image volumes to compute more precise displacement vector fields (DVFs). Our evaluation of the proposed HIR framework, conducted on the DIRLAB 4D CT lung image dataset, yielded an average Target Registration Error (TRE) value of $2.05 \pm 1.43$. We compared this performance with state-of-the-art unsupervised image registration methods, demonstrating the effectiveness and accuracy of our approach.- 4 November 2022
- Published by Elsevier in Clinical Oncology
- Vol. 34, e1-e2
An audit of pattern of failure in patients with buccal mucosa/retromolar trigone carcinoma treated at a single UK centre
- 4 November 2022
- Published by Elsevier in Clinical Oncology
- Vol. 34, e1-e2
An audit of pattern of failure in patients with buccal mucosa/retromolar trigone carcinoma treated at a single UK centre
- Open Access
- 19 May 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 76518-76530
Improved Twofish Algorithm: A Digital Image Enciphering Application
With the growth of networks, the Internet of Things (IoT) and new cyber attacks pose threats to privacy and security, secure communication is therefore becoming one of the most crucial concerns. For this purpose, symmetric algorithms namel; The Rijndael algorithm, the Serpent algorithm, and the TWOFISH algorithm pay equal attention. In this paper, the TWOFISH algorithm’s mathematical complexity is improved by using substitution boxes (S-boxes) drawn from a multiplicative group of units of chain ring$\sum \nolimits _{i=0}^{7} {u^{i}F_{2}} $ . As these S-boxes have the property of having copious generators, they, therefore, produce a rich algebraic complexity. Moreover, the time complexity of the proposed work is modified by processing the 64-bit block throughout the process and reducing the number of subkeys. To measure the strength of the proposed algorithm, various standard color digital images, with a size of$256\times 256$ , are encrypted and tested. The computation speed of the encryption is compared to the standard TWOFISH algorithm’s speed and found that the newly designed algorithm is quite fast. For security analysis and quality assessment, various statistical tests are performed on the standard encrypted images. The results recommend that the proposed algorithm is a strong candidate for digital image encryption.- 19 May 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 76518-76530
Open AccessImproved Twofish Algorithm: A Digital Image Enciphering Application
With the growth of networks, the Internet of Things (IoT) and new cyber attacks pose threats to privacy and security, secure communication is therefore becoming one of the most crucial concerns. For this purpose, symmetric algorithms namel; The Rijndael algorithm, the Serpent algorithm, and the TWOFISH algorithm pay equal attention. In this paper, the TWOFISH algorithm’s mathematical complexity is improved by using substitution boxes (S-boxes) drawn from a multiplicative group of units of chain ring$\sum \nolimits _{i=0}^{7} {u^{i}F_{2}} $ . As these S-boxes have the property of having copious generators, they, therefore, produce a rich algebraic complexity. Moreover, the time complexity of the proposed work is modified by processing the 64-bit block throughout the process and reducing the number of subkeys. To measure the strength of the proposed algorithm, various standard color digital images, with a size of$256\times 256$ , are encrypted and tested. The computation speed of the encryption is compared to the standard TWOFISH algorithm’s speed and found that the newly designed algorithm is quite fast. For security analysis and quality assessment, various statistical tests are performed on the standard encrypted images. The results recommend that the proposed algorithm is a strong candidate for digital image encryption. - Open Access
- 1 January 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 8, 181893-181903
A Dynamic Three-Bit Image Steganography Algorithm for Medical and e-Healthcare Systems
Massive advances in internet infrastructure are impacting e-healthcare services compared to conventional means. Therefore, extra care and protection is needed for extremely confidential patient medical records. With this intention, we have proposed an enhanced image steganography method, to improve imperceptibility and data hiding capacity of stego images. The proposed Image Region Decomposition (IRD) method, embeds more secret information with better imperceptibility, in patient's medical images. The algorithm decomposes the grayscale magnetic resonance imaging (MRI) images into three unique regions: low-intensity, medium-intensity, and high-intensity. Each region is made up of k number of pixels, and in each pixel we operate the block of n least significant bits (LSBs), where 1 ≤ n ≤ 3. Four classes of MRI images of different dimensions are used for embedding. Data with different volumes are used to test the images for imperceptibility and verified with quality factors. The proposed IRD algorithm is tested for performance, on the set of brain MRI images using peak signal-to-noise ratio (PSNR), mean square error (MSE) and structural similarity (SSIM) index. The results elucidated that the MRI stego image is imperceptible, like the original cover image by adjusting 2 nd and 1 st LSBs in the low-intensity region. Our proposed steganography technique provides a better average PSNR (49.27), than other similar methods. The empirical results show that the proposed IRD algorithm, significantly improves the imperceptibility and data embedding capacity, compared to the existing state-of-the-art methods.
- 1 January 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 8, 181893-181903
Open AccessA Dynamic Three-Bit Image Steganography Algorithm for Medical and e-Healthcare Systems
Massive advances in internet infrastructure are impacting e-healthcare services compared to conventional means. Therefore, extra care and protection is needed for extremely confidential patient medical records. With this intention, we have proposed an enhanced image steganography method, to improve imperceptibility and data hiding capacity of stego images. The proposed Image Region Decomposition (IRD) method, embeds more secret information with better imperceptibility, in patient's medical images. The algorithm decomposes the grayscale magnetic resonance imaging (MRI) images into three unique regions: low-intensity, medium-intensity, and high-intensity. Each region is made up of k number of pixels, and in each pixel we operate the block of n least significant bits (LSBs), where 1 ≤ n ≤ 3. Four classes of MRI images of different dimensions are used for embedding. Data with different volumes are used to test the images for imperceptibility and verified with quality factors. The proposed IRD algorithm is tested for performance, on the set of brain MRI images using peak signal-to-noise ratio (PSNR), mean square error (MSE) and structural similarity (SSIM) index. The results elucidated that the MRI stego image is imperceptible, like the original cover image by adjusting 2 nd and 1 st LSBs in the low-intensity region. Our proposed steganography technique provides a better average PSNR (49.27), than other similar methods. The empirical results show that the proposed IRD algorithm, significantly improves the imperceptibility and data embedding capacity, compared to the existing state-of-the-art methods.