Discovery of clinical pathway patterns from event logs using probabilistic topic models

Temporal topic model for clinical pathway mining from electronic medical records

BACKGROUND

In recent years, the discovery of clinical pathways (CPs) from electronic medical records (EMRs) data has received increasing attention because it can directly support clinical doctors with explicit treatment knowledge, which is one of the key challenges in the development of intelligent healthcare services. However, the existing work has focused on topic probabilistic models, which usually produce treatment patterns with similar treatment activities, and such discovered treatment patterns do not take into account the temporal process of patient treatment which does not meet the needs of practical medical applications.

METHODS

Based on the assumption that CPs can be derived from the data of EMRs which usually record the treatment process of patients, this paper proposes a new CPs mining method from EMRs, an extended form of the traditional topic model - the temporal topic model (TTM). The method can capture the treatment topics and the corresponding treatment timestamps for each treatment day.

RESULTS

Experimental research conducted on a real-world dataset of patients' hospitalization processes, and the achieved results demonstrate the applicability and usefulness of the proposed methodology for CPs mining. Compared to existing benchmarks, our model shows significant improvement and robustness.

CONCLUSION

Our TTM provides a more competitive way to mine potential CPs considering the temporal features of the EMR data, providing a very prospective tool to support clinical diagnostic decisions.

Inferring actual treatment pathways from patient records

OBJECTIVE

Treatment pathways are step-by-step plans outlining the recommended medical care for specific diseases; they get revised when different treatments are found to improve patient outcomes. Examining health records is an important part of this revision process, but inferring patients' actual treatments from health data is challenging due to complex event-coding schemes and the absence of pathway-related annotations. The objective of this study is to develop a method for inferring actual treatment steps for a particular patient group from administrative health records - a common form of tabular healthcare data - and address several technique- and methodology-based gaps in treatment pathway-inference research.

METHODS

We introduce Defrag, a method for examining health records to infer the real-world treatment steps for a particular patient group. Defrag learns the semantic and temporal meaning of healthcare event sequences, allowing it to reliably infer treatment steps from complex healthcare data. To our knowledge, Defrag is the first pathway-inference method to utilise a neural network (NN), an approach made possible by a novel, self-supervised learning objective. We also developed a testing and validation framework for pathway inference, which we use to characterise and evaluate Defrag's pathway inference ability, establish benchmarks, and compare against baselines.

RESULTS

We demonstrate Defrag's effectiveness by identifying best-practice pathway fragments for breast cancer, lung cancer, and melanoma in public healthcare records. Additionally, we use synthetic data experiments to demonstrate the characteristics of the Defrag inference method, and to compare Defrag to several baselines, where it significantly outperforms non-NN-based methods.

CONCLUSIONS

Defrag offers an innovative and effective approach for inferring treatment pathways from complex health data. Defrag significantly outperforms several existing pathway-inference methods, but computationally-derived treatment pathways are still difficult to compare against clinical guidelines. Furthermore, the open-source code for Defrag and the testing framework are provided to encourage further research in this area.

Learning the progression patterns of treatments using a probabilistic generative model

Modeling a disease or the treatment of a patient has drawn much attention in recent years due to the vast amount of information that Electronic Health Records contain. This paper presents a probabilistic generative model of treatments that are described in terms of sequences of medical activities of variable length. The main objective is to identify distinct subtypes of treatments for a given disease, and discover their development and progression. To this end, the model considers that a sequence of actions has an associated hierarchical structure of latent variables that both classifies the sequences based on their evolution over time, and segments the sequences into different progression stages. The learning procedure of the model is performed with the Expectation-Maximization algorithm which considers the exponential number of configurations of the latent variables and is efficiently solved with a method based on dynamic programming. The evaluation of the model is twofold: first, we use synthetic data to demonstrate that the learning procedure allows the generative model underlying the data to be recovered; we then further assess the potential of our model to provide treatment classification and staging information in real-world data. Our model can be seen as a tool for classification, simulation, data augmentation and missing data imputation.

Exploring the potential of OMOP common data model for process mining in healthcare

BACKGROUND AND OBJECTIVE

Recently, Electronic Health Records (EHR) are increasingly being converted to Common Data Models (CDMs), a database schema designed to provide standardized vocabularies to facilitate collaborative observational research. To date, however, rare attempts exist to leverage CDM data for healthcare process mining, a technique to derive process-related knowledge (e.g., process model) from event logs. This paper presents a method to extract, construct, and analyze event logs from the Observational Medical Outcomes Partnership (OMOP) CDM for process mining and demonstrates CDM-based healthcare process mining with several real-life study cases while answering frequently posed questions in process mining, in the CDM environment.

METHODS

We propose a method to extract, construct, and analyze event logs from the OMOP CDM for process types including inpatient, outpatient, emergency room processes, and patient journey. Using the proposed method, we extract the retrospective data of several surgical procedure cases (i.e., Total Laparoscopic Hysterectomy (TLH), Total Hip Replacement (THR), Coronary Bypass (CB), Transcatheter Aortic Valve Implantation (TAVI), Pancreaticoduodenectomy (PD)) from the CDM of a Korean tertiary hospital. Patient data are extracted for each of the operations and analyzed using several process mining techniques.

RESULTS

Using process mining, the clinical pathways, outpatient process models, emergency room process models, and patient journeys are demonstrated using the extracted logs. The result shows CDM's usability as a novel and valuable data source for healthcare process analysis, yet with a few considerations. We found that CDM should be complemented by different internal and external data sources to address the administrative and operational aspects of healthcare processes, particularly for outpatient and ER process analyses.

CONCLUSION

To the best of our knowledge, we are the first to exploit CDM for healthcare process mining. Specifically, we provide a step-by-step guidance by demonstrating process analysis from locating relevant CDM tables to visualizing results using process mining tools. The proposed method can be widely applicable across different institutions. This work can contribute to bringing a process mining perspective to the existing CDM users in the changing Hospital Information Systems (HIS) environment and also to facilitating CDM-based studies in the process mining research community.

