ORCID Profile
0000-0002-8302-4571
Current Organisations
AIT Austrian Institute of Technology GmbH
,
Vishwakarma University
Does something not look right? The information on this page has been harvested from data sources that may not be up to date. We continue to work with information providers to improve coverage and quality. To report an issue, use the Feedback Form.
Publisher: MDPI AG
Date: 17-08-2022
DOI: 10.3390/EN15165956
Abstract: The architectural and construction professions are in a state of major transition [...]
Publisher: MDPI AG
Date: 24-04-2022
DOI: 10.3390/ELECTRONICS11091353
Abstract: In the medical field, due to their economic and clinical benefits, there is a growing interest in minimally invasive surgeries and microscopic surgeries. These types of surgeries are often recorded during operations, and these recordings have become a key resource for education, patient disease analysis, surgical error analysis, and surgical skill assessment. However, manual searching in this collection of long-term surgical videos is an extremely labor-intensive and long-term task, requiring an effective content-based video analysis system. In this regard, previous methods for surgical video retrieval are based on handcrafted features which do not represent the video effectively. On the other hand, deep learning-based solutions were found to be effective in both surgical image and video analysis, where CNN-, LSTM- and CNN-LSTM-based methods were proposed in most surgical video analysis tasks. In this paper, we propose a hybrid spatiotemporal embedding method to enhance spatiotemporal representations using an adaptive fusion layer on top of the LSTM and temporal causal convolutional modules. To learn surgical video representations, we propose exploring the supervised contrastive learning approach to leverage label information in addition to augmented versions. By validating our approach to a video retrieval task on two datasets, Surgical Actions 160 and Cataract-101, we significantly improve on previous results in terms of mean average precision, 30.012 ± 1.778 vs. 22.54 ± 1.557 for Surgical Actions 160 and 81.134 ± 1.28 vs. 33.18 ± 1.311 for Cataract-101. We also validate the proposed method’s suitability for surgical phase recognition task using the benchmark Cholec80 surgical dataset, where our approach outperforms (with 90.2% accuracy) the state of the art.
Publisher: MDPI AG
Date: 21-06-2022
DOI: 10.3390/ELECTRONICS11131942
Abstract: Authentication, authorization, and data access control are playing major roles in data security and privacy. The proposed model integrated the multi-factor authentication–authorization process with dependable and non-dependable factors and parameters based on providing security for tenants through a hybrid approach of fully homomorphic encryption methodology: the enhanced homomorphic cryptosystem (EHC) and the Brakersky–Gentry–Vaikuntanathan (BGV) scheme. This research was composed of four major elements: the fully homomorphic encryption blended schemes, EHC and BGV secure token and key implications based on dependable and don-dependable factors an algorithm for generating the tokens and the suitable keys, depending on the user’s role and the execution of experimental test cases by using the EHC algorithm for key and token generation, based on dependable and non-dependable parameters and time periods. The proposed approach was tested with 152 end-users by integrating six multi-tenants, five head tenants, and two enterprise levels and achieved a 92 percent success rate. The research integrated 32-bit plain text in the proposed hybrid approach by taking into consideration the encryption time, decryption time, and key generation time of data transmission via cloud servers. The proposed blended model was efficient in preventing data from ciphertext attacks and achieved a high success rate for transmitting data between the multi-tenants, based on the user-role-user type of enterprise cloud servers.
Publisher: Springer Science and Business Media LLC
Date: 03-2022
Publisher: Wiley
Date: 29-11-2021
DOI: 10.1111/EXSY.12894
Abstract: The conventional technique of leukocyte cell classification involves segmenting the required portion of cells from input image, extracting features of the segmented nuclei, reducing and optimizing these features and then implements the classifier. Thus, designing a good classifier by using such techniques increases the time complexity of the system. In order to resolve such issues, the proposed work implements the deep convolutional neural network (DCNN)‐based models for classifying malignant versus normal WBCs. The proposed system is validated on 108 images of ALL‐IDB 1. Due to limited number of training s les, data augmentation is used to create a similar type of virtual image. In this work, experimentation is carried out for discrimination between normal and infected WBC using DCNN with four different activation functions. By using this method, a set of 6000 s les are generated and used for proper training of the DL model for all activation functions. The performance of each trained model is evaluated in terms of accuracy, recall, precision and F‐measure with the maximum values of 98.1%, 98.3%, 98.3% and 98.3% are achieved, respectively. Finally, it has been concluded that the defined DCNN model and ReLu activation function yield outstanding performance for lymphoblast characterization using microscopic blood images.
Publisher: MDPI AG
Date: 21-05-2022
DOI: 10.3390/SU14106280
Abstract: The rapid development of technology has empowered us to achieve resilient infrastructure to establish a sustainable ecosystem. The construction site is one of the highest risk jobs for accident-related fatalities and injuries globally. From the previous studies, it is concluded that untrained or inexperienced workers were responsible for 40% of work-related accidents and the Health and Safety Executive (HSE) report concludes that inadequate working experience, knowledge, and safety awareness were the key causes of fatal accidents in the construction industry. Moreover, it is identified from previous studies that digital technology such as IoT with the assistance of wireless sensors can enhance the safety of construction sites. Based on this advantage, this study has implemented the hybrid architecture with the integration of the 2.4 GHz Zigbee, 433 MHz long-range (LoRa), and Wi-Fi communication protocol to monitor the health status of workers and construction sites and also to identify workers’ equipment wearing status in real-time scenarios. The proposed architecture is realized by implementing customized hardware, based on 2.4 GHz Zigbee, 433 MHz long-range (LoRa), and Wi-Fi. Furthermore, in the analysis of the evaluation metrics of LoRa, it is concluded that the lowest sensitivity is observed for SF 12 at BW 41.7 kHz and the highest is observed for SF 7 at BW 500 kHz the maximum value data rate is observed at BW 500 kHz at CR 1 for SF 7, and the minimum data rate is observed at BW 41.7 at CR 4 for SF 12. In the future, the customized hardware will be implemented in different construction environments resolving possible challenges that empower to implementation of the proposed architecture in wide extensions.
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 05-2022
No related grants have been discovered for Dr. Mamoon Rashid.