Hochschule Trier, Umwelt-Campus Birkenfeld
+49 6782 17-1727
g.dartmann@umwelt-campus.de

Publikationen

Abstract:

Medical guidelines have a significant role in the field of evidence-based medical treatment. The content of a medical guideline is based on a systematic review of clinical evidence with instructions and recommendations that clinicians can refer to. Most of the guidelines are available in an unstructured text format. Hence, clinicians must take a considerable time to search and find relevant recommendations in their semantic context. Using Machine Learning algorithms, automatic information extraction from medical guidelines has recently become possible. We present a novel system for information extraction and a fuzzy rule database developed for clinical guidelines.

The proposed system, dubbed NLP-FUZZY, combines capabilities of Natural Language Processing (NLP) and Fuzzy Logic approaches. First, the NLP-FUZZY performs a semantic extraction of medical guidelines using a bi-directional Long short-term memory (LSTM). Subsequently, using the extracted semantic, it creates fuzzy rules, which are able to recognize new cases in a learning domain while predicting and extract the grade of recommendation. In order to test the NLP-FUZZY system, we compared its performance with state-of-the-art NLP approaches for clinical information extraction.
 
L. B. Fazlic, A. Hallawa, A. Schmeink, A. Peine, L. Martin and G. Dartmann, „A Novel NLP-FUZZY System Prototype for Information Extraction from Medical Guidelines,“ 2019 42nd International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO), Opatija, Croatia, 2019, pp. 1025-1030.
 
Published in:
Date of Conference:
20-24 May 2019
Date Added to IEEE Xplore:
11 July 2019
 
ISBN Information:
Electronic ISBN: 978-953-233-098-4
Print on Demand(PoD) ISBN: 978-1-5386-9296-7
Electronic ISSN: 2623-8764
 
INSPEC Accession Number:
18820538
DOI:
 
Publisher:
IEEE
 
Conference Location:
Opatija, Croatia, Croatia
 
Funding Agency:
Bundesministerium für Bildung und Forschung – IMEDALytics project

Abstract:

This letter presents an adaptive learning framework for estimating structural parameters in stochastic state-based models (SSMs). SSMs are a useful modeling tool in systems biology and medicine. While models in these disciplines are traditionally hand-crafted, an automated generation based on experimental data becomes a topic of research interest. In particular, our goal is to classify measured processes using the generated models. An innovative likelihood-based adaptive learning approach capable of learning the structural parameters, i.e., the arc weights of SSMs from data and exploiting the reliability of detected inputs is presented in this letter. Its convergence behavior is analyzed and an expression for the error at steady state is derived. Simulations assess the performance of the proposed and existing algorithms for a gene regulatory network.
 
P. M. Vieting, R. C. de Lamare, L. Martin, G. Dartmann and A. Schmeink, „Likelihood-Based Adaptive Learning in Stochastic State-Based Models,“ in IEEE Signal Processing Letters, vol. 26, no. 7, pp. 1031-1035, July 2019.
 

Published in:
IEEE Signal Processing Letters (Volume: 26, Issue: 7, July 2019)

Page(s):
1031 – 1035

Date of Publication:
17 May 2019

ISSN Information:

Print ISSN: 1070-9908
Electronic ISSN: 1558-2361
 
INSPEC Accession Number:
18715550
 

 

Funding Agency:

10.13039/501100002347-Bundesministerium für Bildung und Forschung; IMEDALytics project;

Find also on: ice.rwth-aachen.de/publications

Abstract:

In this work, a new hybrid algorithm for disease risk classification is proposed. The proposed methodology is based on Dynamic Time Warping (DTW). This methodology can be applied to time series from various domains such as vital sign time series available in medical big data. To validate our methodology, we applied it to risk classification for sepsis, which is one of the most challenging problems within the area of medical data analysis. In the first step the algorithm uses different statistical properties of time series features. Furthermore, using differently labeled training data sets, we created a DTW Barycenter Averaging (DBA) on each feature.

In the second step, validation data sets and DTW are used to validate the precision of classification and the final results are compared. The performance of our methodology is validated with real medical data and on six different criteria definitions for the sepsis diseases. Results show that our algorithm performed, in the best case, with precision and recall of 96,38% and 90,90%, respectively.

