Visual Assistance in Development and Validation of Bayesian Networks for Clinical Decision Support.

The development and validation of Clinical Decision Support Models (CDSM) based on Bayesian networks (BN) is commonly done in a collaborative work between medical researchers providing the domain expertise and computer scientists developing the decision support model. Although modern tools provide facilities for data-driven model generation, domain experts are required to validate the accuracy of the learned model and to provide expert knowledge for fine-tuning it while computer scientists are needed to integrate this knowledge in the learned model (hybrid modeling approach). This generally time-expensive procedure hampers CDSM generation and updating. To address this problem, we developed a novel interactive visual approach allowing medical researchers with less knowledge in CDSM to develop and validate BNs based on domain specific data mainly independently and thus, diminishing the need for an additional computer scientist. In this context, we abstracted and simplified the common workflow in BN development as well as adjusted the workflow to medical experts' needs. We demonstrate our visual approach with data of endometrial cancer patients and evaluated it with six medical researchers who are domain experts in the gynecological field.

Improved preoperative risk stratification in endometrial carcinoma patients: external validation of the ENDORISK Bayesian network model in a large population-based case series.

PURPOSE: Preoperative risk stratification of newly diagnosed endometrial carcinoma (EC) patients has been hindered by only moderate prediction performance for many years. Recently ENDORISK, a Bayesian network model, showed high predictive performance. It was the aim of this study to validate ENDORISK by applying the model to a population-based case series of EC patients. METHODS: ENDORISK was applied to a retrospective cohort of women surgically treated for EC from 2003 to 2013. Prediction accuracy for LNM as well as 5-year DSS was investigated. The model's overall performance was quantified by the Brier score, discriminative performance by area under the curve (AUC). RESULTS: A complete dataset was evaluable from 247 patients. 78.1% cases were endometrioid histotype. The majority of patients (n = 156;63.2%) had stage IA disease. Overall, positive lymph nodes were found in 20 (8.1%) patients. Using ENDORISK predicted probabilities, most (n = 156;63.2%) patients have been assigned to low or very low risk group with a false-negative rate of 0.6%. AUC for LNM prediction was 0.851 [95% confidence interval (CI) 0.761-0.941] with a Brier score of 0.06. For 5-year DSS the AUC was 0.698 (95% CI 0.595-0.800) as Brier score has been calculated 0.09. CONCLUSIONS: We were able to successfully validate ENDORISK for prediction of LNM and 5-year DSS. Next steps will now have to focus on ENDORISK performance in daily clinical practice. In addition, incorporating TCGA-derived molecular subtypes will be of key importance for future extended use. This study may support further promoting of data-based decision-making tools for personalized treatment of EC.

External validation study of endometrial cancer preoperative risk stratification model (ENDORISK).

Introduction: Among industrialized countries, endometrial cancer is a common malignancy with generally an excellent outcome. To personalize medicine, we ideally compile as much information as possible concerning patient prognosis prior to effecting an appropriate treatment decision. Endometrial cancer preoperative risk stratification (ENDORISK) is a machine learning-based computational Bayesian networks model that predicts lymph node metastasis and 5-year disease-specific survival potential with percentual probability. Our objective included validating ENDORISK effectiveness in our patient cohort, assessing its application in the current use of sentinel node biopsy, and verifying its accuracy in advanced stages. Methods: The ENDORISK model was evaluated with a retrospective cohort of 425 patients from the University Hospital Brno, Czech Republic. Two hundred ninety-nine patients were involved in our disease-specific survival analysis; 226 cases with known lymph node status were available for lymph node metastasis analysis. Patients were included undergoing either pelvic lymph node dissection (N = 84) or sentinel node biopsy (N =70) to explore the accuracy of both staging procedures. Results: The area under the curve was 0.84 (95% confidence interval [CI], 0.77-0.9) for lymph node metastasis analysis and 0.86 (95% CI, 0.79-0.93) for 5-year disease-specific survival evaluation, indicating quite positive concordance between prediction and reality. Calibration plots to visualize results demonstrated an outstanding predictive value for low-risk cancers (grades 1-2), whereas outcomes were underestimated among high-risk patients (grade 3), especially in disease-specific survival. This phenomenon was even more obvious when patients were subclassified according to FIGO clinical stages. Conclusions: Our data confirmed ENDORISK model's laudable predictive ability, particularly among patients with a low risk of lymph node metastasis and expected favorable survival. For high-risk and/or advanced stages, the ENDORISK network needs to be additionally trained/improved.

