Uzair Iqbal Janjua
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
Uzair Iqbal Janjua has a total of 38 co-authors. Here are the top co-authors by number of shared publications:
- Tahir Mustafa Madni
Department of Computer Science, COMSATS University Islamabad, Islamabad, Pakistan
11 shared publications - Ahmad Raza Shahid
School of Computing, National University of Computer and Emerging Sciences, Islamabad, Pakistan
3 shared publications Fozia Anwar
COMSATS University Islamabad Islamabad Pakistan
3 shared publicationsMuhammad Faisal Cheema
Department of Computer Science, FAST National University, Islamabad, Pakistan
3 shared publications- Adnan Akhunzada
University of Doha for Science and Technology
2 shared publications - Jafreezal Jaafar
Department of Computer and Information Sciences, Universiti Teknologi Petronas, Malaysia
2 shared publications
Latest Publications
- 3 June 2024
- Published by Springer Nature in Journal of Imaging Informatics in Medicine
- Vol. 37 (6) , 2825-2837
CSAMDT: Conditional Self Attention Memory-Driven Transformers for Radiology Report Generation from Chest X-Ray
A radiology report plays a crucial role in guiding patient treatment, but writing these reports is a time-consuming task that demands a radiologist’s expertise. In response to this challenge, researchers in the subfields of artificial intelligence for healthcare have explored techniques for automatically interpreting radiographic images and generating free-text reports, while much of the research on medical report creation has focused on image captioning methods without adequately addressing particular report aspects. This study introduces a Conditional Self Attention Memory-Driven Transformer model for generating radiological reports. The model operates in two phases: initially, a multi-label classification model, utilizing ResNet152 v2 as an encoder, is employed for feature extraction and multiple disease diagnosis. In the second phase, the Conditional Self Attention Memory-Driven Transformer serves as a decoder, utilizing self-attention memory-driven transformers to generate text reports. Comprehensive experimentation was conducted to compare existing and proposed techniques based on Bilingual Evaluation Understudy (BLEU) scores ranging from 1 to 4. The model outperforms the other state-of-the-art techniques by increasing the BLEU 1 (0.475), BLEU 2 (0.358), BLEU 3 (0.229), and BLEU 4 (0.165) respectively. This study’s findings can alleviate radiologists’ workloads and enhance clinical workflows by introducing an autonomous radiological report generation system.
- 3 June 2024
- Published by Springer Nature in Journal of Imaging Informatics in Medicine
- Vol. 37 (6) , 2825-2837
CSAMDT: Conditional Self Attention Memory-Driven Transformers for Radiology Report Generation from Chest X-Ray
A radiology report plays a crucial role in guiding patient treatment, but writing these reports is a time-consuming task that demands a radiologist’s expertise. In response to this challenge, researchers in the subfields of artificial intelligence for healthcare have explored techniques for automatically interpreting radiographic images and generating free-text reports, while much of the research on medical report creation has focused on image captioning methods without adequately addressing particular report aspects. This study introduces a Conditional Self Attention Memory-Driven Transformer model for generating radiological reports. The model operates in two phases: initially, a multi-label classification model, utilizing ResNet152 v2 as an encoder, is employed for feature extraction and multiple disease diagnosis. In the second phase, the Conditional Self Attention Memory-Driven Transformer serves as a decoder, utilizing self-attention memory-driven transformers to generate text reports. Comprehensive experimentation was conducted to compare existing and proposed techniques based on Bilingual Evaluation Understudy (BLEU) scores ranging from 1 to 4. The model outperforms the other state-of-the-art techniques by increasing the BLEU 1 (0.475), BLEU 2 (0.358), BLEU 3 (0.229), and BLEU 4 (0.165) respectively. This study’s findings can alleviate radiologists’ workloads and enhance clinical workflows by introducing an autonomous radiological report generation system.