Lejla Begic Fazlic, Ahmed Hallawa, Matthias Dziubany, Marlies Morgen, Jens Schneider, Marvin Schacht, Anke Schmeink, Lukas Martin, Arne Peine, Thomas Vollmer, Stefan Winter and Guido Dartmann, 8th International Conference on Cyber – Physical Systems and Internet-of-Things (CPS & IoT‘ 2020), Budva, Montenegro, 2020.

Date of Conference:

8-11 June 2020

Date Added to IEEE Xplore:
07 July 2019
 
ISBN Information:
Electronic ISBN: 978-1-7281-6949-1
Print on Demand(PoD) ISBN: 978-1-7281-6950-7
Electronic ISSN: 2637-9511 
 
DOI:
 
Publisher:
IEEE
 
Conference Location:
Budva, Montenegro
 
Funding Agency:
Bundesministerium für Bildung und Forschung – IMEDALytics project

Abstract:

Intensive Care Unit (ICU) professionals have to make lifesaving therapy decisions promptly under high stress and uncertainty. Clinical Decision Support Systems (CDSS) can improve the quality of healthcare by identifying complex statistical connections between patients‘ parameters and by rapidly presenting the statistically most promising treatment options to physicians. However, HCI aspects are rarely considered when developing CDSSs. This paper describes a field study conducted in three ICUs investigating how physicians and nurses form (volume) therapy decisions and monitor their success. Our findings reveal a continuous decision cycle in which nurses and physicians collaborate synchronously and asynchronously to provide optimal care. Furthermore, the desire to understand how a CDSS generated recommendations varies depending on the user’s goals and other contextual factors such as workload. These findings show that CDSSs for ICUs need to (1) specifically facilitate collaboration and (2) support adaptation of the interface to both context and users.
 
Annika Kaltenhauser, Verena Rheinstädter, Andreas Butz, Dieter P. Wallach, „You Have to Piece the Puzzle Together“ — Implications for Designing Decision Support in Intensive Care, DIS ’20: Proceedings of the 2020 ACM Designing Interactive Systems Conference, Eindhoven, Netherlands, 2020, pp. 1509-1522.
 
Published in:
Date of Conference:
July 2020
 
ISBN Information:
978-1-4503-6974-9
 
DOI:
 
Publisher:
Association for Computing Machinery, New York, NY, United States
 
Conference Location:
Eindhoven, Netherlands
 
Funding Agency:
Bundesministerium für Bildung und Forschung – IMEDALytics project

Abstract:

We present the paradigm of Expert of Oz studies for the formative evaluation of hypothetical AI systems. These studies follow the principle of Wizard of Oz studies but use a human expert for simulating the AI. This allows the experimenter to not only investigate the user’s behavior with and reaction to such a system but also to observe and analyze the performance of the expert in the situation at hand. As a consequence, we can learn both about interaction with the hypothetical AI system and about the required AI functionality at the same time and in realistic interaction scenarios.
 
Andreas Butz, Annika Kaltenhauser, Malin Eiband, The Expert of Oz: A Two-sided Study Paradigm for Intelligent Systems, DIS‘ 20 Companion: Companion Publication of the 2020 ACM Designing Interactive Systems Conference, Eindhoven, Netherlands, 2020, pp. 269-273.
 
Published in:
Date of Conference:
July 2020
 
ISBN Information:
978-1-4503-7987-8
 
DOI:
 
Publisher:
Association for Computing Machinery, New York, NY, United States
 
Conference Location:
Eindhoven, Netherlands
 
Funding Agency:
Bundesministerium für Bildung und Forschung – IMEDALytics project

Abstract:

Integrating Artificial Intelligence (AI) technologies promises to open new possibilities for the development of smart systems and the creation of positive user experiences. While the acronym «AI»has often been used inflationary in recent marketese advertisements, the goal of the paper is to explore the relationship of AI and UX in concrete detail by referring to three case studies from our lab. The first case study is taken from a project targeted at the development of a clinical decision support system, while the second study focuses on the development of an autonomous mobility-on-demand system. The final project explores an innovative, AI-injected prototyping tool. We discuss challenges and the application of available guidelines when designing AI-based systems and provide insights into our learnings from the presented case studies.
 
