WebRTC based invariant scattering convolution network for automated validation of ultrasonic videos for IoT enabled tele-sonography

Bharath, R and P, Rajalakshmi (2018) WebRTC based invariant scattering convolution network for automated validation of ultrasonic videos for IoT enabled tele-sonography. In: 4th IEEE World Forum on Internet of Things, WF-IoT, 5-8 February 2018, Singapore.

Full text not available from this repository. (Request a copy)


Tele-sonography works on inherent assumption that the transmitted medical ultrasound videos scanned from remote patients contain the representative data for doing the diagnosis. Due to the high subjectivity involved in scanning and semi-skilled nature of the operating person, this assumption may not always be valid. The remotely scanned ultrasound video contains a lot of redundant information, which is not useful for diagnosis. Transmitting redundant and large volumes of medical data to the expert end for analysis may lead to faulty diagnosis, associated with high transmission cost, and also poses serious challenges on data storage, processing, infrastructure, etc. Addressing these issues, we propose a novel WebRTC based framework to detect representative frames in the ultrasound video and transmit only those frames to the remote sonographer for getting a diagnosis. Detection of representative frames in ultrasound video is done with invariant scattering convolution network. The entire framework is developed using WebRTC, which enables the browser to browser communication thus reducing the computation on end ultrasound scanner and ensures ubiquitous and secured connectivity between technician and the sonographer. The proposed video validation algorithm achieved an accuracy of 96.5% in classifying the representative frames and nonrepresentative frames in the ultrasound video.

[error in script]
IITH Creators:
IITH CreatorsORCiD
Item Type: Conference or Workshop Item (Paper)
Subjects: Electrical Engineering
Divisions: Department of Electrical Engineering
Depositing User: Team Library
Date Deposited: 08 Aug 2018 03:50
Last Modified: 08 Aug 2018 03:51
URI: http://raiith.iith.ac.in/id/eprint/4368
Publisher URL: http://doi.org/10.1109/WF-IoT.2018.8355197
Related URLs:

Actions (login required)

View Item View Item
Statistics for RAIITH ePrint 4368 Statistics for this ePrint Item