- Open Access
- 28 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 92715-92734
An Empirical Study to Investigate the Impact of Factors Influencing Knowledge Sharing in Virtual Teams
This study examines the relationship between virtual team performance and its critical explaining factors, i.e., individual, team, organizational, and technological factors, by exploring the mediating effect of knowledge sharing. For this purpose, a conceptual framework is proposed based on an in-depth Systematic Literature Review (SLR) analysis. Next, data is collected through a survey from 252 IT practitioners working in Pakistan’s small and medium-sized enterprises (SMEs). Statistical tests on collected data are applied in statistical package for social sciences (SPSS) and smart partial least squares (SmartPLS) to evaluate the proposed conceptual framework empirically. Finally, 19 industrial and 5 academic experts verified the framework. There is a consensus among researchers and practitioners regarding the factors that contribute to the explanation of knowledge sharing and virtual team performance. Thus, the current study’s outcome may facilitate the management of software organizations to improvise their knowledge sharing infrastructure, make the development process more efficient, avoid project failure and increase the productivity of virtual teams.
- 28 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 92715-92734
Open AccessAn Empirical Study to Investigate the Impact of Factors Influencing Knowledge Sharing in Virtual Teams
This study examines the relationship between virtual team performance and its critical explaining factors, i.e., individual, team, organizational, and technological factors, by exploring the mediating effect of knowledge sharing. For this purpose, a conceptual framework is proposed based on an in-depth Systematic Literature Review (SLR) analysis. Next, data is collected through a survey from 252 IT practitioners working in Pakistan’s small and medium-sized enterprises (SMEs). Statistical tests on collected data are applied in statistical package for social sciences (SPSS) and smart partial least squares (SmartPLS) to evaluate the proposed conceptual framework empirically. Finally, 19 industrial and 5 academic experts verified the framework. There is a consensus among researchers and practitioners regarding the factors that contribute to the explanation of knowledge sharing and virtual team performance. Thus, the current study’s outcome may facilitate the management of software organizations to improvise their knowledge sharing infrastructure, make the development process more efficient, avoid project failure and increase the productivity of virtual teams.
- Open Access
- 1 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 99499-99518
The Impact of Mitigation Strategies for Socio-Cultural Distance Issues in GSD: An Empirical Study
Global Software Development (GSD) offers several benefits to software development organizations, including reduced development costs, the availability of low-wage and highly skilled employees, and an improved marketplace. Meanwhile, it faces severe communication, coordination, and control issues. The most important of these is the communication issue which is further categorized into socio-cultural, temporal, and geographical issues. Among these issues, researchers believe the socio-cultural issue is the most critical factor and, if not mitigated properly, may lead to software project failure. Although, in the past, many studies have identified socio-cultural distance-related issues, and a few studies proposed mitigation strategies. However, studies have yet to be carried out to prioritize and empirically evaluate all mitigation strategies. Thus the main objectives of this study are: a) to identify socio-cultural distance issues and mitigation strategies through a Systematic Literature Review (SLR), b) to empirically evaluate the impact of identified mitigation strategies on identified socio-cultural distance issues through a survey, and c) to prioritize effective mitigation strategies through a recommended Analytical Hierarchy Process (AHP). A total of six socio-cultural issues and twenty-eight mitigation strategies are identified from the SLR and survey. Out of which, seven mitigation strategies are most effective. This study’s findings will help software organizations to overcome socio-cultural distance issues by using the highest priority mitigation strategies to reduce losses.
- 1 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 99499-99518
Open AccessThe Impact of Mitigation Strategies for Socio-Cultural Distance Issues in GSD: An Empirical Study
Global Software Development (GSD) offers several benefits to software development organizations, including reduced development costs, the availability of low-wage and highly skilled employees, and an improved marketplace. Meanwhile, it faces severe communication, coordination, and control issues. The most important of these is the communication issue which is further categorized into socio-cultural, temporal, and geographical issues. Among these issues, researchers believe the socio-cultural issue is the most critical factor and, if not mitigated properly, may lead to software project failure. Although, in the past, many studies have identified socio-cultural distance-related issues, and a few studies proposed mitigation strategies. However, studies have yet to be carried out to prioritize and empirically evaluate all mitigation strategies. Thus the main objectives of this study are: a) to identify socio-cultural distance issues and mitigation strategies through a Systematic Literature Review (SLR), b) to empirically evaluate the impact of identified mitigation strategies on identified socio-cultural distance issues through a survey, and c) to prioritize effective mitigation strategies through a recommended Analytical Hierarchy Process (AHP). A total of six socio-cultural issues and twenty-eight mitigation strategies are identified from the SLR and survey. Out of which, seven mitigation strategies are most effective. This study’s findings will help software organizations to overcome socio-cultural distance issues by using the highest priority mitigation strategies to reduce losses.