Dieter P. Wallach, Lukas A. Flohr, Annika Kaltenhauser, Beyond the Buzzwords: On the Perspective of AI in UX and Vice Versa, First International Conference, AI-HCI 2020, Held as Part of the 22nd HCI International Conference, HCII 2020, Copenhagen, Denmark, 2020, pp. 146-166
 
Published in:
Date of Conference:
July 19-24, 2020
 
ISBN Information:
Online ISBN: 978-3-030-50334-5
Print ISBN: 978-3-030-50333-8
Series Online ISSN: 1611-3349
Series Print ISSN: 0302-9743
 
DOI:
 
Publisher:
Springer, Cham
 
Conference Location:
Copenhagen, Denmark
 
Funding Agency:
Bundesministerium für Bildung und Forschung – IMEDALytics project

Abstract

Decision support systems in intensive care units are developed with safety, efficiency, and effectiveness in mind. In contrast, user experience (UX) for decision support systems has received limited attention in practice and research. In this paper, we present an application of the Experience Design approach from Hassenzahl [8] for a clinical decision support system (CDSS) for volume therapy – the administration of intravenous fluids – in intensive care units. In semi-structured interviews with five nurses, we gathered nurses‘ work activities around volume therapy, which are perceived as particularly beneficial for their well-being. Lead by the narratives gathered in the interviews we designed positive experience through the CDSS. The resulting concept of a CDSS for nurses aims to create a positive experience during volume therapy, which is supported by addressing and fulfilling the needs for competence and popularity.

Published in:
MuC’20: Mensch und Computer 2020 Magdeburg, Germany 
 
Date of Conference:
September 2020
 
ISBN Information:
978-1-4503-7540-5
 
DOI:
 
Publisher:
Association for Computing Machinery, New York, NY, United States
 
Conference Location:
Magdeburg, Germany
 
Funding Agency:
Bundesministerium für Bildung und Forschung – IMEDALytics project

Abstract

This paper presents an incremental learning approach for estimating the structural parameters in stochastic state-based models (SSMs). SSMs have proven to be useful for modelling biological and medical processes, as they can represent both time dependency and stochastic processes. A major challenge in modelling in bioinformatics is that learning processes usually rely on large publicly accessible databases. In this work ,a new approach is presented, where models are trained incrementally locally at different data sources, e.g., hospitals, without having to pass on sensitive data. After learning, only the parameters of the model are passed on, in this case the arc weights of stochastic Petri nets. As a result, data protection and privacy of patients in hospitals are respected and it is no longer necessary to rely on the existence of a suitable accessible database. Simulations are used to evaluate the performance of the algorithm for a gene regulatory network.

R. Lipp, G. Dartmann, L. Fazlic, T. Vollmer, S. Winter, A. Peine, L. Martin and A. Schmeink, „Incremental Parameter Estimation of Stochastic State-Based Models“, IEEE 19th World Symposium on Applied Machine Intelligence and Informatics (SAMI2021), Herl’any, Slovakia, 2021.

Published in:
 
Date of Conference:
21-23 January 2021
 
 
Publisher:
IEEE
 
Conference Location:
Herl’any, Slovakia
 
Funding Agency:
Bundesministerium für Bildung und Forschung – IMEDALytics project