Applying process mining in health technology assessment

Objective

Propose a process mining-based method for Health Technology Assessment.

Methods

Articles dealing with prior studies in Health Technology Assessment using Process Mining were identified. Five research questions were defined to investigate these studies and present important points and desirable characteristics to be addressed in a proposal. The was defined method with five steps and was submitted to a case study for evaluation.

Results

The Literature search identified six main characteristics. As a result, the five-step method proposed was applied in the radical prostatectomy surgical procedure between the robot assisted technique and laparoscopy.

Conclusion

It was demonstrated in this article the creation of the proposal of an efficient method with its replication for other health technologies, coupled with the good interpretation of the specialists in terms of comprehensibility of the discovered patterns and their correlation with clinical protocols and guidelines.

Opportunities to use electronic health record audit logs to improve cancer care

The rapid adoption of electronic health records (EHRs) has created extensive repositories of digitized data that can be used to inform improvements in care delivery, processes, and patient outcomes. While the clinical data captured in EHRs are widely used for such efforts, EHRs also capture audit log data that reflect how users interact with the EHR to deliver care. Automatically collected audit log data provide a unique opportunity for new insights into EHR user behavior and decision‐making processes. Here, we provide an overview of audit log data and examples that could be used to improve oncology care and outcomes in four domains: diagnostic reasoning and consumption, care team collaboration and communication, patient outcomes and experience, and provider burnout/fatigue. This data source could identify gaps in performance and care, physician uptake of EHR features that enhance decision‐making, and integration of data trends for oncology. Ensuring researchers and oncologists are familiar with the data's potential and developing the data engineering capacity to utilize this rich data source, will expand the breadth of research to improve cancer care.

Clinical and operational insights from data-driven care pathway mapping: a systematic review

Background

Accumulated electronic data from a wide variety of clinical settings has been processed using a range of informatics methods to determine the sequence of care activities experienced by patients. The “as is” or “de facto” care pathways derived can be analysed together with other data to yield clinical and operational information. It seems likely that the needs of both health systems and patients will lead to increasing application of such analyses. A comprehensive review of the literature is presented, with a focus on the study context, types of analysis undertaken, and the utility of the information gained.

Methods

A systematic review was conducted of literature abstracting sequential patient care activities (“de facto” care pathways) from care records. Broad coverage was achieved by initial screening of a Scopus search term, followed by screening of citations (forward snowball) and references (backwards snowball). Previous reviews of related topics were also considered. Studies were initially classified according to the perspective captured in the derived pathways. Concept matrices were then derived, classifying studies according to additional data used and subsequent analysis undertaken, with regard for the clinical domain examined and the knowledge gleaned.

Results

254 publications were identified. The majority (n = 217) of these studies derived care pathways from data of an administrative/clinical type. 80% (n = 173) applied further analytical techniques, while 60% (n = 131) combined care pathways with enhancing data to gain insight into care processes.

Discussion

Classification of the objectives, analyses and complementary data used in data-driven care pathway mapping illustrates areas of greater and lesser focus in the literature. The increasing tendency for these methods to find practical application in service redesign is explored across the variety of contexts and research questions identified. A limitation of our approach is that the topic is broad, limiting discussion of methodological issues.

Conclusion

This review indicates that methods utilising data-driven determination of de facto patient care pathways can provide empirical information relevant to healthcare planning, management, and practice. It is clear that despite the number of publications found the topic reviewed is still in its infancy.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12911-022-01756-2.

Analyzing Patient Trajectories With Artificial Intelligence

In digital medicine, patient data typically record health events over time (eg, through electronic health records, wearables, or other sensing technologies) and thus form unique patient trajectories. Patient trajectories are highly predictive of the future course of diseases and therefore facilitate effective care. However, digital medicine often uses only limited patient data, consisting of health events from only a single or small number of time points while ignoring additional information encoded in patient trajectories. To analyze such rich longitudinal data, new artificial intelligence (AI) solutions are needed. In this paper, we provide an overview of the recent efforts to develop trajectory-aware AI solutions and provide suggestions for future directions. Specifically, we examine the implications for developing disease models from patient trajectories along the typical workflow in AI: problem definition, data processing, modeling, evaluation, and interpretation. We conclude with a discussion of how such AI solutions will allow the field to build robust models for personalized risk scoring, subtyping, and disease pathway discovery.

Towards Predictive Analysis on Disease Progression: A Variational Hawkes Process Model

Massively available longitudinal data about long-term disease trajectories of patients provides a golden mine for the understanding of disease progression and efficient health service delivery. It calls for quantitative modeling of disease progression, which is a tricky problem due to the complexity of the disease progression process as well as the irregularity of time documented in trajectories. In this study, we tackle the problem with the goal of predictively analyzing disease progression. Specifically, we propose a novel Variational Hawkes Process (VHP) model to generalize disease progression and predict future patient states based on the clinical observational data of past disease trajectories. First, Hawkes Process captures the intensity of irregular visits in a trajectory documented on medical facilities and controls the aforementioned information flowing into future visits. Thereafter, the captured intensity is incorporated into a Variational Auto-Encoder to generate the representation of the future partial disease trajectory for a target patient in a predictive manner. To further improve the prediction performance, we equip the proposed model with a disease trajectory discriminator to distinguish the generated trajectories from real ones. We evaluate the proposed model on two public datasets from the MIMIC-III database pertaining to heart failure and sepsis patients, respectively, and one real-world dataset from a Chinese hospital pertaining to heart failure patients with multiple admissions. Experimental results demonstrate that the proposed model significantly outperforms state-of-the-art baselines, and may derive a set of practical implications that can benefit a wide spectrum of management and applications on disease progression.