- 14 January 2023
- Published by Wiley in International Journal of Imaging Systems and Technology
- Vol. 33 (3) , 866-880
An automatic gastric polyp detection technique using deep learning
Over the last few years, researchers have focused on computer‐aided polyp detection in gastroscopy. Deep learning (DL) has shown great promise for polyps' identification. The most exceptional contribution of DL methods in gastroenterology is their ability to identify polyps quickly and accurately using convolution neural network. Nonetheless, despite significant advancements, automatic detection of small polyps remains a challenging and complex task. Furthermore, due to multiple pooling operations, the features of small polyps are lost, resulting in low detection accuracy. This paper proposes an efficient object detection method for polyp detection using gastric images to address this issue. A single‐shot multi‐box detector (SSD) was combined with the feature extractor VGG‐16, and the Refined Map Block (RMB) was integrated into SSD's high‐resolution feature maps to get more semantic information. The RMB output was used as the input to the successive layers. The RMB comprises of attention cascade and feature map concatenation cascade. The attention cascade improved the localization accuracy, while the feature map concatenation cascade improved the classification accuracy. Using the former, the proposed attention‐based SSD for gastric polyps (ASSD‐GPNet) model focused on the specific information, a polyp, rather than the background. Furthermore, the feature map concatenation cascade adds semantic information while reducing computational complexity. The output of these two cascades was combined to produce a refined feature map that enhances the detection of small polyps. The model was trained and tested on 1970 gastric images and Pascal VOC07 + 12. Image augmentation was applied to increase the training data of gastric images to reduce overfitting and skip connections were used to overcome the vanishing gradient problem. Overall, the experimental results demonstrated that the proposed model outperformed than compared models in both medical and natural images. The ASSD‐GPNet obtained mean average precision (mAP) of 94.2% on gastric images and 76.9% on Pascal VOC.
- 14 January 2023
- Published by Wiley in International Journal of Imaging Systems and Technology
- Vol. 33 (3) , 866-880
An automatic gastric polyp detection technique using deep learning
Over the last few years, researchers have focused on computer‐aided polyp detection in gastroscopy. Deep learning (DL) has shown great promise for polyps' identification. The most exceptional contribution of DL methods in gastroenterology is their ability to identify polyps quickly and accurately using convolution neural network. Nonetheless, despite significant advancements, automatic detection of small polyps remains a challenging and complex task. Furthermore, due to multiple pooling operations, the features of small polyps are lost, resulting in low detection accuracy. This paper proposes an efficient object detection method for polyp detection using gastric images to address this issue. A single‐shot multi‐box detector (SSD) was combined with the feature extractor VGG‐16, and the Refined Map Block (RMB) was integrated into SSD's high‐resolution feature maps to get more semantic information. The RMB output was used as the input to the successive layers. The RMB comprises of attention cascade and feature map concatenation cascade. The attention cascade improved the localization accuracy, while the feature map concatenation cascade improved the classification accuracy. Using the former, the proposed attention‐based SSD for gastric polyps (ASSD‐GPNet) model focused on the specific information, a polyp, rather than the background. Furthermore, the feature map concatenation cascade adds semantic information while reducing computational complexity. The output of these two cascades was combined to produce a refined feature map that enhances the detection of small polyps. The model was trained and tested on 1970 gastric images and Pascal VOC07 + 12. Image augmentation was applied to increase the training data of gastric images to reduce overfitting and skip connections were used to overcome the vanishing gradient problem. Overall, the experimental results demonstrated that the proposed model outperformed than compared models in both medical and natural images. The ASSD‐GPNet obtained mean average precision (mAP) of 94.2% on gastric images and 76.9% on Pascal VOC.
