A study on machine learning methods for accuracy improvement of in-body device localization

Dmitrieva D, Anzai D, Kirchner J, Fischer G, Wang J (2022)

Publication Language: English

Publication Type: Conference contribution

Publication year: 2022

Event location: Nagoya JP

URI: https://www.bioem.org/

Abstract

Wireless devices for medical purposes play a promising role in implant body area networks. This study presented an artificial intelligence-based method to improve the localization of the wireless capsule endoscope during the liquid phantom experiment. The channel characteristics of signal at 400 MHz medical implant communication service (MICS) band were calculated based on the experimental data and used for conventional maximum likelihood estimation as well as to simulate the synthetic training data to introduce a possible solution for training data harvesting. The proposed method showed the root mean square localization error of 1.74 cm compared with 6.66 cm of the conventional method

Authors with CRIS profile

Daria Dmitrieva Professur für Technische Elektronik Jens Kirchner Lehrstuhl für Technische Elektronik Georg Fischer Professur für Technische Elektronik

Involved external institutions

Nagoya Institute of Technology / 名古屋工業大学 Nagoya Kōgyō Daigaku JP Japan (JP)

How to cite

APA:

Dmitrieva, D., Anzai, D., Kirchner, J., Fischer, G., & Wang, J. (2022). A study on machine learning methods for accuracy improvement of in-body device localization. In Proceedings of the BIOEM The Joint Annual Meeting of the Bioelectromagnetics Society and the European Bioelectromagnetics Association. Nagoya, JP.

MLA:

Dmitrieva, Daria, et al. "A study on machine learning methods for accuracy improvement of in-body device localization." Proceedings of the BIOEM The Joint Annual Meeting of the Bioelectromagnetics Society and the European Bioelectromagnetics Association, Nagoya 2022.

BibTeX: Download