Modified Needleman-Wunsch algorithm for clinical pathway clustering

Clinical pathways are used to guide clinicians to provide a standardised delivery of care. Because of their standardisation, the aim of clinical pathways is to reduce variation in both care process and patient outcomes. When learning clinical pathways from data through data mining, it is common practice to represent each patient pathway as a string corresponding to their movements through activities. Clustering techniques are popular methods for pathway mining, and therefore this paper focuses on distance metrics applied to string data for k-medoids clustering. The two main aims are to firstly, develop a technique that seamlessly integrates expert information with data and secondly, to develop a string distance metric for the purpose of process data. The overall goal was to allow for more meaningful clustering results to be found by adding context into the string similarity calculation. Eight common distance metrics and their applicability are discussed. These distance metrics prove to give an arbitrary distance, without consideration for context, and each produce different results. As a result, this paper describes the development of a new distance metric, the modified Needleman-Wunsch algorithm, that allows for expert interaction with the calculation by assigning groupings and rankings to activities, which provide context to the strings. This algorithm has been developed in partnership with UK's National Health Service (NHS) with the focus on a lung cancer pathway, however the handling of the data and algorithm allows for application to any disease type. This method is contained within Sim.Pro.Flow, a publicly available decision support tool.

Real-World Utilization of Biomarker Testing for Patients with Advanced Non-Small Cell Lung Cancer in a Tertiary Referral Center and Referring Hospitals

The continued introduction of biomarkers and innovative testing methods makes already complex diagnosis in patients with stage IV non-small-cell lung cancer (NSCLC) even more complex. This study primarily analyzed variations in biomarker testing in clinical practice in patients referred to a comprehensive cancer center in the Netherlands. The secondary aim was to compare the cost of biomarker testing with the cost of whole-genome sequencing. The cohort included 102 stage IV NSCLC patients who received biomarker testing in 2017 or 2018 at the comprehensive cancer center. The complete biomarker testing history of the cohort was identified using linked data from the comprehensive cancer center and the nationwide network and registry of histopathology and cytopathology in the Netherlands. Unique biomarker-test combinations, costs, turnaround times, and test utilization were examined. The results indicate substantial variation in test utilization and sequences. The mean cost per patient of biomarker testing was 2259.92 ± 1217.10 USD, or 1881.23 ± 1013.15 EUR. Targeted gene panels were most frequently conducted, followed by IHC analysis for programmed cell death protein ligand 1. Typically, the most common biomarkers were assessed within the first tests, and emerging biomarkers were tested further down the test sequence. At the cost of current biomarker testing, replacing current testing with whole-genome sequencing would have led to cost-savings in only two patients (2%).

Identifying common treatments from Electronic Health Records with missing information. An application to breast cancer

The aim of this paper is to analyze the sequence of actions in the health system associated with a particular disease. In order to do that, using Electronic Health Records, we define a general methodology that allows us to: (i) identify the actions in the health system associated with a disease; (ii) identify those patients with a complete treatment for the disease; (iii) and discover common treatment pathways followed by the patients with a specific diagnosis. The methodology takes into account the characteristics of the EHRs, such as record heterogeneity and missing information. As an example, we use the proposed methodology to analyze breast cancer disease. For this diagnosis, 5 groups of treatments, which fit in with medical practice guidelines and expert knowledge, were obtained.

A Multiview Model for Detecting the Inappropriate Use of Prescription Medication: Machine Learning Approach

Background

The inappropriate use of prescription medication has recently garnered worldwide attention, but most national policies do not effectively provide for early detection or timely intervention.

Objective

This study aimed to develop and assess the validity of a model that can detect the inappropriate use of prescription medication. This effort combines a multiview and topic matching method. The study also assessed the validity of this approach.

Methods

A multiview extension of the latent Dirichlet allocation algorithm for topic modeling was chosen to generate diagnosis-medication topics, with data obtained from the Chinese Monitoring Network for Rational Use of Drugs (CMNRUD) database. Topic mapping allowed for calculating the degree to which diagnoses and medications were similarly distributed and, by setting a threshold, for identifying prescription misuse. The Beijing Regional Prescription Review Database (BRPRD) database was used as the gold standard to assess the model’s validity. We also conducted a sensitivity analysis using random samples of validated prescriptions and evaluated the model’s performance.

Results

A total of 44 million prescriptions were used to generate topics using the diagnoses and medications from the CMNRUD database. A random sample (15,000 prescriptions) from the BRPRD was used for validation, and it was found that the model had a sensitivity of 81.8%, specificity of 47.4%, positive-predictive value of 14.5%, and negative-predictive value of 96.0%. The model showed superior stability under different sampling proportions.

Conclusions

A method that combines multiview topic modeling and topic matching can detect the inappropriate use of prescription medication. This model, which has mediocre specificity and moderate sensitivity, can be used as a primary screening tool and will likely complement and improve the process of manually reviewing prescriptions.

Mining post-surgical care processes in breast cancer patients

In this work we describe the application of a careflow mining algorithm to detect the most frequent patterns of care in a cohort of 3000 breast cancer patients. The applied method relies on longitudinal data extracted from electronic health records, recorded from the first surgical procedure after a breast cancer diagnosis. Careflows are mined from events data recorded for administrative purposes, including procedures from ICD9 - CM billing codes and chemotherapy treatments. Events data have been pre-processed with Topic Modelling to create composite events based on concurrent procedures. The results of the careflow mining algorithm allow the discovery of electronic temporal phenotypes across the studied population. These phenotypes are further characterized on the basis of clinical traits and tumour histopathology, as well as in terms of relapses, metastasis occurrence and 5-year survival rates. Results are highly significant from a clinical perspective, since phenotypes describe well characterized pathology classes, and the careflows are well matched with existing clinical guidelines. The analysis thus facilitates deriving real-world evidence that can inform clinicians as well as hospital decision makers.

Quantification and visualisation methods of data-driven chronic care delivery pathways: protocol for a systematic review and content analysis

INTRODUCTION

Chronic conditions require long periods of care and often involve repeated interactions with multiple healthcare providers. Faced with increasing illness burden and costs, healthcare systems are currently working towards integrated care to streamline these interactions and improve efficiency. To support this, one promising resource is the information on routine care delivery stored in various electronic healthcare databases (EHD). In chronic conditions, care delivery pathways (CDPs) can be constructed by linking multiple data sources and extracting time-stamped healthcare utilisation events and other medical data related to individual or groups of patients over specific time periods; CDPs may provide insights into current practice and ways of improving it. Several methods have been proposed in recent years to quantify and visualise CDPs. We present the protocol for a systematic review aiming to describe the content and development of CDP methods, to derive common recommendations for CDP construction.