- Open Access
- 29 November 2022
- Published by Karachi Institute of Economics and Technology in KIET Journal of Computing and Information Sciences
- Vol. 6 (1) , 1-27
Temporal Distance Issues and their Mitigation Strategies in GSD: An Empirical Study
GSD has several advantages that make it easier for software companies to follow this trend, like skillful workers, lower costs, and an improved marketplace. To achieve the benefits mentioned above, distributed teams must communicate effectively. Communication is a major problem for development, especially in GSD. Communication issues are classified based on temporal, geographical, and sociocultural distances. The temporal distance problem affects more GSD operations than the other hazards combined. Organizations can mitigate these obstacles and attain the desired quality by employing mitigation strategies while adhering to budget and time constraints. The authors of this study conducted a systematic literature review to identify temporal distance issues and their respective mitigation strategies to develop a conceptual framework. Additionally, a questionnaire interview with closed-ended questions was conducted. Without an empirical study, it is difficult to determine the precise relationship between temporal distance issues and mitigation strategies. The primary objectives of this study are to experimentally evaluate the proposed conceptual framework and the effects of selected risk mitigation measures. This study identified forty-four mitigation techniques and nine temporal distance challenges with a focus on a GSD-based environment, empirical research was conducted using data collected from Pakistani software companies. The findings of this study indicate, with a p-value of 0.01, that various mitigation techniques significantly impact various temporal distance issues. Based on the findings, software organizations can employ appropriate mitigation techniques to address temporal distance issues and reduce the failure rate of software projects.
- 29 November 2022
- Published by Karachi Institute of Economics and Technology in KIET Journal of Computing and Information Sciences
- Vol. 6 (1) , 1-27
Open AccessTemporal Distance Issues and their Mitigation Strategies in GSD: An Empirical Study
GSD has several advantages that make it easier for software companies to follow this trend, like skillful workers, lower costs, and an improved marketplace. To achieve the benefits mentioned above, distributed teams must communicate effectively. Communication is a major problem for development, especially in GSD. Communication issues are classified based on temporal, geographical, and sociocultural distances. The temporal distance problem affects more GSD operations than the other hazards combined. Organizations can mitigate these obstacles and attain the desired quality by employing mitigation strategies while adhering to budget and time constraints. The authors of this study conducted a systematic literature review to identify temporal distance issues and their respective mitigation strategies to develop a conceptual framework. Additionally, a questionnaire interview with closed-ended questions was conducted. Without an empirical study, it is difficult to determine the precise relationship between temporal distance issues and mitigation strategies. The primary objectives of this study are to experimentally evaluate the proposed conceptual framework and the effects of selected risk mitigation measures. This study identified forty-four mitigation techniques and nine temporal distance challenges with a focus on a GSD-based environment, empirical research was conducted using data collected from Pakistani software companies. The findings of this study indicate, with a p-value of 0.01, that various mitigation techniques significantly impact various temporal distance issues. Based on the findings, software organizations can employ appropriate mitigation techniques to address temporal distance issues and reduce the failure rate of software projects.
- Open Access
- 9 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 10, 64234-64256
Identification and Analysis of Factors Influencing Turnover Intention of Pakistan IT Professionals: An Empirical Study
Information Technology is one of the fast-growing industries. However, it is facing a high employee turnover. Lack of management support, low level of compensation plans, poor relationship with supervisors, limited development and training programs, and work stress causes high employee turnover. Therefore, this study is conducted to identify and empirically evaluate the factors of turnover intention in Pakistan’s software and IT industries. For this purpose, a systematic literature review is conducted to identify turnover intention factors from the literature, and a survey is conducted to evaluate the identified factors empirically. In the survey, a total of two hundred and fifty professionals from Pakistan’s IT and software industry participated either through online Google Form or printed survey questionnaires. The collected data were analyzed using regression analysis performed in SPSS and Warp-PLS. According to our findings, IT professionals prefer to work long tenure in the organization if they are provided with better opportunities in terms of compensation plans, career growth, and management support. The findings of this study help the organization’s top management to retain skilled employees.
