Modeling clinical pathways: MSc defense Lonneke Vermeulen

imageLonneke Vermeulen has defended successfully her M.Sc. thesis titled “A Process Modelling Method for Care Pathways”. She developed a methodology for modeling care pathways using BPMN. Lonneke studied Operations Management & Logistics, and her project was a collaboration with Catharina Hospital Eindhoven.

The goal of this research is to design a process modelling method for care pathways (CP) in hospitals, based on the existing literature of the field as well as experiences from practice that is applicable on any kind of pathway. The method focuses on setting the right requirements for the modelling language and tool such that the model can decide on the best possible model for the project starting from the goals (e.g. communication tool, mapping, checklists). Furthermore, special attention is paid on the missing literature aspects of information gathering necessary to model the CP, the relationship between the goals of the model and the necessary granularity levels, and how to set those granularity levels.

Financial forecasting for healthcare insurers: MSc defense Sebastiaan van Zelst

Today, Sebastiaan van Zelst defended successfully his M.Sc. thesis titled “Defining a Financial Forecasting Model for Healthcare Insurance Companies: a collaborative Markov chain approach incorporating institutional care pathway traversal”. Sebastiaan studied Business Information Systems and his project was a collaboration with KPMG.

This document concerns an exploratory research towards the application of collaborative Markov chains as a forecasting model within the field of healthcare insurance. The model proposed is based on both predicting care demand and associated institutional pathway traversal. A system of collaborative Markov chains allows the user to jointly model several probabilistic elements that share dependencies. It describes a collection of Markov chains in which the state of a certain chain within the collection influences transition probabilities in other chains within the collection. It allows the use of different types of techniques to estimate forecasting parameters as an input within one model. It entails a modular structure which allows the user to perform case-based analyses. Simulation of simplified proof-of-concept cases has shown accurate predictive behavior. Due to the fact that Markov chains and consequently systems of collaborative Markov chains are probabilistic in nature, simulation of a sufficient number of sampling replication yields results that tend to follow a Normal distribution. The statistical nature of the simulation results lends itself perfectly for consecutive statistical post-processing. Systems of collaborative Markov chains provide in modeling complex probabilistic systems in which several dependencies might exist. Current challenges within the application of systems of collaborative Markov chains involve the complexity in terms of the number of parameters to estimate and associated running times. Additionally the existence of potential “inactive elements” with respect to the field of healthcare insurance introduces additional challenges in parameter estimation of the model.

RFID technology in outpatient logistics: MSc defense Burak Onurlu

Burak OnurluBurak Onurlu has defended successfully his M.Sc. thesis titled “RFID Technology in Outpatient Logistics: an analysis of its potential and acceptance ”. As student of Innovation Management, Burak studied the potential of RFID technology in outpatient logistics and the acceptance of the technology by healthcare practitioners.

The dramatic and continuous increase in health care spending threatens the economic stability and safety of the countries. Therefore, the governments and the health care organizations are forced to rethink and redesign their strategies in order to provide timely and cost effective healthcare services. Within this perspective, today’s technological developments have drawn attention of decision makers in health care industry due to their potential benefits in terms of efficiency and cost reduction. In this research, we focus on the main outpatient logistics problems and the recent use of a relatively new technology, Radio Frequency Identification (RFID), in order to solve those problems. In addition, the main concentration of the research is to explore the potential and the adoption of RFID in outpatient logistics. 109 Dutch health care professionals with diverse backgrounds, from doctors and nurses to managers and IT specialists, participated in a web-survey which was based on one of the most popular technology adoption theories, the Technology Acceptance Model. The proposed research model and the collected data were analyzed by using Structural Equation Modeling. The results of the research shows that perceived ease of use and perceived usefulness are the significant determinants of behavioral intention to use RFID technology in outpatient logistics. They can also explain/predict 67% of the variance in health care professionals’ behavioral intention to use RFID technology in outpatient logistics. Moreover, based on the Method Evaluation Model, the results suggest that RFID technology will likely be adopted in practice by health care professionals.

Performance measurement for clinical pathways: MSc defense Parvathy Meenakshy

Today, Parvathy Meenakshy defended successfully her M.Sc. thesis titled “A Performance Measurement Framework for Clinical Pathways Monitoring”. Parvathy studied Business Information Systems and her project was a collaboration with Catharina Hospital Eindhoven.

Healthcare domain around the world is facing challenges due to the skyrocketing cost, chronic illness, ever increasing aging population, sedentary life style and changing demographics. There is an intense pressure to balance cost and quality of care. Several researches are conducted to investigate the benefits of adopting process orientation and process oriented information systems in healthcare, which ascertain the fact that process orientation can improve the quality of care, efficiency and cost effectiveness. Clinical pathway is an example of process management tool in healthcare to streamline process and standardize care. The actual execution and continuous monitoring of the pathway is of utmost importance for improvement. The monitoring of the pathway is essentially the measurement of indicators and its reporting to analyze the performance of the pathway. In this thesis, we develop a framework for performance measurement system for clinical pathway monitoring. The system uses an ontology of pathway indicators with a formal method to define them. A proof of concpet system has also been implemented.

Machine learning for medical prediction: MSc defense Luis De Luna Orozco

Luis De LunaLuis Jorge De Luna Orozco has defended successfully his M.Sc. thesis titled “Performance of Machine Learning Algorithms to Predict Anastomotic Failure in Bariatric Surgery”. Luis studied Business Information Systems and his project was a collaboration with Catharina Hospital Eindhoven.