Automated color detection in orchids using color labels and deep learning.

The color of particular parts of a flower is often employed as one of the features to differentiate between flower types. Thus, color is also used in flower-image classification. Color labels, such as 'green', 'red', and 'yellow', are used by taxonomists and lay people alike to describe the color of plants. Flower image datasets usually only consist of images and do not contain flower descriptions. In this research, we have built a flower-image dataset, especially regarding orchid species, which consists of human-friendly textual descriptions of features of specific flowers, on the one hand, and digital photographs indicating how a flower looks like, on the other hand. Using this dataset, a new automated color detection model was developed. It is the first research of its kind using color labels and deep learning for color detection in flower recognition. As deep learning often excels in pattern recognition in digital images, we applied transfer learning with various amounts of unfreezing of layers with five different neural network architectures (VGG16, Inception, Resnet50, Xception, Nasnet) to determine which architecture and which scheme of transfer learning performs best. In addition, various color scheme scenarios were tested, including the use of primary and secondary color together, and, in addition, the effectiveness of dealing with multi-class classification using multi-class, combined binary, and, finally, ensemble classifiers were studied. The best overall performance was achieved by the ensemble classifier. The results show that the proposed method can detect the color of flower and labellum very well without having to perform image segmentation. The result of this study can act as a foundation for the development of an image-based plant recognition system that is able to offer an explanation of a provided classification.

Preoperative risk stratification in endometrial cancer (ENDORISK) by a Bayesian network model: A development and validation study.

BACKGROUND: Bayesian networks (BNs) are machine-learning-based computational models that visualize causal relationships and provide insight into the processes underlying disease progression, closely resembling clinical decision-making. Preoperative identification of patients at risk for lymph node metastasis (LNM) is challenging in endometrial cancer, and although several biomarkers are related to LNM, none of them are incorporated in clinical practice. The aim of this study was to develop and externally validate a preoperative BN to predict LNM and outcome in endometrial cancer patients. METHODS AND FINDINGS: Within the European Network for Individualized Treatment of Endometrial Cancer (ENITEC), we performed a retrospective multicenter cohort study including 763 patients, median age 65 years (interquartile range [IQR] 58-71), surgically treated for endometrial cancer between February 1995 and August 2013 at one of the 10 participating European hospitals. A BN was developed using score-based machine learning in addition to expert knowledge. Our main outcome measures were LNM and 5-year disease-specific survival (DSS). Preoperative clinical, histopathological, and molecular biomarkers were included in the network. External validation was performed using 2 prospective study cohorts: the Molecular Markers in Treatment in Endometrial Cancer (MoMaTEC) study cohort, including 446 Norwegian patients, median age 64 years (IQR 59-74), treated between May 2001 and 2010; and the PIpelle Prospective ENDOmetrial carcinoma (PIPENDO) study cohort, including 384 Dutch patients, median age 66 years (IQR 60-73), treated between September 2011 and December 2013. A BN called ENDORISK (preoperative risk stratification in endometrial cancer) was developed including the following predictors: preoperative tumor grade; immunohistochemical expression of estrogen receptor (ER), progesterone receptor (PR), p53, and L1 cell adhesion molecule (L1CAM); cancer antigen 125 serum level; thrombocyte count; imaging results on lymphadenopathy; and cervical cytology. In the MoMaTEC cohort, the area under the curve (AUC) was 0.82 (95% confidence interval [CI] 0.76-0.88) for LNM and 0.82 (95% CI 0.77-0.87) for 5-year DSS. In the PIPENDO cohort, the AUC for 5-year DSS was 0.84 (95% CI 0.78-0.90). The network was well-calibrated. In the MoMaTEC cohort, 249 patients (55.8%) were classified with <5% risk of LNM, with a false-negative rate of 1.6%. A limitation of the study is the use of imputation to correct for missing predictor variables in the development cohort and the retrospective study design. CONCLUSIONS: In this study, we illustrated how BNs can be used for individualizing clinical decision-making in oncology by incorporating easily accessible and multimodal biomarkers. The network shows the complex interactions underlying the carcinogenetic process of endometrial cancer by its graphical representation. A prospective feasibility study will be needed prior to implementation in the clinic.

