Journal
ADVANCES IN INTELLIGENT NETWORKING AND COLLABORATIVE SYSTEMS (INCOS-2021)
Volume 312, Issue -, Pages 233-242Publisher
SPRINGER INTERNATIONAL PUBLISHING AG
DOI: 10.1007/978-3-030-84910-8_25
Keywords
-
Categories
Funding
- ERDF in project A Research Platform focused on Industry [CZ.02.1.01/0.0/0.0/17 049/0008425]
- Technology Agency of the Czech Republic [TN01000024]
- National Competence Center -Cybernetics and Artificial Intelligence [SP2021/24]
- VSB -Technical University of Ostrava
Ask authors/readers for more resources
Medical imaging and image analysis play an important role in modern diagnosis and treatment, and intelligent image processing and data analysis can enhance the detection, segmentation, and understanding of medical images. However, modern image analysis methods are often computationally complex and slow to be applied. This study analyzes the performance and costs of several neural models for medical image analysis using different NVIDIA Jetson modules, focusing on lung X-ray images related to COVID-19.
Medical imaging and image analysis are important elements of modern diagnostic and treatment methods. Intelligent image processing, pattern recognition, and data analysis can be leveraged to introduce a new level of detection, segmentation, and, in general, understanding to medical image analysis. However, modern image analysis methods such as deep neural networks are often connected with significant computational complexity, slowing their adoption. Recent embedded systems such as the NVIDIA Jetson general-purpose GPUs became a viable platform for efficient execution of some computational models. This work analyzes the performance and time and energy costs of several neural models for medical image analysis on different kinds of NVIDIA Jetson modules. The experiments are performed with the lung X-ray medical images in connection with the COVID-19 disease.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available