METHODS AND ANALYSIS

This protocol followed the Preferred Reporting Items for Systematic review and Meta-Analysis Protocols. A literature search will be performed in PubMed (MEDLINE), Scopus, IEEE, CINAHL and EMBASE, without date restrictions, to review published papers reporting data-driven chronic CDPs quantification and visualisation methods. We will describe them using several characteristics relevant for EHD use in long-term care, grouped into three domains: (1) clinical (what clinical information does the method use and how was it considered relevant?), (2) data science (what are the method's development and implementation characteristics?) and (3) behavioural (which behaviours and interactions does the method aim to promote among users and how?). Data extraction will be performed via deductive content analysis using previously defined characteristics and accompanied by an inductive analysis to identify and code additional relevant features. Results will be presented in descriptive format and used to compare current CDPs and generate recommendations for future CDP development initiatives.

ETHICS AND DISSEMINATION

Database searches will be initiated in May 2019. The review is expected to be completed by February 2020. Ethical approval is not required for this review. Results will be disseminated in peer-reviewed journals and conference presentations.

PROSPERO REGISTRATION NUMBER

CRD42019140494.

Variance Risk Identification and Treatment of Clinical Pathway by Integrated Bayesian Network and Association Rules Mining

With the continuous development of data mining techniques in the medical field, variance analysis in clinical pathways based on data mining approaches have attracted increasing attention from scholars and decision makers. However, studies on variance analysis and treatment of specific kinds of disease are still relatively scarce. In order to reduce the hazard of postpartum hemorrhage after cesarean section, we conducted a detailed analysis on the relevant risk factors and treatment mechanisms, adopting the integrated Bayesian network and association rule mining approaches. By proposing a Bayesian network model based on regression analysis, we calculated the probability of risk factors determining the key factors that result in postpartum hemorrhage after cesarean section. In addition, we mined a few association rules regarding the treatment of postpartum hemorrhage on the basis of different clinical features. We divided the risk factors into primary and secondary risk factors by realizing the classification of different causes of postpartum hemorrhage after cesarean section and sorted the posterior probability to obtain the key factors in the primary and secondary risk factors: uterine atony and prolonged labor. The rules of clinical features associated with the management of postpartum hemorrhage during cesarean section were obtained. Finally, related strategies were proposed for improving medical service quality and enhancing the rescue efficiency of clinical pathways in China.

Characterizing Artificial Intelligence Applications in Cancer Research: A Latent Dirichlet Allocation Analysis

Background

Artificial intelligence (AI)–based therapeutics, devices, and systems are vital innovations in cancer control; particularly, they allow for diagnosis, screening, precise estimation of survival, informing therapy selection, and scaling up treatment services in a timely manner.

Objective

The aim of this study was to analyze the global trends, patterns, and development of interdisciplinary landscapes in AI and cancer research.

Methods

An exploratory factor analysis was conducted to identify research domains emerging from abstract contents. The Jaccard similarity index was utilized to identify the most frequently co-occurring terms. Latent Dirichlet Allocation was used for classifying papers into corresponding topics.

Results

From 1991 to 2018, the number of studies examining the application of AI in cancer care has grown to 3555 papers covering therapeutics, capacities, and factors associated with outcomes. Topics with the highest volume of publications include (1) machine learning, (2) comparative effectiveness evaluation of AI-assisted medical therapies, and (3) AI-based prediction. Noticeably, this classification has revealed topics examining the incremental effectiveness of AI applications, the quality of life, and functioning of patients receiving these innovations. The growing research productivity and expansion of multidisciplinary approaches are largely driven by machine learning, artificial neural networks, and AI in various clinical practices.

Conclusions

The research landscapes show that the development of AI in cancer care is focused on not only improving prediction in cancer screening and AI-assisted therapeutics but also on improving other corresponding areas such as precision and personalized medicine and patient-reported outcomes.

Clinical Information Systems and Artificial Intelligence: Recent Research Trends

OBJECTIVES

This survey aims at reviewing the literature related to Clinical Information Systems (CIS), Hospital Information Systems (HIS), Electronic Health Record (EHR) systems, and how collected data can be analyzed by Artificial Intelligence (AI) techniques.

METHODS

We selected the major journals (11 journals) collecting papers (more than 7,000) over the last five years from the top members of the research community, and read and analyzed the papers (more than 200) covering the topics. Then, we completed the analysis using search engines to also include papers from major conferences over the same five years.

RESULTS

We defined a taxonomy of major features and research areas of CIS, HIS, EHR systems. We also defined a taxonomy for the use of Artificial Intelligence (AI) techniques on healthcare data. In the light of these taxonomies, we report on the most relevant papers from the literature.

CONCLUSIONS

We highlighted some major research directions and issues which seem to be promising and to need further investigations over a medium- or long-term period.

Can process mining automatically describe care pathways of patients with long-term conditions in UK primary care? A study protocol

INTRODUCTION

In the UK, primary care is seen as the optimal context for delivering care to an ageing population with a growing number of long-term conditions. However, if it is to meet these demands effectively and efficiently, a more precise understanding of existing care processes is required to ensure their configuration is based on robust evidence. This need to understand and optimise organisational performance is not unique to healthcare, and in industries such as telecommunications or finance, a methodology known as 'process mining' has become an established and successful method to identify how an organisation can best deploy resources to meet the needs of its clients and customers. Here and for the first time in the UK, we will apply it to primary care settings to gain a greater understanding of how patients with two of the most common chronic conditions are managed.

