Authors:	Mähringer-Kunz, Aline Wagner, Franziska Hahn, Felix Weinmann, Arndt Brodehl, Sebastian Schotten, Sebastian Hinrichs, Jan B. Düber, Christoph Galle, Peter R. Pinto dos Santos, Daniel Kloeckner, Roman
Title:	Predicting survival after transarterial chemoembolization for hepatocellular carcinoma using a neural network : a pilot study
Online publication date:	26-Aug-2021
Year of first publication:	2020
Language:	english
Abstract:	BACKGROUND AND AIMS Deciding when to repeat and when to stop transarterial chemoembolization (TACE) in patients with hepatocellular carcinoma (HCC) can be difficult even for experienced investigators. Our aim was to develop a survival prediction model for such patients undergoing TACE using novel machine learning algorithms and to compare it to conventional prediction scores, ART, ABCR and SNACOR. METHODS For this retrospective analysis, 282 patients who underwent TACE for HCC at our tertiary referral centre between January 2005 and December 2017 were included in the final analysis. We built an artificial neural network (ANN) including all parameters used by the aforementioned risk scores and other clinically meaningful parameters. Following an 80:20 split, the first 225 patients were used for training; the more recently treated 20% were used for validation. RESULTS The ANN had a promising performance at predicting 1-year survival, with an area under the ROC curve (AUC) of 0.77 ± 0.13. Internal validation yielded an AUC of 0.83 ± 0.06, a positive predictive value of 87.5% and a negative predictive value of 68.0%. The sensitivity was 77.8% and specificity 81.0%. In a head-to-head comparison, the ANN outperformed the aforementioned scoring systems, which yielded lower AUCs (SNACOR 0.73 ± 0.07, ABCR 0.70 ± 0.07 and ART 0.54 ± 0.08). This difference reached significance for ART (P < .001); for ABCR and SNACOR significance was not reached (P = .143 and P = .201). CONCLUSIONS Artificial neural networks could be better at predicting patient survival after TACE for HCC than traditional scoring systems. Once established, such prediction models could easily be deployed in clinical routine and help determine optimal patient care.
DDC:	610 Medizin 610 Medical sciences
Institution:	Johannes Gutenberg-Universität Mainz
Department:	FB 04 Medizin
Place:	Mainz
ROR:	https://ror.org/023b0x485
DOI:	http://doi.org/10.25358/openscience-6311
Version:	Published version
Publication type:	Zeitschriftenaufsatz
License:	CC BY
Information on rights of use:	https://creativecommons.org/licenses/by/4.0/
Journal:	Liver international 40 3
Pages or article number:	694 703
Publisher:	Wiley-Blackwell
Publisher place:	Oxford
Issue date:	2020
ISSN:	1478-3231
Publisher URL:	https://doi.org/10.1111/liv.14380
Publisher DOI:	10.1111/liv.14380
Appears in collections:	JGU-Publikationen