- 9 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 10, 64234-64256
Open AccessIdentification and Analysis of Factors Influencing Turnover Intention of Pakistan IT Professionals: An Empirical Study
Information Technology is one of the fast-growing industries. However, it is facing a high employee turnover. Lack of management support, low level of compensation plans, poor relationship with supervisors, limited development and training programs, and work stress causes high employee turnover. Therefore, this study is conducted to identify and empirically evaluate the factors of turnover intention in Pakistan’s software and IT industries. For this purpose, a systematic literature review is conducted to identify turnover intention factors from the literature, and a survey is conducted to evaluate the identified factors empirically. In the survey, a total of two hundred and fifty professionals from Pakistan’s IT and software industry participated either through online Google Form or printed survey questionnaires. The collected data were analyzed using regression analysis performed in SPSS and Warp-PLS. According to our findings, IT professionals prefer to work long tenure in the organization if they are provided with better opportunities in terms of compensation plans, career growth, and management support. The findings of this study help the organization’s top management to retain skilled employees.
- 26 January 2022
- Published by Springer Nature in Journal of Digital Imaging
- Vol. 35 (2) , 374-384
Automated 2D Slice-Based Skull Stripping Multi-View Ensemble Model on NFBS and IBSR Datasets
This study proposed and evaluated a two-dimensional (2D) slice-based multi-view U-Net (MVU-Net) architecture for skull stripping. The proposed model fused all three TI-weighted brain magnetic resonance imaging (MRI) views, i.e., axial, coronal, and sagittal. This 2D method performed equally well as a three-dimensional (3D) model of skull stripping. while using fewer computational resources. The predictions of all three views were fused linearly, producing a final brain mask with better accuracy and efficiency. Meanwhile, two publicly available datasets—the Internet Brain Segmentation Repository (IBSR) and Neurofeedback Skull-stripped (NFBS) repository—were trained and tested. The MVU-Net, U-Net, and skip connection U-Net (SCU-Net) architectures were then compared. For the IBSR dataset, compared to U-Net and SC-UNet, the MVU-Net architecture attained better mean dice score coefficient (DSC), sensitivity, and specificity, at 0.9184, 0.9397, and 0.9908, respectively. Similarly, the MVU-Net architecture achieved better mean DSC, sensitivity, and specificity, at 0.9681, 0.9763, and 0.9954, respectively, than the U-Net and SC-UNet for the NFBS dataset.
- 26 January 2022
- Published by Springer Nature in Journal of Digital Imaging
- Vol. 35 (2) , 374-384
Automated 2D Slice-Based Skull Stripping Multi-View Ensemble Model on NFBS and IBSR Datasets
This study proposed and evaluated a two-dimensional (2D) slice-based multi-view U-Net (MVU-Net) architecture for skull stripping. The proposed model fused all three TI-weighted brain magnetic resonance imaging (MRI) views, i.e., axial, coronal, and sagittal. This 2D method performed equally well as a three-dimensional (3D) model of skull stripping. while using fewer computational resources. The predictions of all three views were fused linearly, producing a final brain mask with better accuracy and efficiency. Meanwhile, two publicly available datasets—the Internet Brain Segmentation Repository (IBSR) and Neurofeedback Skull-stripped (NFBS) repository—were trained and tested. The MVU-Net, U-Net, and skip connection U-Net (SCU-Net) architectures were then compared. For the IBSR dataset, compared to U-Net and SC-UNet, the MVU-Net architecture attained better mean dice score coefficient (DSC), sensitivity, and specificity, at 0.9184, 0.9397, and 0.9908, respectively. Similarly, the MVU-Net architecture achieved better mean DSC, sensitivity, and specificity, at 0.9681, 0.9763, and 0.9954, respectively, than the U-Net and SC-UNet for the NFBS dataset.