A probabilistic framework for predicting disease dynamics: A case study of psychotic depression.

Unsupervised learning is often used to obtain insight into the underlying structure of medical data, but it is not always clear how to use such structure in an effective way. In this paper, we propose a probabilistic framework for predicting disease dynamics guided by latent states. The framework is based on hidden Markov models and aims to facilitate the selection of hypotheses that might yield insight into the dynamics. We demonstrate this by using clinical trial data for psychotic depression treatment as a case study. The discovered latent structure and proposed outcome are then validated using standard depression criteria, and are shown to provide new insight into the heterogeneity of psychotic depression in terms of predictive symptoms for different interventions.

A comparison between discrete and continuous time Bayesian networks in learning from clinical time series data with irregularity.

BACKGROUND: Recently, mobile devices, such as smartphones, have been introduced into healthcare research to substitute paper diaries as data-collection tools in the home environment. Such devices support collecting patient data at different time points over a long period, resulting in clinical time-series data with high temporal complexity, such as time irregularities. Analysis of such time series poses new challenges for machine-learning techniques. The clinical context for the research discussed in this paper is home monitoring in chronic obstructive pulmonary disease (COPD). OBJECTIVE: The goal of the present research is to find out which properties of temporal Bayesian network models allow to cope best with irregularly spaced multivariate clinical time-series data. METHODS: Two mainstream temporal Bayesian network models of multivariate clinical time series are studied: dynamic Bayesian networks, where the system is described as a snapshot at discrete time points, and continuous time Bayesian networks, where transitions between states are modeled in continuous time. Their capability of learning from clinical time series that vary in nature are extensively studied. In order to compare the two temporal Bayesian network types for regularly and irregularly spaced time-series data, three typical ways of observing time-series data were investigated: (1) regularly spaced in time with a fixed rate; (2) irregularly spaced and missing completely at random at discrete time points; (3) irregularly spaced and missing at random at discrete time points. In addition, similar experiments were carried out using real-world COPD patient data where observations are unevenly spaced. RESULTS: For regularly spaced time series, the dynamic Bayesian network models outperform the continuous time Bayesian networks. Similarly, if the data is missing completely at random, discrete-time models outperform continuous time models in most situations. For more realistic settings where data is not missing completely at random, the situation is more complicated. In simulation experiments, both models perform similarly if there is strong prior knowledge available about the missing data distribution. Otherwise, continuous time Bayesian networks perform better. In experiments with unevenly spaced real-world data, we surprisingly found that a dynamic Bayesian network where time is ignored performs similar to a continuous time Bayesian network. CONCLUSION: The results confirm conventional wisdom that discrete-time Bayesian networks are appropriate when learning from regularly spaced clinical time series. Similarly, we found that time series where the missingness occurs completely at random, dynamic Bayesian networks are an appropriate choice. However, for complex clinical time-series data that motivated this research, the continuous-time models are at least competitive and sometimes better than their discrete-time counterparts. Furthermore, continuous-time models provide additional benefits of being able to provide more fine-grained predictions than discrete-time models, which will be of practical relevance in clinical applications.

Exploiting Experts' Knowledge for Structure Learning of Bayesian Networks.

