Busca
1 - 7 de 7
Número de resultados para mostrar por página
Resultados da Busca
-
- Descrição:
- Internet of Things (IoT), a strong integration of radio frequency identifier (RFID), wireless devices, and sensors, has provided a difficult yet strong chance to shape existing systems into intelligent ones. Many new applications have been created in the last few years. As many as a million objects are anticipated to be linked together to form a network that can infer meaningful conclusions based on raw data. This means any IoT system is heterogeneous when it comes to the types of devices that are used in the system and how they communicate with each other. In most cases, an IoT network can be described as a layered network, with multiple tiers stacked on top of each other. IoT network performance improvement typically focuses on a single layer. As a result, effectiveness in one layer may rise while that of another may fall. Ultimately, the achievement issue must be addressed by considering improvements in all layers of an IoT network, or at the very least, by considering contiguous hierarchical levels. Using a parallel and clustered architecture in the device layer, this paper examines how to improve the performance of an IoT network’s controller layer. A particular clustered architecture at the device level has been shown to increase the performance of an IoT network by 16% percent. Using a clustered architecture at the device layer in conjunction with a parallel architecture at the controller layer boosts performance by 24% overall.
- Palavra-chave:
- Layered Networking, Parallel Architectures, Device Level Clustering, Internet of Things, Topology Binding, Performance Optimization, and Radio-frequency Identification
- Sujeito:
- Artificial Intelligence and Data Science
- O Criador:
- Singh, Ninni , Roy, Sudipta , Saad, Ismail , Rahebi, Javad , Farzamnia, Ali , Gunjan, Vinit Kumar , Davanam, Ganesh , and Kallam, Suresh
- Owner:
- n.sakthivel@jioinstitute.edu.in
- Editor:
- MDPI
- Localização:
- Switzerland
- Língua:
- English
- Data carregada:
- 11-02-2023
- Data modificada:
- 16-02-2023
- Data Criada:
- 01-11-2022
- Rights Statement Tesim:
- In Copyright
- License Tesim:
- All rights reserved
- Resource Type:
- Article
- Identificador:
- 10.3390/en15228738
-
- Descrição:
- Fetal brain segmentation and gestational age prediction have been under active research in the field of medical image processing for a long time. However, both these tasks are challenging due to factors like difficulty in acquiring a proper fetal brain image owing to the fetal movement during the scan. With the recent advancements in deep learning, many models have been proposed for performing both the tasks, individually, with good accuracy. In this paper, we present Multi-Tasking Single Encoder U-Net, MTSE U-Net, a deep learning architecture for performing three tasks on fetal brain images. The first task is the segmentation of the fetal brain into its seven components: intracranial space and extra-axial cerebrospinal fluid spaces, gray matter, white matter, ventricles, cerebellum, deep gray matter, and brainstem, and spinal cord. The second task is the prediction of the type of the fetal brain (pathological or neurotypical). The third task is the prediction of the gestational age of the fetus from its brain. All of this will be performed by a single model. The fetal brain images can be obtained by segmenting it from the fetal magnetic resonance images using any of the previous works on fetal brain segmentation, thus showing our work as an extension of the already existing segmentation works. The Jaccard similarity and Dice score for the segmentation task by this model are 77 and 82%, respectively, accuracy for the type of prediction task is 89% and the mean absolute error for the gestational age task is 0.83 weeks. The salient region identification by the model is also tested and these results show that a single model can perform multiple, but related, tasks simultaneously with good accuracy, thus eliminating the need to use separate models for each task.