- 8 February 2021
- Published by SPIE-Intl Soc Optical Eng in Journal of Medical Imaging
- Vol. 8 (01) , 014003-15
Extraction of region of interest from brain MRI by converting images into neutrosophic domain using the modified S-function
Purpose: A brain tumor is deadly as its exact extraction is tricky. However, at times, its removal is the only way to save a patient, leaving very little room for the doctors to make a mistake. Image segmentation algorithms can be used to detect tumor in magnetic resonance imaging (MRI). Irregularity in size, location, and shape of tumor in brain with imbalanced distribution of classes in the dataset make this a challenging task. To deal with these challenges, a region of interest (ROI) is extracted from images by removing redundant information. Approach: We present a process to extract ROIs by converting images into neutrosophic domain. Two modalities FLAIR and T2 were diffused to reduce inhomogeneity in nontumorous regions and then anisotropic diffusion is applied to reduce the noise. The ROIs, which are tumorous regions, were extracted using neutrosophic technique based on the modified S-function. The extracted ROIs were refined by applying the morphological operations in the end. Results: We evaluated our proposed method using three datasets including BraTS 2019 and compared the results with state-of-the-art methods. The parameters sensitivity, false negative rate, and ratio of ROI area to slice area were calculated to evaluate the proposed method. These parameters indicate that the proposed method achieved more than 98% sensitivity, 1.5% false negative rate, and removed more than 80% redundancy. Conclusions: Evaluating parameters indicate that the method proposed has removed most of the redundant data from MRI images and extracted ROIs are composed of tumorous region.
- 8 February 2021
- Published by SPIE-Intl Soc Optical Eng in Journal of Medical Imaging
- Vol. 8 (01) , 014003-15
Extraction of region of interest from brain MRI by converting images into neutrosophic domain using the modified S-function
Purpose: A brain tumor is deadly as its exact extraction is tricky. However, at times, its removal is the only way to save a patient, leaving very little room for the doctors to make a mistake. Image segmentation algorithms can be used to detect tumor in magnetic resonance imaging (MRI). Irregularity in size, location, and shape of tumor in brain with imbalanced distribution of classes in the dataset make this a challenging task. To deal with these challenges, a region of interest (ROI) is extracted from images by removing redundant information. Approach: We present a process to extract ROIs by converting images into neutrosophic domain. Two modalities FLAIR and T2 were diffused to reduce inhomogeneity in nontumorous regions and then anisotropic diffusion is applied to reduce the noise. The ROIs, which are tumorous regions, were extracted using neutrosophic technique based on the modified S-function. The extracted ROIs were refined by applying the morphological operations in the end. Results: We evaluated our proposed method using three datasets including BraTS 2019 and compared the results with state-of-the-art methods. The parameters sensitivity, false negative rate, and ratio of ROI area to slice area were calculated to evaluate the proposed method. These parameters indicate that the proposed method achieved more than 98% sensitivity, 1.5% false negative rate, and removed more than 80% redundancy. Conclusions: Evaluating parameters indicate that the method proposed has removed most of the redundant data from MRI images and extracted ROIs are composed of tumorous region.
- 28 January 2021
- Published by Wiley in International Journal of Imaging Systems and Technology
- Vol. 31 (3) , 1519-1535
Multi‐level dilated convolutional neural network for brain tumour segmentation and multi‐view‐based radiomics for overall survival prediction
Glioblastoma (GBM) is the most high‐risk and grievous tumour in the brain that causes the death of more than 50% of the patients within one to 2 years after diagnosis. Accurate detection and prognosis of this disease are critical to provide essential guidelines for treatment planning. This study proposed using a deep learning‐based network for the GBM segmentation and radiomic features for the patient's overall survival (OS) time prediction. The segmentation model used in this study was a modified U‐Net‐based deep 3D multi‐level dilated convolutional neural network. It uses multiple kernels of altered sizes to capture contextual information at different levels. The proposed scheme for OS time prediction overcomes the problem of information loss caused by the derivation of features in a single view due to the variation in the neighbouring pixels of the tumorous region. The selected features were based on texture, shape, and volume. These features were computed from the segmented components of tumour in axial, coronal, and sagittal views of magnetic resonance imaging slices. The proposed models were trained and evaluated on the BraTS 2019 dataset. Experimental results of OS time prediction on the validation data have showed an accuracy of 48.3%, with the mean squared error of 92 599.598. On the validation data, the segmentation model achieved a mean dice similarity coefficient of 0.75, 0.89, and 0.80 for enhancing tumour, whole tumour, and tumour core, respectively. Future work is warranted to improve the overall performance of OS time prediction based on the findings in this study.
