@inproceedings{balaka2026voidlytics,
author = {Balaka, Hanna and Zhang, Ke Er Amy and Czinkoczky, Marta Menta and Vaudel, Marc and Hauser, Helwig and Garrison, Laura Ann},
booktitle = {Eurographics Workshop on Visual Computing for Biology and Medicine},
title = {Voidlytics for Leveraging Informative Missingness in the Study of Hyperemesis Gravidarum},
year = {2026},
editor = {Krueger, Robert and Mörth, Eric and Furmanová, Katarina},
publisher = {The Eurographics Association},
ISSN = {2070-5786},
ISBN = {978-3-03868-324-7},
abstract = {Public health studies commonly use questionnaires to get insights into human health. The Norwegian Mother, Father and Child Cohort Study (MoBa) is a longitudinal study that captures self-reported data on pregnancy symptoms across gestation, in particular also with a potential for an improved understanding of hyperemesis gravidarum (HG), a severe pregnancy complication. Although missing values are prevalent in these data and can hinder the analysis of the collected information, researchers studying MoBa data hypothesize that missingness within observed responses may signal HG. We present a design study exploring the problem space of hypothesis generation regarding relationships between missingness response patterns in the MoBa data and HG. Through this study, we also introduce Voidlytics, a visualization tool that supports hypothesis generation. It facilitates the exploration of missingness response patterns, the explicit formulation of assumptions about how suggestive of HG specific patterns are, as well as testing these assumptions. We provide initial evidence that the tool may support the generation of hypotheses about how indicative of HG missingness response patterns are. Supplementary materials are available at https://osf.io/3ycp5.},
pdf = {pdfs/balaka2026voidlytics.pdf},
images = {images/balaka2026voidlytics.png},
thumbnails = {images/balaka2026voidlytics_thumb.png},
}