Title: Similarity of chemotherapy histories based on imputed values

Authors: Mourad Khayati; Jay Martin Anderson; Michael H. Böhlen; Johann Gamper; Manfred Mitterer

Addresses: Department of Informatics, University of Zürich, Binzmüehlestrasse, 14 CH-8050, Zurich, Switzerland. ' Franklin and Marshall College, Lancaster, PA 17604-3003, USA. ' Department of Informatics, University of Zürich, Binzmüehlestrasse, 14 CH-8050, Zurich, Switzerland. ' Faculty of Computer Science, Free University of Bozen/Bolzano, Dominikanerplatz 3/piazza, Domenicani, 3, I-39100 Bozen/Bolzano, Italy. ' Department fr Labormedizin, Krankenhaus Meran/Ospedale Merano, Rossini-Straße 5, I-39012, Meran/Merano, Italy

Abstract: The comparison of time series of multivariate data is a long-standing problem in many applications in the clinical domain. We propose two approaches to retrieve from a hospital data warehouse the k patients P1, ..., Pn with a chemotherapy history that is most similar to patient Q: the first is based on warping distance, together with an initial alignment using imputed values. The second is based on the volume of the Kiviat tube. In implementing the Euclidean distance, we investigate the addition of null events to achieve similar cardinality, and dynamic time warping, a widely-used technique in the comparison of time series data. The investigations are based on a real world clinical database.

Keywords: treatment courses; treatment similarity; chemotherapy histories; Euclidean distance; dynamic time warping; DTW; alignment with imputed values; buckets of events; multivariate time series; hospital data warehousing; clinical databases; cancer treatment.

DOI: 10.1504/IJMEI.2012.048389

International Journal of Medical Engineering and Informatics, 2012 Vol.4 No.3, pp.282 - 298

Received: 08 May 2021
Accepted: 12 May 2021

Published online: 02 Aug 2012 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article