Learning Bayesian network structures from data is known to be hard, mainly because the number of candidate graphs is super-exponential in the number of variables. Furthermore, using observational data alone, the true causal graph is not discernible from other graphs that model the same set of conditional independencies. In this paper, it is investigated whether Bayesian network structure learning can be improved by exploiting the opinions of multiple domain experts regarding cause-effect relationships. In practice, experts have different individual probabilities of correctly labeling the inclusion or exclusion of edges in the structure. The accuracy of each expert is modeled by three parameters. Two new scoring functions are introduced that score each candidate graph based on the data and experts' opinions, taking into account their accuracy parameters. In the first scoring function, the experts' accuracies are estimated using an expectation-maximization-based algorithm and the estimated accuracies are explicitly used in the scoring process. The second function marginalizes out the accuracy parameters to obtain more robust scores when it is not possible to obtain a good estimate of experts' accuracies. The experimental results on simulated and real world datasets show that exploiting experts' knowledge can improve the structure learning if we take the experts' accuracies into account.

Understanding disease processes by partitioned dynamic Bayesian networks.

For many clinical problems in patients the underlying pathophysiological process changes in the course of time as a result of medical interventions. In model building for such problems, the typical scarcity of data in a clinical setting has been often compensated by utilizing time homogeneous models, such as dynamic Bayesian networks. As a consequence, the specificities of the underlying process are lost in the obtained models. In the current work, we propose the new concept of partitioned dynamic Bayesian networks to capture distribution regime changes, i.e. time non-homogeneity, benefiting from an intuitive and compact representation with the solid theoretical foundation of Bayesian network models. In order to balance specificity and simplicity in real-world scenarios, we propose a heuristic algorithm to search and learn these non-homogeneous models taking into account a preference for less complex models. An extensive set of experiments were ran, in which simulating experiments show that the heuristic algorithm was capable of constructing well-suited solutions, in terms of goodness of fit and statistical distance to the original distributions, in consonance with the underlying processes that generated data, whether it was homogeneous or non-homogeneous. Finally, a study case on psychotic depression was conducted using non-homogeneous models learned by the heuristic, leading to insightful answers for clinically relevant questions concerning the dynamics of this mental disorder.

Smartphone-based analysis of biochemical tests for health monitoring support at home.

In the context of home-based healthcare monitoring systems, it is desirable that the results obtained from biochemical tests - tests of various body fluids such as blood and urine - are objective and automatically generated to reduce the number of man-made errors. The authors present the StripTest reader - an innovative smartphone-based interpreter of biochemical tests based on paper-based strip colour using image processing techniques. The working principles of the reader include image acquisition of the colour strip pads using the camera phone, analysing the images within the phone and comparing them with reference colours provided by the manufacturer to obtain the test result. The detection of kidney damage was used as a scenario to illustrate the application of, and test, the StripTest reader. An extensive evaluation using laboratory and human urine samples demonstrates the reader's accuracy and precision of detection, indicating the successful development of a cheap, mobile and smart reader for home-monitoring of kidney functioning, which can facilitate the early detection of health problems and a timely treatment intervention.

Learning Bayesian networks for clinical time series analysis.

INTRODUCTION: Autonomous chronic disease management requires models that are able to interpret time series data from patients. However, construction of such models by means of machine learning requires the availability of costly health-care data, often resulting in small samples. We analysed data from chronic obstructive pulmonary disease (COPD) patients with the goal of constructing a model to predict the occurrence of exacerbation events, i.e., episodes of decreased pulmonary health status. METHODS: Data from 10 COPD patients, gathered with our home monitoring system, were used for temporal Bayesian network learning, combined with bootstrapping methods for data analysis of small data samples. For comparison a temporal variant of augmented naive Bayes models and a temporal nodes Bayesian network (TNBN) were constructed. The performances of the methods were first tested with synthetic data. Subsequently, different COPD models were compared to each other using an external validation data set. RESULTS: The model learning methods are capable of finding good predictive models for our COPD data. Model averaging over models based on bootstrap replications is able to find a good balance between true and false positive rates on predicting COPD exacerbation events. Temporal naive Bayes offers an alternative that trades some performance for a reduction in computation time and easier interpretation.

Describing disease processes using a probabilistic logic of qualitative time.

