Search

Objective: Machine learning techniques can be used to extract predictive models for diseases from electronic medical records (EMRs). However, the nature of EMRs makes it difficult to apply off-the-shelf machine learning techniques while still exploiting the rich content of the EMRs. In this paper, we explore the usage of a range of natural language processing (NLP) techniques to extract valuable predictors from uncoded consultation notes and study whether they can help to improve predictive performance.Methods: We study a number of existing techniques for the extraction of predictors from the consultation notes, namely a bag of words based approach and topic modeling. In addition, we develop a dedicated technique to match the uncoded consultation notes with a medical ontology. We apply these techniques as an extension to an existing pipeline to extract predictors from EMRs. We evaluate them in the context of predictive modeling for colorectal cancer (CRC), a disease known to be difficult to diagnose before performing an endoscopy.Results: Our results show that we are able to extract useful information from the consultation notes. The predictive performance of the ontology-based extraction method moves significantly beyond the benchmark of age and gender alone (area under the receiver operating characteristic curve (AUC) of 0.870 versus 0.831). We also observe more accurate predictive models by adding features derived from processing the consultation notes compared to solely using coded data (AUC of 0.896 versus 0.882) although the difference is not significant. The extracted features from the notes are shown be equally predictive (i.e. there is no significant difference in performance) compared to the coded data of the consultations.Conclusion: It is possible to extract useful predictors from uncoded consultation notes that improve predictive performance. Techniques linking text to concepts in medical ontologies to derive these predictors are shown to perform best for predicting CRC in our EMR dataset. [ABSTRACT FROM AUTHOR]

Abstract: When considering intelligent agents that interact with humans, having an idea of the trust levels of the human, for example in other agents or services, can be of great importance. Most models of human trust that exist assume trust in one trustee is independent of trust in another trustee. The model introduced here addresses so-called relative trust. The idea of relative trust is that trust in a certain trustee not only depends on the experiences with that trustee, but also on trustees that are perceived competitors of that trustee. Such models for relative trust contain parameters to represent the specific dependency between trust for different trustees. In order to tailor the model towards a specific human, dedicated parameter estimation techniques are used. The validation shows that such a model for relative trust is able to predict human trust based behavior significantly better compared to a benchmark model. [Copyright &y& Elsevier]

A model-based agent system model for medicine usage management is presented and formally analysed. The model incorporates an intelligent ambient agent model that has an explicit representation of a dynamical system model to estimate the medicine level in the patient s body by simulation, is able to analyse whether the patient intends to take the medicine too early or too late, and can take measures to prevent this. [ABSTRACT FROM AUTHOR]

Abstract: In this paper a computational model is presented for how a desire triggers responses and feelings. The model shows how these feelings can be biased, for example due to addicting experiences in the past. Both the strength of a response and of the associated feeling result from a converging dynamic pattern modeled by reciprocal causal interactions between the two. The model has been used to conduct a number of simulation experiments under varying circumstances. Moreover, it has been evaluated by formal analysis of emerging patterns entailed by the model. Furthermore, it has been pointed out how the computational model can be applied within an ambient agent system supporting a human in not being tempted. In a simple example scenario it is shown such an ambient agent system is able to predict and assess a human’s desire state, and use this assessment to suggest alternatives to avoid falling for certain temptations. [Copyright &y& Elsevier]

Abstract: Trust dynamics can be modeled in relation to experiences. In this paper two models to represent human trust dynamics are introduced, namely a model on a cognitive level and a neural model. These models include a number of parameters, providing the possibility to express certain relations between trustees. The behavior of each of the models is further analyzed by means of simulation experiments and formal verification techniques. Thereafter, both models have been compared to see whether they can produce patterns that are comparable. As each of the models has its own specific set of parameters, with values that depend on the type of person modeled, such a comparison is non-trivial. To address this, a special comparison approach is introduced, based on mutual mirroring of the models in each other. More specifically, for a given parameter values set for one model, by an automated parameter estimation procedure the most optimal values for the parameter values of the other model are determined in order to show the same behavior. Roughly spoken the results are that the models can mirror each other up to an accuracy of around 90%. [Copyright &y& Elsevier]