- 28 January 2021
- Published by Wiley in International Journal of Imaging Systems and Technology
- Vol. 31 (3) , 1519-1535
Multi‐level dilated convolutional neural network for brain tumour segmentation and multi‐view‐based radiomics for overall survival prediction
Glioblastoma (GBM) is the most high‐risk and grievous tumour in the brain that causes the death of more than 50% of the patients within one to 2 years after diagnosis. Accurate detection and prognosis of this disease are critical to provide essential guidelines for treatment planning. This study proposed using a deep learning‐based network for the GBM segmentation and radiomic features for the patient's overall survival (OS) time prediction. The segmentation model used in this study was a modified U‐Net‐based deep 3D multi‐level dilated convolutional neural network. It uses multiple kernels of altered sizes to capture contextual information at different levels. The proposed scheme for OS time prediction overcomes the problem of information loss caused by the derivation of features in a single view due to the variation in the neighbouring pixels of the tumorous region. The selected features were based on texture, shape, and volume. These features were computed from the segmented components of tumour in axial, coronal, and sagittal views of magnetic resonance imaging slices. The proposed models were trained and evaluated on the BraTS 2019 dataset. Experimental results of OS time prediction on the validation data have showed an accuracy of 48.3%, with the mean squared error of 92 599.598. On the validation data, the segmentation model achieved a mean dice similarity coefficient of 0.75, 0.89, and 0.80 for enhancing tumour, whole tumour, and tumour core, respectively. Future work is warranted to improve the overall performance of OS time prediction based on the findings in this study.
- Open Access
- 3 July 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 8, 125755-125775
The Impact of Scope Creep on Project Success: An Empirical Investigation
Advocates of software engineering and software project management stated in the literature that creeping of scope is one of the most common causes for the failure of software projects. Also, advocates believed that it could occur in almost every software project, which leads to compromise in quality, delayed schedules, increase cost and decreased customer satisfaction. However, the lack of empirical evidence demands a comprehensive investigation to identify the factors of scope creep and to propose a conceptual framework to empirically evaluate the impact of scope creep on software project success. To determine the scope creep factors in this study, two exploratory methods, i.e. a Systematic Literature Review (SLR) and interview from experts are performed. Following the analysis of these methods, a conceptual framework is proposed. To empirically evaluate the proposed conceptual framework, data is collected through a survey method. Next, the collected data is analyzed through Partial Least Squares’ Structural Equation Modelling (PLS-SEM). From the results, it is evident that the identified factors of scope creep are negatively associated with software project success. The results of empirical evaluation also second the findings of SLR. The outcome of the study may help the practitioners to understand the dynamics of factors, which undermine scope creep in software SMEs and to assist them in the development of effective control and mitigation strategies, therefore, to increase the project success rate.
- 3 July 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 8, 125755-125775
Open AccessThe Impact of Scope Creep on Project Success: An Empirical Investigation
Advocates of software engineering and software project management stated in the literature that creeping of scope is one of the most common causes for the failure of software projects. Also, advocates believed that it could occur in almost every software project, which leads to compromise in quality, delayed schedules, increase cost and decreased customer satisfaction. However, the lack of empirical evidence demands a comprehensive investigation to identify the factors of scope creep and to propose a conceptual framework to empirically evaluate the impact of scope creep on software project success. To determine the scope creep factors in this study, two exploratory methods, i.e. a Systematic Literature Review (SLR) and interview from experts are performed. Following the analysis of these methods, a conceptual framework is proposed. To empirically evaluate the proposed conceptual framework, data is collected through a survey method. Next, the collected data is analyzed through Partial Least Squares’ Structural Equation Modelling (PLS-SEM). From the results, it is evident that the identified factors of scope creep are negatively associated with software project success. The results of empirical evaluation also second the findings of SLR. The outcome of the study may help the practitioners to understand the dynamics of factors, which undermine scope creep in software SMEs and to assist them in the development of effective control and mitigation strategies, therefore, to increase the project success rate.