The main purpose of this report is to perform a machine learning study to find out to which degree these algorithms can predict occurrence of anastomotic failure in bariatric surgery. The information used in this study was obtained from two databases provided by the Catharina Hospital in Eindhoven which contain preoperative information about patients (Sleeves) and perioperative blood pressure (4KP) recorded for 1116 surgeries since 2006. First part of this study covers an analysis to find out which features or variables can be obtained from this set of data and to which degree they have predictive power to identify complications after bariatric surgery, more specifically intestinal leakage when using them in classification predictive models.

Second, investigation and performance analysis of three different machine learning classification algorithms is carried using techniques to compensate imbalanced sets of data due to the fact that only 3% of the patients present anastomotic failure after surgery. Finally relation between hypotension episodes an anastomotic failure is studied and explained.

Main results of this study show that Random Forest classifying technique can offer prediction accuracy up to 90% when taking into consideration perioperative and preoperative bariatric surgery information like smoking assessment, surgery duration, medication assessment, surgery technique and occurrence of hypotension episodes under sedation. At the end all these results are discussed and a Random Forest model in WEKA is provided, moreover possible future work is suggested to continue with the creation of a full clinical decision support system.

Comparison of workflow support and clinical decision support systems: MSc defense Xuchen Wang

Xuchen ThesisXuchen Wang has successfully defended her M.Sc. thesis titled “A Comparison of Workflow Management Systems and Clinical Decision Support Systems in Supporting Clinical Processes”. Xuchen studied Operations Management and Logistics.

Healthcare institutions are facing the challenge to deliver safe care against affordable costs. Both workflow management systems and clinical decision support systems are powerful tools in enhance clinical services but from two different fields. This projects aims to investigate the commonalities and differences of applying workflow management systems (WFMSs) and clinical decision support systems (CDSSs) in clinical environment. By conducting theoretical analyses and practical implementations, the functionalities and the advantage and disadvantages of both systems will be discussed. In the end of the project, the proposal of combining the two types of systems to achieve better clinical care is put forward.

Analysis of operative times in bariatric surgery: MSc defense Pablo Perdiguer

Pablo PerdiguerPablo Perdiguer has successfully defended his M.Sc. thesis titled “An Analysis of Operative Times in Bariatric Surgery and Modelling for Predicting the Consequences of Intra-Operative Complications in Anesthetic Procedures: data mining application in healthcare”. Pablo was an exchange student from Spain and studied Operations Management and Logistics in Eindhoven. His project was a collaboration with Catharina Hospital Eindhoven.

In this report, two different studies are carried out based on data obtained from the 4kp, a database at the Catharina Hospital in Eindhoven. The first analysis addresses the investigation of operative times in bariatric surgery, with a special focus on understanding their variation over recent years. Initially, some background is provided by a look at some of the literature on bariatric surgery, which is followed by an in-depth treatment of data mining. Indeed, the use of data mining techniques is one of the main features of this project and, in particular, the application of regression analysis and k-nearest neighbor algorithm for the prediction and classification of operative times.

Secondly, a modeling for the prediction of the consequences of intraoperative complications in anesthesia-related procedures is completed. Before the data mining techniques are applied, the complications and their evolution over time are explored, which serves as a first step in the subsequent analysis. For the study of the complications of complications, the accuracy of the algorithms involving decision tree classification and neuronal network prediction is tested, and alternative research models are also presented. All processes are part of the Knowledge Discovery in Databases framework, which provided a structured methodology with which to work.

The two aforementioned studies are part of work performed with the database, which is referred to as the extraction information. This process is fully supported by practicing doctors from the Catharina Hospital who contributed with their expert knowledge in order to better understand and discuss the results obtained. The information extracting procedure is aimed at providing useful results on bariatric surgery operative times and intraoperative complications in anesthesia-related operations, both general and bariatric-specific. In addition, the understanding of changes in operative times leads to the identification of what extra information is needed to improve the quality of the studies, and the modeling of complications might be helpful for practitioners in order to better comprehend the nature of patients’ complications.

All the studies discussed here are based upon prior work on the database, which can be regarded as the creating information phase. In this process, a well-filtered bariatric surgery database was created, and some guidelines for improving the current database, encompassing both data collection and data available, are given. Furthermore, data cleaning is done and transformations of the database through workflows that can be easily checked (automation of the process) are made available.

Finally, all results are discussed, including the incorporation of expert knowledge and advice, and the possibility of future work and its directions is addressed. In addition, a CD containing useful information and data used in this research is made available along with a hard copy of the project. Specifically, it contains Excel files with data directly extracted from the database, KNIME projects where workflows are created and this paper itself, accompanied by the final presentation in PowerPoint format.

Process mining in the anesthesia care: MSc defense Arnau Carbonell

Arnau CarboneMasterThesisFinalll has successfully defended his M.Sc. thesis titled “Analysis of the Treatment of Pain and Anxiety in the Anesthesia Care in an ERCP: a process mining application in heath care”. Arnau was an exchange student from Spain and studied Operations Management and Logistics in Eindhoven. His project was a collaboration with Harvard affiliated Beth Israel Deaconess Medical Center in Boston.

This work is an initial study into the application of process mining techniques in a clinical environment. Healthcare processes have to deal with an extraordinary uncertainty and healthcare organizations, because of their processes, are seen as highly dynamic, complex, ad hoc, and multidisciplinary. Unfortunately, process mining techniques are usually not meant for the medical environment, since they are more likely to be used in administrative processes. Therefore, in this study we search for appropriate mining methods to be applied in the medical setting. To carry out the study, two databases of the anesthesia care of ERCP (Endoscopic Retrograde Cholangio Pancreatography) are used. The study shows that existing process mining methods have a limited application with clinical data.