Abstract: This paper presents analysis and simulation of meta-reasoning processes based on an agent-based meta-level architecture for strategic reasoning in naval planning. The architecture was designed as a generic agent model and instantiated with decision knowledge acquired from naval domain experts and was specified as an executable agent-based model which has been used to perform a number of simulations. To evaluate the simulation results, relevant properties for the planning decision were identified and formalized. These properties have been validated for the simulation traces. [Copyright &y& Elsevier]

Abstract: This paper presents an agent-based approach for error detection in incident management organizations. The approach consists of several parts. First, a formal approach for the specification and hierarchical verification of both traces and properties. Incomplete traces are enriched by enrichment rules. Furthermore, a classification mechanism is presented for the different properties in incident management that is based on psychological literature. Classification of errors provides insight in the functioning of the agents involved with respect to their roles. This insight enables the provision of dedicated training sessions and allows software support to give appropriate warning messages during incident management. [Copyright &y& Elsevier]

Abstract: Within the field of incident management split second decisions have to be made, usually on the basis of incomplete and partially incorrect information. As a result of these conditions, errors occur in such decision processes. In order to avoid repetition of such errors, historic cases, disaster plans, and training logs need to be thoroughly analysed. This paper presents a formal approach for such an analysis that pays special attention to spatial and temporal aspects, to information exchange, and to organisational structure. The formal nature of the approach enables automation of analysis, which is illustrated by case studies of two disasters. [Copyright &y& Elsevier]

• Machine learning has the potential to improve the positive predictive value of neonatal screening for primary and central congenital hypothyroidism. • Currently, detection of central congenital hypothyroidism is accompanied by a high number of false-positives. • In order of importance, variables determining detection of congenital hypothyroidism were TSH, T4/TBG, (gestational) age, TBG, T4, and age at NBS sampling. • From the current routine screening parameters, T4/TBG is the most important parameter to detect central congenital hypothyroidism with a low positive predictive value. • New screening parameters are needed to improve the detection of central congenital hypothyroidism. The Dutch Congenital hypothyroidism (CH) Newborn Screening (NBS) algorithm for thyroidal and central congenital hypothyroidism (CH-T and CH-C, respectively) is primarily based on determination of thyroxine (T4) concentrations in dried blood spots, followed by thyroid-stimulating hormone (TSH) and thyroxine-binding globulin (TBG) measurements enabling detection of both CH-T and CH-C, with a positive predictive value (PPV) of 21%. A calculated T4/TBG ratio serves as an indirect measure for free T4. The aim of this study is to investigate whether machine learning techniques can help to improve the PPV of the algorithm without missing the positive cases that should have been detected with the current algorithm. NBS data and parameters of CH patients and false-positive referrals in the period 2007–2017 and of a healthy reference population were included in the study. A random forest model was trained and tested using a stratified split and improved using synthetic minority oversampling technique (SMOTE). NBS data of 4668 newborns were included, containing 458 CH-T and 82 CH-C patients, 2332 false-positive referrals and 1670 healthy newborns. Variables determining identification of CH were (in order of importance) TSH, T4/TBG ratio, gestational age, TBG, T4 and age at NBS sampling. In a Receiver-Operating Characteristic (ROC) analysis on the test set, current sensitivity could be maintained, while increasing the PPV to 26%. Machine learning techniques have the potential to improve the PPV of the Dutch CH NBS. However, improved detection of currently missed cases is only possible with new, better predictors of especially CH-C and a better registration and inclusion of these cases in future models. [ABSTRACT FROM AUTHOR]