BACKGROUND: Clinical knowledge about progress of diseases is characterised by temporal information as well as uncertainty. However, precise timing information is often unavailable in medicine. In previous research this problem has been tackled using Allen's qualitative algebra of time, which, despite successful medical application, does not deal with the associated uncertainty. OBJECTIVES: It is investigated whether and how Allen's temporal algebra can be extended to handle uncertainty to better fit available knowledge and data of disease processes. METHODS: To bridge the gap between probability theory and qualitative time reasoning, methods from probabilistic logic are explored. The relation between the probabilistic logic representation and dynamic Bayesian networks is analysed. By studying a typical, and clinically relevant problem, the detection of exacerbations of chronic obstructive pulmonary disease (COPD), it is determined whether the developed probabilistic logic of qualitative time is medically useful. RESULTS: The probabilistic logic extension of Allen's temporal algebra, called Qualitative Time CP-logic provides a tool to model disease processes at a natural level of abstraction and is sufficiently powerful to reason with imprecise, uncertain knowledge. The representation of the COPD disease process gives evidence that the framework can be applied functionally to a clinical problem. CONCLUSION: The combination of qualitative time and probabilistic logic offers a useful framework for modelling knowledge and data to describe disease processes in clinical medicine.

Multilevel temporal Bayesian networks can model longitudinal change in multimorbidity.

OBJECTIVES: Although the course of single diseases can be studied using traditional epidemiologic techniques, these methods cannot capture the complex joint evolutionary course of multiple disorders. In this study, multilevel temporal Bayesian networks were adopted to study the course of multimorbidity in the expectation that this would yield new clinical insight. STUDY DESIGN AND SETTING: Clinical data of patients were extracted from 90 general practice registries in the Netherlands. One and half million patient-years were used for analysis. The simultaneous progression of six chronic cardiovascular conditions was investigated, correcting for both patient and practice-related variables. RESULTS: Cumulative incidence rates of one or more new morbidities rapidly increase with the number of morbidities present at baseline, ranging up to 47% and 76% for 3- and 5-year follow-ups, respectively. Hypertension and lipid disorders, as health risk factors, increase the cumulative incidence rates of both individual and multiple disorders. Moreover, in their presence, the observed cumulative incidence rates of combinations of cardiovascular disorders, that is, multimorbidity differs significantly from the expected rates. CONCLUSION: There are clear synergies between health risks and chronic diseases when multimorbidity within a patient progresses over time. The method used here supports a more comprehensive analysis of such synergies compared with what can be obtained by traditional statistics.

An autonomous mobile system for the management of COPD.

INTRODUCTION: Managing chronic disease through automated systems has the potential to both benefit the patient and reduce health-care costs. We have developed and evaluated a disease management system for patients with chronic obstructive pulmonary disease (COPD). Its aim is to predict and detect exacerbations and, through this, help patients self-manage their disease to prevent hospitalisation. MATERIALS: The carefully crafted intelligent system consists of a mobile device that is able to collect case-specific, subjective and objective, physiological data, and to alert the patient by a patient-specific interpretation of the data by means of probabilistic reasoning. Collected data are also sent to a central server for inspection by health-care professionals. METHODS: We evaluated the probabilistic model using cross-validation and ROC analyses on data from an earlier study and by an independent data set. Furthermore a pilot with actual COPD patients has been conducted to test technical feasibility and to obtain user feedback. RESULTS: Model evaluation results show that we can reliably detect exacerbations. Pilot study results suggest that an intervention based on this system could be successful.

On the interplay of machine learning and background knowledge in image interpretation by Bayesian networks.

