JMIR Publications

ABSTRACT

Background:

A significant contribution of ICT to healthcare is constituted by systems automating and enhancing the management of research and clinical data. A recent advancement is the development of imaging biobanks which are now enabling collection and storing of diagnostic images, clinical reports, and demographic data to allow researchers to identify associations between lifestyle and genetic factors, and imaging-derived phenotypes.

Objective:

In this work we describe and evaluate the network and system performance of an operating biobank for diagnostic images, based on the XNAT [1] open-source platform.

Methods:

Three usage cases are evaluated focusing on the point of view of the user (researcher or radiologist) which interacts with the biobank and inspects chest CT exams of COVID19 cohort. The experiments consider four network setups: (i) Local (virtual machines communicating in a single host), (ii) LAN (organization-local access), (iii) VPN (remote secure access through the Internet), (iv) WAN (as in VPN, but without the IP-level tunnelling). The experimental setup records the activity of a human user interacting with the biobank system and replays it consistently multiple times. Several metrics are extracted from network traffic traces and server logs that are captured during the activity replay.

Results:

There are two principal outputs: one related to the diagnostic data transfer and one related to the two different users’ access, Research and Radiologist. Regarding the diagnostic data transfer, two kinds of containers are considered: the Web and the Database. The results show that the first one seems to be the most memory-hungry and to have the higher computational load with respect the second one. Regarding the users’ case access: the results show that the two users have the same network performances but the higher resources consumption is registered for two different actions respectively.

Conclusions:

The analysis shows that the current setup is well provisioned for satisfying the planned number of concurrent users. More importantly, it also highlights and quantifies the resource demands of specific user actions and thus provides a guideline for planning, setting up and using an images biobanking system.