Ričards Marcinkevičs

Ričards Marcinkevičs

Finding the question is often more important than finding the answer — John W. Tukey

Selected Publications

You can find the full list of my articles on my Google Scholar profile. Below is a list of selected publications:

Interpretable and intervenable ultrasonography-based machine learning models for pediatric appendicitis

Published in Medical Image Analysis, 2024

This work presents interpretable machine learning models for predicting the diagnosis, management and severity of suspected appendicitis using ultrasound images. Our approach utilizes concept bottleneck models that facilitate interpretation and interaction with high-level concepts understandable to clinicians.

Recommended citation: Marcinkevičs, R., Reis Wolfertstetter, P., Klimiene, U., Chin-Cheong, K., Paschke, A., Zerres, J., … Vogt, J. E. (2024). Interpretable and intervenable ultrasonography-based machine learning models for pediatric appendicitis. Medical Image Analysis, 91, 103042. http://rmarcinkevics.github.io/files/2024-01-01-ml-app-us.pdf

Interpretable and Explainable Machine Learning: A Methods-centric Overview with Concrete Examples

Published in Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 2023

This review is intended for a general machine learning audience interested in exploring the challenges of interpretation and explanation beyond the logistic regression or random forest variable importance. We examine inductive biases behind interpretable and explainable machine learning and illustrate them with concrete examples from the literature.

Recommended citation: Marcinkevičs, R., & Vogt, J. E. (2023). Interpretable and explainable machine learning: A methods‐centric overview with concrete examples. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, e1493. http://rmarcinkevics.github.io/files/2023-02-28-interexplain.pdf

Debiasing Deep Chest X-Ray Classifiers using Intra- and Post-processing Methods

Published in 7th Machine Learning for Healthcare Conference, MLHC 2022, 2022

This work presents two intra-processing techniques based on fine-tuning and pruning an already-trained neural network that can help achieve group fairness of deep medical image classifiers when deploying them in domains with different fairness considerations and constraints.

Recommended citation: Marcinkevičs, R., Ozkan, E., & Vogt, J.E. (2022). Debiasing Deep Chest X-Ray Classifiers using Intra- and Post-processing Methods. Proceedings of the 7th Machine Learning for Healthcare Conference, in Proceedings of Machine Learning Research 182:504-536. http://rmarcinkevics.github.io/files/2022-08-26-prune.pdf

A Deep Variational Approach to Clustering Survival Data

Published in 10th International Conference on Learning Representations, ICLR 2022, 2022

We introduce a novel semi-supervised probabilistic approach to cluster survival data by leveraging recent advances in stochastic gradient variational inference. Our method employs a deep generative model to uncover the underlying distribution of both the explanatory variables and censored survival times.

Recommended citation: Manduchi, L., Marcinkevičs, R., Massi, M. C., Weikert, T. J., Sauter, A., Gotta, V., … Vogt, J. E. (2022). A Deep Variational Approach to Clustering Survival Data. 10th International Conference on Learning Representations, ICLR 2022. http://rmarcinkevics.github.io/files/2022-04-25-vadesc.pdf

Rapid and Reversible Control of Human Metabolism by Individual Sleep States

Published in Cell Reports, 2021

This paper explores the regulation of metabolites detected in human breath by sleep states using neural-network-based inference techniques for Granger causality.

Recommended citation: Nowak, N., Gaisl, T., Miladinovic, D., Marcinkevičs, R., Osswald, M., Bauer, S., ... & Kohler, M. (2021). Rapid and reversible control of human metabolism by individual sleep states. Cell Reports, 37(4), 109903. http://rmarcinkevics.github.io/files/2021-10-26-slimmba.pdf

Interpretable Models for Granger Causality Using Self-explaining Neural Networks

Published in 9th International Conference on Learning Representations, ICLR 2021, 2021

This paper proposes a novel framework for inferring multivariate Granger causality under nonlinear dynamics based on an extension of self-explaining neural networks.

Recommended citation: Marcinkevičs, R., & Vogt, J. E. (2021). Interpretable Models for Granger Causality Using Self-explaining Neural Networks. 9th International Conference on Learning Representations, ICLR 2021. http://rmarcinkevics.github.io/files/2021-10-14-gvar.pdf

Exploring Relationships between Cerebral and Peripheral Biosignals with Neural Networks

Published in 2021 IEEE International Conference on Digital Health, ICDH, 2021

Neural networks are applied to explore high-dimensional nonlinear correlations between the cerebral brainwaves and variations in heart rate and electrodermal activity.

Recommended citation: Hatteland, A. H., Marcinkevičs, R., Marquis, R., Frick, T., Hubbard, I., Vogt, J. E., ... & Ryvlin, P. (2021). Exploring Relationships between Cerebral and Peripheral Biosignals with Neural Networks. In 2021 IEEE International Conference on Digital Health (ICDH) (pp. 103-113). IEEE. http://rmarcinkevics.github.io/files/2021-09-05-brainperi.pdf

Using Machine Learning to Predict the Diagnosis, Management and Severity of Pediatric Appendicitis

Published in Frontiers in Pediatrics, 2021

This pilot study is the first to simultaneously predict the diagnosis, management, and severity of the disease in children with suspected appendicitis. We present the first ML model for appendicitis that was deployed as an open access easy-to-use online tool.

Recommended citation: Marcinkevičs, R., Reis Wolfertstetter, P., Wellmann, S., Knorr, C., & Vogt, J. E. (2021). Using Machine Learning to Predict the Diagnosis, Management and Severity of Pediatric Appendicitis. Frontiers in Pediatrics, 360. http://rmarcinkevics.github.io/files/2021-04-29-ml-app.pdf