- Palavra-chave:
- Medical image processing, Fetal brain segmentation, Deep learning, Fetal gestational age prediction, and Convolutional neural networks
- Sujeito:
- Data Science and Artificial Intelligence
- O Criador:
- Ganguly, Debayan , Chatterjee, Kingshuk , Gangopadhyay, Tuhinangshu , Sarkar, Surjadeep , Halder, Shinjini , Dasgupta, Paramik , and Roy, Sudipta
- Owner:
- n.sakthivel@jioinstitute.edu.in
- Editor:
- Springer Nature
- Localização:
- Switzerland
- Língua:
- English
- Data carregada:
- 11-02-2023
- Data modificada:
- 16-02-2023
- Data Criada:
- 01-11-2022
- Rights Statement Tesim:
- In Copyright
- License Tesim:
- All rights reserved
- Resource Type:
- Article
- Identificador:
- 10.1007/s13721-022-00394-y
-
- Descrição:
- Purpose For radiologists, identifying and assessing thelung nodules of cancerous form from CT scans is a difficult and laborious task. As a result, early lung growing prediction is required for the investigation technique, and hence it increases the chances of a successful treatment. To ease this problem, computer-aided diagnostic (CAD) solutions have been deployed. The main purpose of the work is to detect the nodules are malignant or not and to provide the results with better accuracy. Methods A neural network model that incorporates a feedback loop is the recurrent neural network. Evolutionary algorithms such as the Grey Wolf Optimization Algorithm and Recurrent Neural Network (RNN) Techniques are investigated utilising the Matlab Tool in this work. Statistical attributes are also produced and compared with other RNN with Genetic Algorithm (GA) and Particle Swarm Optimization (PSO)combinations for study. Results The proposed method produced very high accuracy, sensitivity, specificity, and precision and compared with other state of art methods. Because of its simplicity and possible global search capabilities, evolutionary algorithms have shown tremendous promise in the area of feature selection in the latest years. Conclusion The proposed techniques have demonstrated outstanding outcomes in various disciplines, outperforming classical methods. Early detection of lung nodules will aid in determining whether the nodules will become malignant or not.
- Palavra-chave:
- Optimization, Lung Cancer, Recurrent Neural Network, and CT images
- Sujeito:
- Data Science and Artificial Intelligence
- O Criador:
- Roy, Sudipta, Gunjan, Vinit Kumar , Shaik, Fahimudin, and Singh, Ninni
- Owner:
- n.sakthivel@jioinstitute.edu.in
- Editor:
- Springer Nature
- Localização:
- Cham and Switzerland
- Língua:
- English
- Data carregada:
- 10-02-2023
- Data modificada:
- 16-02-2023
- Data Criada:
- 01-11-2022
- Rights Statement Tesim:
- In Copyright
- License Tesim:
- All rights reserved
- Resource Type:
- Article
- Identificador:
- 10.1007/s12553-022-00700-8
-
- Palavra-chave:
- Artificial intelligence, Explainable AI, Deep learning, and Medical image
- O Criador:
- Tanushree Meena, Debojyoti Pal , and Sudipta Roy
- Owner:
- nancy1.singh@jioinstitute.edu.in
- Data carregada:
- 16-04-2024
- Data modificada:
- 16-04-2024
- Rights Statement Tesim:
- In Copyright
- License Tesim:
- All rights reserved
-
- Descrição:
- Architected cellular materials, such as lattice structures, offer potential for tunable mechanical properties for dynamic applications of energy absorption and impact mitigation. In this work, the static and dynamic behavior of polymeric lattice structures was investigated through experiments on octet-truss, Kelvin, and cubic topologies with relative densities around 8%. Dynamic testing was conducted via direct impact experiments (25–70 m/s) with high-speed imaging coupled with digital image correlation and a polycarbonate Hopkinson pressure bar. Mechanical properties such as elastic wave speed, deformation modes, failure properties, particle velocities, and stress histories were extracted from experimental results. At low impact velocities, a transient dynamic response was observed which was composed of a compaction front initiating at the impact surface and additional deformation bands whose characteristics matched low strain-rate behavior. For higher impact velocities, shock analysis was carried out using compaction wave velocity and Eulerian Rankine–Hugoniot jump conditions with parameters determined from full-field measurements.