METHODS AND ANALYSIS

The study will be conducted in three phases; first, we will apply process mining algorithms to the data held on the clinical management system of four practices of varying characteristics in the West Midlands to determine how each interacts with patients with hypertension or type 2 diabetes. Second, we will use traditional process mapping exercises at each practice to manually produce maps of care processes for the selected condition. Third, with the aid of staff and patients at each practice, we will compare and contrast the process models produced by process mining with the process maps produced via manual techniques, review differences and similarities between them and the relative importance of each. The first pilot study will be on hypertension and the second for patients diagnosed with type 2 diabetes.

ETHICS AND DISSEMINATION

Ethical approval has been provided by East Midlands-Leicester South Regional Ethics Committee (REC reference 18/EM/0284). Having refined the automated production of maps of care processes, we can explore pinch points and bottlenecks, process variants and unexpected behaviour, and make informed recommendations to improve the quality and efficiency of care. The results of this study will be submitted for publication in peer-reviewed journals.

Probabilistic modeling personalized treatment pathways using electronic health records

BACKGROUND

Modeling personalized treatment pathways plays an important role in understanding essential/critical treatment behaviors performed on patients during their hospitalizations and thus provides the opportunity for the improvement of better health service delivery in treatment pathways.

OBJECTIVE

Unlike traditional business process mining, modeling personalized treatment pathways is more challenging because they are typically case-specific. Although several studies have been devoted to modeling patient treatment pathways, limited efforts have been made on the extraction of latent semantics and their transitions behind patient treatment pathways, which are often ambiguous and poorly understood.

METHODS

In this article, we propose an extension of the Hidden Markov Model to mine and model personalized treatment pathways by extracting latent treatment topics and identifying their sequential dependencies in pathways, in the form of probabilistic distributions and transitions of patients' raw Electronic Health Record (EHR) data.

RESULTS

We evaluated the proposed model on 48,024 patients with cardiovascular diseases. A total of 15 treatment topics and their typical transition routes were discovered from EHR data that contained 1,391,251 treatment events with 2786 types of interventions and that were evaluated by ten clinicians manually. The obtained p-values are 0.000146 and 0.009106 in comparison with both Latent Dirichlet Allocation and Sequent Naïve Bayes models, respectively; this outcome indicate that our approach achieves a better understanding of human evaluators on modeling personalized treatment pathway than that of benchmark models.

CONCLUSION

The experimental results on a real-world data set clearly suggest that the proposed model has efficiency in mining and modeling personalized treatment pathways. We argue that the discovered treatment topics and their transition routes, as actionable knowledge that represents the practice of treating individual patients in their clinical pathways, can be further exploited to help physicians better understand their specialty and learn from previous experiences for treatment analysis and improvement.

A data-driven framework of typical treatment process extraction and evaluation

BACKGROUND

A clinical pathway (CP) defines a standardized care process for a well-defined patient group that aims to improve patient outcomes and promote patient safety. However, the construction of a new pathway from scratch is a time-consuming task for medical staff because it involves many factors, including objects, multidisciplinary collaboration, sequential design, and outcome measurements. Recently, the rapid development of hospital information systems has allowed the storage of large volumes of electronic medical records (EMRs), and this information constitutes an abundant data resource for building CPs using data-mining methods.

METHODS

We provide an automatic method for extracting typical treatment processes from EMRs that consists of four key steps. First, a novel similarity method is proposed to measure the similarity of two treatment records. Then, we perform an affinity propagation (AP) clustering algorithm to cluster doctor order set sequences (DOSSs). Next, a framework is proposed to extract a high-level description of each treatment cluster. Finally, we evaluate the extracted typical treatment processes by matching the treatment cluster with external information, such as the treatment efficacy, length of stay, and treatment cost.

RESULTS

By experiments on EMRs of 8287 cerebral infarction patients, it is concluded that our proposed method can effectively extract typical treatment processes from treatment records, and also has great potential to improve treatment outcome by personalizing the treatment process for patients with different conditions.

CONCLUSION

The extracted typical treatment processes are intuitive and can provide managerial guidance for CP redesign and optimization. In addition, our work can assist clinicians in clearly understanding their routine treatment processes and recommend optimal treatment pathways for patients.

Simulation of patient flow in multiple healthcare units using process and data mining techniques for model identification

INTRODUCTION

An approach to building a hybrid simulation of patient flow is introduced with a combination of data-driven methods for automation of model identification. The approach is described with a conceptual framework and basic methods for combination of different techniques. The implementation of the proposed approach for simulation of the acute coronary syndrome (ACS) was developed and used in an experimental study.

METHODS

A combination of data, text, process mining techniques, and machine learning approaches for the analysis of electronic health records (EHRs) with discrete-event simulation (DES) and queueing theory for the simulation of patient flow was proposed. The performed analysis of EHRs for ACS patients enabled identification of several classes of clinical pathways (CPs) which were used to implement a more realistic simulation of the patient flow. The developed solution was implemented using Python libraries (SimPy, SciPy, and others).

RESULTS

The proposed approach enables more a realistic and detailed simulation of the patient flow within a group of related departments. An experimental study shows an improved simulation of patient length of stay for ACS patient flow obtained from EHRs in Almazov National Medical Research Centre in Saint Petersburg, Russia.

CONCLUSION

The proposed approach, methods, and solutions provide a conceptual, methodological, and programming framework for the implementation of a simulation of complex and diverse scenarios within a flow of patients for different purposes: decision making, training, management optimization, and others.

A two-step approach for mining patient treatment pathways in administrative healthcare databases

Clustering electronic medical records allows the discovery of information on healthcare practices. Entries in such medical records are usually composed of a succession of diagnostics or therapeutic steps. The corresponding processes are complex and heterogeneous since they depend on medical knowledge integrating clinical guidelines, the physician's individual experience, and patient data and conditions. To analyze such data, we are first proposing to cluster medical visits, consultations, and hospital stays into homogeneous groups, and then to construct higher-level patient treatment pathways over these different groups. These pathways are then also clustered to distill typical pathways, enabling interpretation of clusters by experts. This approach is evaluated on a real-world administrative database of elderly people in Québec suffering from heart failures.