Reinforcement Learning (RL) has recently found many applications in the healthcare domain thanks to its natural fit to clinical decision-making and ability to learn optimal decisions from observational data. A key challenge in adopting RL-based solution in clinical practice, however, is the inclusion of existing knowledge in learning a suitable solution. Existing knowledge from e.g. medical guidelines may improve the safety of solutions, produce a better balance between short- and long-term outcomes for patients and increase trust and adoption by clinicians. We present a framework for including knowledge available from medical guidelines in RL. The framework includes components for enforcing safety constraints and an approach that alters the learning signal to better balance short- and long-term outcomes based on these guidelines. We evaluate the framework by extending an existing RL-based mechanical ventilation (MV) approach with clinically established ventilation guidelines. Results from off-policy policy evaluation indicate that our approach has the potential to decrease 90-day mortality while ensuring lung protective ventilation. This framework provides an important stepping stone towards implementations of RL in clinical practice and opens up several avenues for further research. • Reinforcement learning based on observational data and existing knowledge. • Treatment advice and physiological knowledge in guidelines are injected into learner. • Ventilation case study indicates adherence to guidelines, clinician outperformance. [ABSTRACT FROM AUTHOR]

Abstract: Within agent-based Ambient Intelligence applications agents react to humans based on information obtained by sensoring and their knowledge about human functioning. Appropriate types of reactions depend on the extent to which an agent understands the human and is able to interpret the available information (which is often incomplete, and hence multi-interpretable) in order to create a more complete internal image of the environment, including humans. Such an understanding requires that the agent has knowledge to a certain depth about the human’s physiological and mental processes in the form of an explicitly represented model of the causal and dynamic relations describing these processes. In addition, given such a model representation, the agent needs reasoning methods to derive conclusions from the model and interpret the (partial) information available by sensoring. This paper presents the development of a toolbox that can be used by a modeller to design Ambient Intelligence applications. This toolbox contains a number of model-based reasoning methods and approaches to control such reasoning methods. Formal specifications in an executable temporal format are offered, which allows for simulation of reasoning processes and automated verification of the resulting reasoning traces in a dedicated software environment. A number of such simulation experiments and their formal analysis are described. The main contribution of this paper is that the reasoning methods in the toolbox have the possibility to reason using both quantitative and qualitative aspects in combination with a temporal dimension, and the possibility to perform focused reasoning based upon certain heuristic information. [Copyright &y& Elsevier]

Introduction: In recent years, reinforcement learning (RL) has gained traction in the healthcare domain. In particular, RL methods have been explored for haemodynamic optimization of septic patients in the Intensive Care Unit. Most hospitals however, lack the data and expertise for model development, necessitating transfer of models developed using external datasets. This approach assumes model generalizability across different patient populations, the validity of which has not previously been tested. In addition, there is limited knowledge on safety and reliability. These challenges need to be addressed to further facilitate implementation of RL models in clinical practice.Method: We developed and validated a new reinforcement learning model for hemodynamic optimization in sepsis on the MIMIC intensive care database from the USA using a dueling double deep Q network. We then transferred this model to the European AmsterdamUMCdb intensive care database. T-Distributed Stochastic Neighbor Embedding and Sequential Organ Failure Assessment scores were used to explore the differences between the patient populations. We apply off-policy policy evaluation methods to quantify model performance. In addition, we introduce and apply a novel deep policy inspection to analyse how the optimal policy relates to the different phases of sepsis and sepsis treatment to provide interpretable insight in order to assess model safety and reliability.Results: The off-policy evaluation revealed that the optimal policy outperformed the physician policy on both datasets despite marked differences between the two patient populations and physician's policies. Our novel deep policy inspection method showed insightful results and unveiled that the model could initiate therapy adequately and adjust therapy intensity to illness severity and disease progression which indicated safe and reliable model behaviour. Compared to current physician behavior, the developed policy prefers a more liberal use of vasopressors with a more restrained use of fluid therapy in line with previous work.Conclusion: We created a reinforcement learning model for optimal bedside hemodynamic management and demonstrated model transferability between populations from the USA and Europe for the first time. We proposed new methods for deep policy inspection integrating expert domain knowledge. This is expected to facilitate progression to bedside clinical decision support for the treatment of critically ill patients. [ABSTRACT FROM AUTHOR]