- Palavra-chave:
- Failure, Digital image correlation, Transient dynamic, Shock, Lattice structure, and Compaction
- Sujeito:
- Applied Science and Engineering
- O Criador:
- Weeks, J. S. and Ravichandran, Guruswami
- Owner:
- n.sakthivel@jioinstitute.edu.in
- Editor:
- Springer Nature
- Localização:
- Switzerland
- Língua:
- English
- Data carregada:
- 21-03-2023
- Data modificada:
- 21-03-2023
- Data Criada:
- 01-12-2022
- Rights Statement Tesim:
- In Copyright
- License Tesim:
- All rights reserved
- Resource Type:
- Article
- Identificador:
- 10.1007/s40870-022-00359-2
-
- Descrição:
- Architected cellular materials, such as lattice structures, offer potential for tunable mechanical properties for dynamic applications of energy absorption and impact mitigation. In this work, the static and dynamic behavior of polymeric lattice structures was investigated through experiments on octet-truss, Kelvin, and cubic topologies with relative densities around 8%. Dynamic testing was conducted via direct impact experiments (25–70 m/s) with high-speed imaging coupled with digital image correlation and a polycarbonate Hopkinson pressure bar. Mechanical properties such as elastic wave speed, deformation modes, failure properties, particle velocities, and stress histories were extracted from experimental results. At low impact velocities, a transient dynamic response was observed which was composed of a compaction front initiating at the impact surface and additional deformation bands whose characteristics matched low strain-rate behavior. For higher impact velocities, shock analysis was carried out using compaction wave velocity and Eulerian Rankine–Hugoniot jump conditions with parameters determined from full-field measurements.
- Palavra-chave:
- Transient dynamic, Failure, Shock, Compaction, Lattice structure, and Digital image correlation
- Sujeito:
- Material Science, Chemistry, and Chemical Engineering
- O Criador:
- Weeks, J. S. and Ravichandran, G.
- Owner:
- n.sakthivel@jioinstitute.edu.in
- Editor:
- Springer Nature
- Localização:
- Switzerland
- Língua:
- English
- Data carregada:
- 11-02-2023
- Data modificada:
- 16-02-2023
- Data Criada:
- 01-12-2022
- Rights Statement Tesim:
- In Copyright
- License Tesim:
- All rights reserved
- Resource Type:
- Article
- Identificador:
- 10.1007/s40870-022-00359-2
-
7. Bone Fracture Detection Using Deep Supervised Learning from Radiological Images: A Paradigm Shift
- Descrição:
- Bone diseases are common and can result in various musculoskeletal conditions (MC). An estimated 1.71 billion patients suffer from musculoskeletal problems worldwide. Apart from musculoskeletal fractures, femoral neck injuries, knee osteoarthritis, and fractures are very common bone diseases, and the rate is expected to double in the next 30 years. Therefore, proper and timely diagnosis and treatment of a fractured patient are crucial. Contrastingly, missed fractures are a common prognosis failure in accidents and emergencies. This causes complications and delays in patients’ treatment and care. These days, artificial intelligence (AI) and, more specifically, deep learning (DL) are receiving significant attention to assist radiologists in bone fracture detection. DL can be widely used in medical image analysis. Some studies in traumatology and orthopaedics have shown the use and potential of DL in diagnosing fractures and diseases from radiographs. In this systematic review, we provide an overview of the use of DL in bone imaging to help radiologists to detect various abnormalities, particularly fractures. We have also discussed the challenges and problems faced in the DL-based method, and the future of DL in bone imaging.
- Palavra-chave:
- Deep Learning, Fractures, Bone Imaging, Artificial Intelligence, Radiology, and Computer Vision
- Sujeito:
- Data Science and Artificial Intelligence
- O Criador:
- Roy, Sudipta and Meena, Tanushree
- Contribuinte:
- Jio Institute CVMI-Computer Vision in Medical Imaging Project
- Owner:
- n.sakthivel@jioinstitute.edu.in
- Editor:
- MDPI
- Localização:
- Switzerland
- Língua:
- English
- Data carregada:
- 10-02-2023
- Data modificada:
- 16-02-2023
- Data Criada:
- 01-10-2022
- Rights Statement Tesim:
- In Copyright
- License Tesim:
- All rights reserved
- Resource Type:
- Article
- Identificador:
- 10.3390/diagnostics12102420