Learning bundled care opportunities from electronic medical records

OBJECTIVE

The traditional fee-for-service approach to healthcare can lead to the management of a patient's conditions in a siloed manner, inducing various negative consequences. It has been recognized that a bundled approach to healthcare - one that manages a collection of health conditions together - may enable greater efficacy and cost savings. However, it is not always evident which sets of conditions should be managed in a bundled manner. In this study, we investigate if a data-driven approach can automatically learn potential bundles.

METHODS

We designed a framework to infer health condition collections (HCCs) based on the similarity of their clinical workflows, according to electronic medical record (EMR) utilization. We evaluated the framework with data from over 16,500 inpatient stays from Northwestern Memorial Hospital in Chicago, Illinois. The plausibility of the inferred HCCs for bundled care was assessed through an online survey of a panel of five experts, whose responses were analyzed via an analysis of variance (ANOVA) at a 95% confidence level. We further assessed the face validity of the HCCs using evidence in the published literature.

RESULTS

The framework inferred four HCCs, indicative of (1) fetal abnormalities, (2) late pregnancies, (3) prostate problems, and (4) chronic diseases, with congestive heart failure featuring prominently. Each HCC was substantiated with evidence in the literature and was deemed plausible for bundled care by the experts at a statistically significant level.

CONCLUSIONS

The findings suggest that an automated EMR data-driven framework conducted can provide a basis for discovering bundled care opportunities. Still, translating such findings into actual care management will require further refinement, implementation, and evaluation.

Discovering health-care processes using DeciClareMiner

Flexible, human-centric and knowledge-intensive processes occur in many service industries and are prominent in the health-care sector. Knowledge workers (e.g., doctors or other health-care personnel) are given the flexibility to address each process instance (i.e., episode of care) in the way that they deem most suitable. As a result, the knowledge of these processes is generally of a tacit nature, with many stakeholders lacking a clear view of a process. In this paper, we propose an algorithm called DeciClareMiner that combines process and decision mining to extract a process model and the corresponding knowledge from past executions of these processes. The algorithm was evaluated by applying it to a realistic health-care case and comparing the results to a complete search benchmark. In a relatively short time (10 min), DeciClareMiner was able to produce a DeciClare model that represents 93% of episodes of care with atomic constraints. Compared to the 50 h required to calculate the 100%-episode model via an exhaustive search approach, our result is considered a major improvement.

Computer technologies to integrate medical treatments to manage multimorbidity

The high prevalence of multimorbid cases is a challenge for Health-Care Systems today. Clinical practice guidelines are the means to register and transmit the available evidence-based medical knowledge concerning concrete diseases. Several computer languages have been defined to represent this knowledge in a way that computers could use to help physicians in the daily practice of medicine. The generation of guidelines for all possible multimorbidities entails several issues that are difficult to address. Consequently, numerous medical informatics technologies have appeared merging computer information structures in a way that the treatment knowledge about single diseases could be combined in order to deliver health-care to patients suffering from multimorbidity. This paper proposes a classification of the most promising current technologies addressing this issue and provides an analysis of their maturity, strengths, and weaknesses. We conclude with an enumeration of ten relevant issues to consider when developing such technologies.

Incorporating Topic Assignment Constraint and Topic Correlation Limitation into Clinical Goal Discovering for Clinical Pathway Mining

Clinical pathways are widely used around the world for providing quality medical treatment and controlling healthcare cost. However, the expert-designed clinical pathways can hardly deal with the variances among hospitals and patients. It calls for more dynamic and adaptive process, which is derived from various clinical data. Topic-based clinical pathway mining is an effective approach to discover a concise process model. Through this approach, the latent topics found by latent Dirichlet allocation (LDA) represent the clinical goals. And process mining methods are used to extract the temporal relations between these topics. However, the topic quality is usually not desirable due to the low performance of the LDA in clinical data. In this paper, we incorporate topic assignment constraint and topic correlation limitation into the LDA to enhance the ability of discovering high-quality topics. Two real-world datasets are used to evaluate the proposed method. The results show that the topics discovered by our method are with higher coherence, informativeness, and coverage than the original LDA. These quality topics are suitable to represent the clinical goals. Also, we illustrate that our method is effective in generating a comprehensive topic-based clinical pathway model.

Utilizing Electronic Medical Records to Discover Changing Trends of Medical Behaviors Over Time

Objectives

Medical behaviors are playing significant roles in the delivery of high quality and cost-effective health services. Timely discovery of changing frequencies of medical behaviors is beneficial for the improvement of health services. The main objective of this work is to discover the changing trends of medical behaviors over time.

Methods

This study proposes a two-steps approach to detect essential changing patterns of medical behaviors from Electronic Medical Records (EMRs). In detail, a probabilistic topic model, i.e., Latent Dirichlet allocation (LDA), is firstly applied to disclose yearly treatment patterns in regard to the risk stratification of patients from a large volume of EMRs. After that, the changing trends by comparing essential/critical medical behaviors in a specific time period are detected and analyzed, including changes of significant patient features with their values, and changes of critical treatment interventions with their occurring time stamps.

Results

We verify the effectiveness of the proposed approach on a clinical dataset containing 12,152 patient cases with a time range of 10 years. Totally, 135 patients features and 234 treatment interventions in three treatment patterns were selected to detect their changing trends. In particular, evolving trends of yearly occurring probabilities of the selected medical behaviors were categorized into six content changing patterns (i.e, 112 growing, 123 declining, 43 up-down, 16 down-up, 35 steady, and 40 jumping), using the proposed approach. Besides, changing trends of execution time of treatment interventions were classified into three occurring time changing patterns (i.e., 175 early-implemented, 50 steady-implemented and 9 delay-implemented).

Conclusions

Experimental results show that our approach has an ability to utilize EMRs to discover essential evolving trends of medical behaviors, and thus provide significant potential to be further explored for health services redesign and improvement.

