1,721,154 research outputs found
Computer aided detection as a decision aid in medical screening
Full text linkContains fulltext :
91305.pdf (Publisher’s version ) (Open Access)Radboud Universiteit Nijmegen, 14 december 2011Promotores : Karssemeijer, N., Lucas, P.J.F.176 p
Temporal probabilistic models for disease management : with applications in COPD care
Full text linkContains fulltext :
115726.pdf (Publisher’s version ) (Open Access)Radboud Universiteit Nijmegen, 04 december 2013Promotor : Lucas, P.J.F. Co-promotor : Schermer, T.R.J.129 p
New network models for the analysis of disease interaction with applications in multimorbidity
Full text linkContains fulltext :
128893.pdf (Publisher’s version ) (Open Access)Radboud Universiteit Nijmegen, 02 juni 2014Promotor : Lucas, P.J.F. Co-promotor : Hommersom, A.J.195 p
Time and Bayesian Networks
Full text linkContains fulltext :
194846.pdf (Publisher’s version ) (Open Access)Radboud University, 30 augustus 2018Promotor : Lucas, P.J.F. Co-promotores : Hommersom, A.J., Heijden, M. van derv, 187 p
Formal Methods for Verification of Clinical Practice Guidelines
Full text linkContains fulltext :
72447.pdf (Author’s version preprint ) (Open Access
Hybrid Probabilistic Logics: Theoretical Aspects, Algorithms and Experiments
Full text linkContains fulltext :
157121.pdf (Publisher’s version ) (Open Access)Steffen Michels
Hybrid Probabilistic Logics: Theoretical Aspects, Algorithms and Experiments
Probabilistic logics aim at combining the properties of logic, that is they provide a structured way of expressing knowledge and a mechanical way of reasoning about such knowledge, with the ability of probability theory to deal with uncertainty. Such probabilistic logics can serve as a basis to automate uncertainty reasoning, based on a structured and interpretable representation of knowledge. There is a wide spectrum of probabilistic logic languages, which differ in the fundamental balance between how expressive a language is and how hard it is to reason about knowledge expressed in the language. On the one hand, a language should be expressive enough to allow one to express all knowledge required to model a problem at hand. On the other hand, it should still be possible to draw useful conclusions in reasonable time.
In this thesis we propose probabilistic logics with a unique balance between expressiveness and hardness of reasoning, which perfectly matches the requirements for many problem domains. On the expressivity side, we want to support hybrid models, i.e. combining discrete and continuous factors. On the reasoning side, we want to preserve soundness of reasoning, which means that conclusions drawn from knowledge agree with that knowledge. This may seem a basic requirement, but all common inference methods for hybrid models are either restricted to only a small class of such models, or provide only unsound reasoning, based on approximations that come without any guarantees.
This thesis covers a wide range of results. We provide a theoretical basis for probabilistic logics with the properties outlined above, but we also propose efficient inference algorithms, which we evaluate theoretically and experimentally. We also design practical languages, suitable for knowledge representation, and show applicability using a challenging real-world problem.RU Radboud Universiteit, 04 mei 2016Promotor : Lucas, P.J.F. Co-promotores : Velikova, M.V., Hommerson, A.J.237 p
Adaptation of Clinical Practice Guidelines
Full text linkContains fulltext :
72650.pdf (author's version ) (Open Access
A Constraint-based Approach to Medical Guidelines and Protocols
Full text linkContains fulltext :
72742.pdf (Author’s version preprint ) (Open Access
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
- …