Abstract

The aim of this work was to develop and evaluate the reinforcement learning algorithm VentAI, which is able to suggest a dynamically optimized mechanical ventilation regime for critically-ill patients. We built, validated and tested its performance on 11,943 events of volume-controlled mechanical ventilation derived from 61,532 distinct ICU admissions and tested it on an independent, secondary dataset (200,859 ICU stays; 25,086 mechanical ventilation events). A patient „data fingerprint“ of 44 features was extracted as multidimensional time series in 4-hour time steps. We used a Markov decision process, including a reward system and a Q-learning approach, to find the optimized settings for positive end-expiratory pressure (PEEP), fraction of inspired oxygen (FiO2) and ideal body weight-adjusted tidal volume (Vt). The observed outcome was in-hospital or 90-day mortality. VentAI reached a significantly increased estimated performance return of 83.3 (primary dataset) and 84.1 (secondary dataset) compared to physicians‘ standard clinical care (51.1). The number of recommended action changes per mechanically ventilated patient constantly exceeded those of the clinicians. VentAI chose 202.9% more frequently ventilation regimes with lower Vt (5-7.5 mL/kg), but 50.8% less for regimes with higher Vt (7.5-10 mL/kg). VentAI recommended 29.3% more frequently PEEP levels of 5-7 cm H2O and 53.6% more frequently PEEP levels of 7-9 cmH2O. VentAI avoided high (>55%) FiO2 values (59.8% decrease), while preferring the range of 50-55% (140.3% increase). In conclusion, VentAI provides reproducible high performance by dynamically choosing an optimized, individualized ventilation strategy and thus might be of benefit for critically ill patients.

Arne Peine, Ahmed Hallawa,  Johannes Bickenbach,  Guido Dartmann, Lejla Begic Fazlic, Anke Schmeink, Gerd Ascheid, Christoph Thiemermann, Andreas Schuppert, Ryan Kindle, Leo Celi, Gernot Marx  & Lukas Martin, „Development and validation of a reinforcement learning algorithm to dynamically optimize mechanical ventilation in critical care“, npj Digit. Med. 4, 32 (2021).

 
 
Publisher:
Nature Research Journal  npj digital medicine
Funding Agency:
Project is partly founded  by Bundesministerium für Bildung und Forschung – IMEDALytics project
 

 

Abstract:

In this paper, we propose a method we call MEMEography for HCI research to understand people and their interactional contexts from the remixed internet memes they post in internet communities. While memes might not be the most obvious choice of a research subject, they allow us to investigate unfamiliar domains even when access to the field is beyond reach. We describe an initial approach of data selection, collection, prioritization and analysis. In addition, we demonstrate the kinds of insights we can gain through MEMEographies by analyzing a corpus of memes in the intensive care unit (ICU) context posted 2020 on Instagram. ICU memes open up insights into the environment, work practices, challenges, emotions and familiarized us with ICU practitioners’ language, even though access to an actual ICU was completely impossible during 2020.

Annika Kaltenhauser, Nađa Terzimehić, Andreas Butz, „MEMEography: Understanding Users Through Internet Memes“, CHI EA ’21: Extended Abstracts of the 2021 CHI Conference on Human Factors in Computing Systems, Japan (remote), 2021.

 
 
Published in:
Date of Conference:
May 2021
 
ISBN Information:
 
Publisher:
Association for Computing Machinery, New York, NY, United States
Conference Location: Yokohama, Japan (Remote)
 
Conference Location:
Japan
 
Funding Agency:
Bundesministerium für Bildung und Forschung – IMEDALytics project
 

Abstract:

This paper presents an analytic learning approach to estimate the kinetic parameters in stochastic Petri Nets. Stochastic Petri Nets are very useful to describe time dependent processes in medical and biomedical applications. To be able to model processes more accurately and reliably without prior knowledge of the process, it is essential to automatically build these mathematical models based on measured data. Besides the structural parameters, stochastic Petri Nets are defined by a set of kinetic parameters that give the model its dynamics. Efficient and reliable estimation of these kinetic parameters is enormously important for the simulation of the model to output useful results. We therefore present a new method for determining these kinetic parameters based on estimation theory. Simulations show that our new algorithm is similarly accurate but faster by a factor of 7600 compared to an evolutionary algorithm.

D. Schuck, R. Lipp, L. B. Fazlic, G. Dartmann and A. Schmeink, „Estimation of Kinetic Parameters in Stochastic Biomedical Models Using Estimation Theory”, IEEE 19th International Symposium on Intelligent Systems and Informatics (SISY 2021), Subotica, Serbia, September 2021, accepted.

Published in:

IEEE 19th International Symposium on Intelligent Systems and Informatics (SISY 2021)

Date if Conference:

September 16-18, 2021

Conference Location: Subotica, Serbia

 
Funding Agency:
Bundesministerium für Bildung und Forschung – IMEDALytics project