Using telephony data to facilitate discovery of clinical workflows

BACKGROUND

Discovery of clinical workflows to target for redesign using methods such as Lean and Six Sigma is difficult. VoIP telephone call pattern analysis may complement direct observation and EMR-based tools in understanding clinical workflows at the enterprise level by allowing visualization of institutional telecommunications activity.

OBJECTIVE

To build an analytic framework mapping repetitive and high-volume telephone call patterns in a large medical center to their associated clinical units using an enterprise unified communications server log file and to support visualization of specific call patterns using graphical networks.

METHODS

Consecutive call detail records from the medical center's unified communications server were parsed to cross-correlate telephone call patterns and map associated phone numbers to a cost center dictionary. Hashed data structures were built to allow construction of edge and node files representing high volume call patterns for display with an open source graph network tool.

RESULTS

Summary statistics for an analysis of exactly one week's call detail records at a large academic medical center showed that 912,386 calls were placed with a total duration of 23,186 hours. Approximately half of all calling called number pairs had an average call duration under 60 seconds and of these the average call duration was 27 seconds.

CONCLUSIONS

Cross-correlation of phone calls identified by clinical cost center can be used to generate graphical displays of clinical enterprise communications. Many calls are short. The compact data transfers within short calls may serve as automation or re-design targets. The large absolute amount of time medical center employees were engaged in VoIP telecommunications suggests that analysis of telephone call patterns may offer additional insights into core clinical workflows.

Distinguishing surgical behavior by sequential pattern discovery

OBJECTIVE

Each surgical procedure is unique due to patient's and also surgeon's particularities. In this study, we propose a new approach to distinguish surgical behaviors between surgical sites, levels of expertise and individual surgeons thanks to a pattern discovery method.

METHODS

The developed approach aims to distinguish surgical behaviors based on shared longest frequent sequential patterns between surgical process models. To allow clustering, we propose a new metric called SLFSP. The approach is validated by comparison with a clustering method using Dynamic Time Warping as a metric to characterize the similarity between surgical process models.

RESULTS

Our method outperformed the existing approach. It was able to make a perfect distinction between surgical sites (accuracy of 100%). We reached an accuracy superior to 90% and 85% for distinguishing levels of expertise and individual surgeons.

CONCLUSION

Clustering based on shared longest frequent sequential patterns outperformed the previous study based on time analysis.

SIGNIFICANCE

The proposed method shows the feasibility of comparing surgical process models, not only by their duration but also by their structure of activities. Furthermore, patterns may show risky behaviors, which could be an interesting information for surgical training to prevent adverse events.

Temporal electronic phenotyping by mining careflows of breast cancer patients

In this work we present a careflow mining approach designed to analyze heterogeneous longitudinal data and to identify phenotypes in a patient cohort. The main idea underlying our approach is to combine methods derived from sequential pattern mining and temporal data mining to derive frequent healthcare histories (careflows) in a population of patients. This approach was applied to an integrated data repository containing clinical and administrative data of more than 4000 breast cancer patients. We used the mined histories to identify sub-cohorts of patients grouped according to healthcare activities pathways, then we characterized these sub-cohorts with clinical data. In this way, we were able to perform temporal electronic phenotyping of electronic health records (EHR) data.

An overview of topic modeling and its current applications in bioinformatics

BACKGROUND

With the rapid accumulation of biological datasets, machine learning methods designed to automate data analysis are urgently needed. In recent years, so-called topic models that originated from the field of natural language processing have been receiving much attention in bioinformatics because of their interpretability. Our aim was to review the application and development of topic models for bioinformatics.

DESCRIPTION

This paper starts with the description of a topic model, with a focus on the understanding of topic modeling. A general outline is provided on how to build an application in a topic model and how to develop a topic model. Meanwhile, the literature on application of topic models to biological data was searched and analyzed in depth. According to the types of models and the analogy between the concept of document-topic-word and a biological object (as well as the tasks of a topic model), we categorized the related studies and provided an outlook on the use of topic models for the development of bioinformatics applications.

CONCLUSION

Topic modeling is a useful method (in contrast to the traditional means of data reduction in bioinformatics) and enhances researchers' ability to interpret biological information. Nevertheless, due to the lack of topic models optimized for specific biological data, the studies on topic modeling in biological data still have a long and challenging road ahead. We believe that topic models are a promising method for various applications in bioinformatics research.

Knowledge Discovery from Biomedical Ontologies in Cross Domains

In recent years, there is an increasing demand for sharing and integration of medical data in biomedical research. In order to improve a health care system, it is required to support the integration of data by facilitating semantic interoperability systems and practices. Semantic interoperability is difficult to achieve in these systems as the conceptual models underlying datasets are not fully exploited. In this paper, we propose a semantic framework, called Medical Knowledge Discovery and Data Mining (MedKDD), that aims to build a topic hierarchy and serve the semantic interoperability between different ontologies. For the purpose, we fully focus on the discovery of semantic patterns about the association of relations in the heterogeneous information network representing different types of objects and relationships in multiple biological ontologies and the creation of a topic hierarchy through the analysis of the discovered patterns. These patterns are used to cluster heterogeneous information networks into a set of smaller topic graphs in a hierarchical manner and then to conduct cross domain knowledge discovery from the multiple biological ontologies. Thus, patterns made a greater contribution in the knowledge discovery across multiple ontologies. We have demonstrated the cross domain knowledge discovery in the MedKDD framework using a case study with 9 primary biological ontologies from Bio2RDF and compared it with the cross domain query processing approach, namely SLAP. We have confirmed the effectiveness of the MedKDD framework in knowledge discovery from multiple medical ontologies.