OBJECTIVES: To obtain a balanced view on the role and place of expert knowledge and learning methods in building Bayesian networks for medical image interpretation. METHODS AND MATERIALS: The interpretation of mammograms was selected as the example medical image interpretation problem. Medical image interpretation has its own common standards and procedures. The impact of these on two complementary methods for Bayesian network construction was explored. Firstly, methods for the discretisation of continuous features were investigated, yielding multinomial distributions that were compared to the original Gaussian probabilistic parameters of the network. Secondly, the structure of a manually constructed Bayesian network was tested by structure learning from image data. The image data used for the research came from screening mammographic examinations of 795 patients, of whom 344 were cancerous. RESULTS: The experimental results show that there is an interesting interplay of machine learning results and background knowledge in medical image interpretation. Networks with discretised data lead to better classification performance (increase in the detected cancers of up to 11.7%), easier interpretation, and a better fit to the data in comparison to the expert-based Bayesian network with Gaussian probabilistic parameters. Gaussian probability distributions are often used in medical image interpretation because of the continuous nature of many of the image features. The structures learnt supported many of the expert-originated relationships but also revealed some novel relationships between the mammographic features. Using discretised features and performing structure learning on the mammographic data has further improved the cancer detection performance of up to 17% compared to the manually constructed Bayesian network model. CONCLUSION: Finding the right balance between expert knowledge and data-derived knowledge, both at the level of network structure and parameters, is key to using Bayesian networks for medical image interpretation. A balanced approach to building Bayesian networks for image interpretation yields more accurate and understandable Bayesian network models.

Multilevel Bayesian networks for the analysis of hierarchical health care data.

OBJECTIVE: Large health care datasets normally have a hierarchical structure, in terms of levels, as the data have been obtained from different practices, hospitals, or regions. Multilevel regression is the technique commonly used to deal with such multilevel data. However, for the statistical analysis of interactions between entities from a domain, multilevel regression yields little to no insight. While Bayesian networks have proved to be useful for analysis of interactions, they do not have the capability to deal with hierarchical data. In this paper, we describe a new formalism, which we call multilevel Bayesian networks; its effectiveness for the analysis of hierarchically structured health care data is studied from the perspective of multimorbidity. METHODS: Multilevel Bayesian networks are formally defined and applied to analyze clinical data from family practices in The Netherlands with the aim to predict interactions between heart failure and diabetes mellitus. We compare the results obtained with multilevel regression. RESULTS: The results obtained by multilevel Bayesian networks closely resembled those obtained by multilevel regression. For both diseases, the area under the curve of the prediction model improved, and the net reclassification improvements were significantly positive. In addition, the models offered considerable more insight, through its internal structure, into the interactions between the diseases. CONCLUSIONS: Multilevel Bayesian networks offer a suitable alternative to multilevel regression when analyzing hierarchical health care data. They provide more insight into the interactions between multiple diseases. Moreover, a multilevel Bayesian network model can be used for the prediction of the occurrence of multiple diseases, even when some of the predictors are unknown, which is typically the case in medicine.

A probabilistic framework for image information fusion with an application to mammographic analysis.

The recent increased interest in information fusion methods for solving complex problem, such as in image analysis, is motivated by the wish to better exploit the multitude of information, available from different sources, to enhance decision-making. In this paper, we propose a novel method, that advances the state of the art of fusing image information from different views, based on a special class of probabilistic graphical models, called causal independence models. The strength of this method is its ability to systematically and naturally capture uncertain domain knowledge, while performing information fusion in a computationally efficient way. We examine the value of the method for mammographic analysis and demonstrate its advantages in terms of explicit knowledge representation and accuracy (increase of at least 6.3% and 5.2% of true positive detection rates at 5% and 10% false positive rates) in comparison with previous single-view and multi-view systems, and benchmark fusion methods such as naïve Bayes and logistic regression.

Probabilistic causal models of multimorbidity concepts.

Multimorbidity, i.e., the presence of multiple diseases within one person, is a significant health-care problem for western societies: diagnosis, prognosis and treatment in the presence of of multiple diseases can be complex due to the various interactions between diseases. A literature review reveals that there is a variety of definitions that describe different concepts with respect to multimorbidity, both for the cause of multimorbidity as well as the implications of multimorbidity. To be able to aid computerized decision support systems within patient care, e.g. electronic clinical guidelines that can be personalized given the patient's problems, these multimorbidity aspects need to be defined rigorously in a formal language. In this paper, we employ causal Bayesian networks to define and analyze a novel framework that can be used to model a spectrum of aspects related to multimorbidity. We conclude that this framework provides a solid basis for modeling interactions between multiple diseases.