DOI: 10.4018/IJEHMC.2021010102 International Journal of E-Health and Medical Communications Volume 12 • Issue 1 • January-March 2021 Copyright © 2021, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. 16 Data Mining of MR Technical Parameters: A Case Study for SAR in a Large- Scale MR Repository Adriana Murraças, University of Aveiro, Portugal Paula Maria Vaz Martins, School of Health Sciences, University of Aveiro, Portugal Carlos Daniel Cipriani Ferreira, Perspectum Diagnostics, Ltd, Oxford, UK & Institute of Nuclear Sciences Applied to Health (ICNAS), University of Coimbra, Portugal Tiago Marques Godinho, University of Aveiro, Portugal Augusto Marques Ferreira da Silva, University of Aveiro, Portugal ABSTRACT Exposure to radiofrequency (RF) energy during a magnetic resonance imaging exam is a safety concern related to biological thermal effects. Estimation of the specific absorption rate (SAR) is done by manufacturer scanner integrated tools to monitor RF energy. This work presents an exploratory approach of DICOM metadata focused in whole-body SAR values, patient dependent parameters, and pulse sequences. Previously acquired abdominopelvic and head studies were retrieved from a 3 Tesla scanner. Dicoogle tool was used for metadata indexing, mining, and extraction. Specifically weighted pulse sequences were related with weight, BMI, and gender through boxplot diagrams and effect size analysis. A decrease of SAR values with increasing body weight and BMI categories is observable for abdominopelvic studies. Head studies showed different trends regarding distinct pulse sequences; in addition, underage patients register higher SAR values compared to adults. Male individuals register marginally higher SAR values. Metadata recording practices and standardization need to be improved. KEywoRDS Cliff’s Delta, Data Mining, Dicoogle, Effect Size, Large Repository, Magnetic Resonance, Pulse Sequences, Specific Absorption Rate INTRoDUCTIoN The imaging modality of magnetic resonance (MR) involves the absorption of radiofrequency (RF) energy by the human body, comprising one of the main patient safety concerns during the exam due to heating risk of tissues. Monitoring RF absorption is achieved by the estimation of specific absorption rate (SAR) expressed in watts per kilogram (W/kg) (Hartwig, 2015). Several studies deal with SAR values over computational simulations, but the usage of the ones obtained in real patient in everyday MR practice are hardly ever analyzed. Guidelines and reviews address SAR as dependent of the patient weight, introduced by the MR technologist, but are not specific in what way weight influences SAR. The only way to assess SAR during the MR exam is to trust in the methods applied by the manufacturers, that return an output of an estimated SAR that can limit acquisition conditions according to guideline values. This article, published as an Open Access article on January 7, 2021 in the gold Open Access journal, International Journal of E-Health and Medical Communications (converted to gold Open Access January 1, 2021), is distributed under the terms of the Creative Commons At- tribution License (http://creativecommons.org/licenses/by/4.0/) which permits unrestricted use, distribution, and production in any medium, provided the author of the original work and original publication source are properly credited.