Predicate Oriented Pattern Analysis for Biomedical Knowledge Discovery

In the current biomedical data movement, numerous efforts have been made to convert and normalize a large number of traditional structured and unstructured data (e.g., EHRs, reports) to semi-structured data (e.g., RDF, OWL). With the increasing number of semi-structured data coming into the biomedical community, data integration and knowledge discovery from heterogeneous domains become important research problem. In the application level, detection of related concepts among medical ontologies is an important goal of life science research. It is more crucial to figure out how different concepts are related within a single ontology or across multiple ontologies by analysing predicates in different knowledge bases. However, the world today is one of information explosion, and it is extremely difficult for biomedical researchers to find existing or potential predicates to perform linking among cross domain concepts without any support from schema pattern analysis. Therefore, there is a need for a mechanism to do predicate oriented pattern analysis to partition heterogeneous ontologies into closer small topics and do query generation to discover cross domain knowledge from each topic. In this paper, we present such a model that predicates oriented pattern analysis based on their close relationship and generates a similarity matrix. Based on this similarity matrix, we apply an innovated unsupervised learning algorithm to partition large data sets into smaller and closer topics and generate meaningful queries to fully discover knowledge over a set of interlinked data sources. We have implemented a prototype system named BmQGen and evaluate the proposed model with colorectal surgical cohort from the Mayo Clinic.

On mining latent topics from healthcare chat logs

BACKGROUND

Public and internet-based social media such as online healthcare-oriented chat groups provide a convenient channel for patients and people concerned about health to communicate and share information with each other. The chat logs of an online healthcare-oriented chat group can potentially be used to extract latent topics, to encourage participation, and to recommend relevant healthcare information to users.

OBJECTIVE

This paper addresses the use of online healthcare chat logs to automatically discover both underlying topics and user interests.

METHOD

We present a new probabilistic model that exploits healthcare chat logs to find hidden topics and changes in these topics over time. The proposed model uses separate but associated hidden variables to explore both topics and individual interests such that it can provide useful insights to the participants of online healthcare chat groups about their interests in terms of weighted topics or vice versa.

RESULTS

We evaluate the proposed model on a real-world chat log by comparing its performance to benchmark topic models, i.e., latent Dirichlet allocation (LDA) and Author Topic Model (ATM), on the topic extraction task. The chat log is obtained from an online chat group of pregnant women, which consists of 233,452 chat word tokens contributed by 118 users. Both detected individual interests and underlying topics with their progressive information over time are demonstrated. The results show that the performance of the proposed model exceeds that of the benchmark models.

CONCLUSION

The experimental results illustrate that the proposed model is a promising method for extracting healthcare knowledge from social media data.

Modeling healthcare data using multiple-channel latent Dirichlet allocation

Information and communications technologies have enabled healthcare institutions to accumulate large amounts of healthcare data that include diagnoses, medications, and additional contextual information such as patient demographics. To gain a better understanding of big healthcare data and to develop better data-driven clinical decision support systems, we propose a novel multiple-channel latent Dirichlet allocation (MCLDA) approach for modeling diagnoses, medications, and contextual information in healthcare data. The proposed MCLDA model assumes that a latent health status group structure is responsible for the observed co-occurrences among diagnoses, medications, and contextual information. Using a real-world research testbed that includes one million healthcare insurance claim records, we investigate the utility of MCLDA. Our empirical evaluation results suggest that MCLDA is capable of capturing the comorbidity structures and linking them with the distribution of medications. Moreover, MCLDA is able to identify the pairing between diagnoses and medications in a record based on the assigned latent groups. MCLDA can also be employed to predict missing medications or diagnoses given partial records. Our evaluation results also show that, in most cases, MCLDA outperforms alternative methods such as logistic regressions and the k-nearest-neighbor (KNN) model for two prediction tasks, i.e., medication and diagnosis prediction. Thus, MCLDA represents a promising approach to modeling healthcare data for clinical decision support.

Process Mining Methodology for Health Process Tracking Using Real-Time Indoor Location Systems

The definition of efficient and accurate health processes in hospitals is crucial for ensuring an adequate quality of service. Knowing and improving the behavior of the surgical processes in a hospital can improve the number of patients that can be operated on using the same resources. However, the measure of this process is usually made in an obtrusive way, forcing nurses to get information and time data, affecting the proper process and generating inaccurate data due to human errors during the stressful journey of health staff in the operating theater. The use of indoor location systems can take time information about the process in an unobtrusive way, freeing nurses, allowing them to engage in purely welfare work. However, it is necessary to present these data in a understandable way for health professionals, who cannot deal with large amounts of historical localization log data. The use of process mining techniques can deal with this problem, offering an easily understandable view of the process. In this paper, we present a tool and a process mining-based methodology that, using indoor location systems, enables health staff not only to represent the process, but to know precise information about the deployment of the process in an unobtrusive and transparent way. We have successfully tested this tool in a real surgical area with 3613 patients during February, March and April of 2015.

A Method to Evaluate Critical Factors for Successful Implementation of Clinical Pathways

OBJECTIVE

Clinical pathways (CPs) have been viewed as a multidisciplinary tool to improve the quality and efficiency of evidence-based care. Despite widespread enthusiasm for CPs, research has shown that many CP initiatives are unsuccessful. To this end, this study provides a methodology to evaluate critical success factors (CSFs) that can aid healthcare organizations to achieve successful CP implementation.

DESIGN

This study presents a new approach to evaluate CP implementation CSFs, with the aims being: (1) to identify CSFs for implementation of CPs through a comprehensive literature review and interviews with collaborative experts; (2) to use a filed study data with a robust fuzzy DEMATEL (the decision making trial and evaluation laboratory) approach to visualize the structure of complicated causal relationships between CSFs and obtain the influence level of these factors.

PARTICIPANTS

The filed study data is provided by ten clinical experts of a Chinese hospital.

RESULTS

23 identified CSF factors which are initially identified through a review of the literature and interviews with collaborative experts. Then, a number of direct and indirect relationships are derived from the data such that different perceptions can be integrated into a compromised cause and effect model of CP implementation.

CONCLUSIONS

The results indicate that the proposed approach can systematically evaluate CSFs and realize the importance of each factor such that the most common causes of failure of CP implementation could be eliminated or avoided. Therefore, the tool proposed would help healthcare organizations to manage CP implementation in a more effective and proactive way.