Using recommendation to support adaptive clinical pathways

Developing a Data-Driven Approach in Order to Improve the Safety and Quality of Patient Care

Objective: To improve the safety and quality of patient care in hospitals by shaping clinical pathways throughout the patient journey.

Study Setting: A risk model designed for healthcare organizations in the context of the challenges arising from comorbidity and other treatment-related complexities.

Study Design: The core of the model is the patient and his intra-hospital journey, which is analyzed using a data-driven approach. The structure of a predictive model to support organizational and clinical decision-making activities is explained. Data relating to each step of the intra-hospital journey (from hospital admission to discharge) are extracted from clinical records.

Principal Findings: The proposed approach is feasible and can be used effectively to improve safety and quality. It enables the evaluation of clinical risks at each step of the patient journey.

Conclusion: Based on data from real cases, the model can record and calculate, over time, variables and behaviors that affect the safety and quality of healthcare organizations. This provides a greater understanding of healthcare processes and their complexity which can, in turn, advance research relating to clinical pathways and improve strategies adopted by organizations.

Treatment effect prediction with adversarial deep learning using electronic health records

BACKGROUND

Treatment effect prediction (TEP) plays an important role in disease management by ensuring that the expected clinical outcomes are obtained after performing specialized and sophisticated treatments on patients given their personalized clinical status. In recent years, the wide adoption of electronic health records (EHRs) has provided a comprehensive data source for intelligent clinical applications including the TEP investigated in this study.

METHOD

We examined the problem of using a large volume of heterogeneous EHR data to predict treatment effects and developed an adversarial deep treatment effect prediction model to address the problem. Our model employed two auto-encoders for learning the representative and discriminative features of both patient characteristics and treatments from EHR data. The discriminative power of the learned features was further enhanced by decoding the correlational information between the patient characteristics and subsequent treatments by means of a generated adversarial learning strategy. Thereafter, a logistic regression layer was appended on the top of the resulting feature representation layer for TEP.

RESULT

The proposed model was evaluated on two real clinical datasets collected from the cardiology department of a Chinese hospital. In particular, on acute coronary syndrome (ACS) dataset, the proposed adversarial deep treatment effect prediction (ADTEP) (0.662) exhibited 1.4, 2.2, and 6.3% performance gains in terms of the area under the ROC curve (AUC) over deep treatment effect prediction (DTEP) (0.653), logistic regression (LR) (0.648), and support vector machine (SVM) (0.621), respectively. As for heart failure (HF) case study, the proposed ADTEP also outperformed all benchmarks. The experimental results demonstrated that our proposed model achieved competitive performance compared to state-of-the-art models in tackling the TEP problem.

CONCLUSION

In this work, we propose a novel model to address the TEP problem by utilizing a large volume of observational data from EHR. With adversarial learning strategy, our proposed model can further explore the correlational information between patient statuses and treatments to extract more robust and discriminative representation of patient samples from their EHR data. Such representation finally benefits the model on TEP. The experimental results of two case studies demonstrate the superiority of our proposed method compared to state-of-the-art methods.

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.

Characterizing treatment pathways at scale using the OHDSI network

Observational research promises to complement experimental research by providing large, diverse populations that would be infeasible for an experiment. Observational research can test its own clinical hypotheses, and observational studies also can contribute to the design of experiments and inform the generalizability of experimental research. Understanding the diversity of populations and the variance in care is one component. In this study, the Observational Health Data Sciences and Informatics (OHDSI) collaboration created an international data network with 11 data sources from four countries, including electronic health records and administrative claims data on 250 million patients. All data were mapped to common data standards, patient privacy was maintained by using a distributed model, and results were aggregated centrally. Treatment pathways were elucidated for type 2 diabetes mellitus, hypertension, and depression. The pathways revealed that the world is moving toward more consistent therapy over time across diseases and across locations, but significant heterogeneity remains among sources, pointing to challenges in generalizing clinical trial results. Diabetes favored a single first-line medication, metformin, to a much greater extent than hypertension or depression. About 10% of diabetes and depression patients and almost 25% of hypertension patients followed a treatment pathway that was unique within the cohort. Aside from factors such as sample size and underlying population (academic medical center versus general population), electronic health records data and administrative claims data revealed similar results. Large-scale international observational research is feasible.

Integration of Personalized Healthcare Pathways in an ICT Platform for Diabetes Managements: A Small-Scale Exploratory Study

The availability of new tools able to support patient monitoring and personalized care may substantially improve the quality of chronic disease management. A personalized healthcare pathway (PHP) has been developed for diabetes disease management and integrated into an information and communication technology system to accomplish a shift from organization-centered care to patient-centered care. A small-scale exploratory study was conducted to test the platform. Preliminary results are presented that shed light on how the PHP influences system usage and performance outcomes.

Developing a semantic web model for medical differential diagnosis recommendation

In this paper we describe a novel model for differential diagnosis designed to make recommendations by utilizing semantic web technologies. The model is a response to a number of requirements, ranging from incorporating essential clinical diagnostic semantics to the integration of data mining for the process of identifying candidate diseases that best explain a set of clinical features. We introduce two major components, which we find essential to the construction of an integral differential diagnosis recommendation model: the evidence-based recommender component and the proximity-based recommender component. Both approaches are driven by disease diagnosis ontologies designed specifically to enable the process of generating diagnostic recommendations. These ontologies are the disease symptom ontology and the patient ontology. The evidence-based diagnosis process develops dynamic rules based on standardized clinical pathways. The proximity-based component employs data mining to provide clinicians with diagnosis predictions, as well as generates new diagnosis rules from provided training datasets. This article describes the integration between these two components along with the developed diagnosis ontologies to form a novel medical differential diagnosis recommendation model. This article also provides test cases from the implementation of the overall model, which shows quite promising diagnostic recommendation results.

Online treatment compliance checking for clinical pathways

Compliance checking for clinical pathways (CPs) is getting increasing attention in health-care organizations due to stricter requirements for cost control and treatment excellence. Many compliance measures have been proposed for treatment behavior inspection in CPs. However, most of them look at aggregated data seen from an external perspective, e.g. length of stay, cost, infection rate, etc., which may provide only a posterior impression of the overall conformance with the established CPs such that in-depth and in near real time checking on the compliance of the essential/critical treatment behaviors of CPs is limited. To provide clinicians real time insights into violations of the established CP specification and support online compliance checking, this article presents a semantic rule-based CP compliance checking system. In detail, we construct a CP ontology (CPO) model to provide a formal grounding of CP compliance checking. Using the proposed CPO, domain treatment constraints are modeled into Semantic Web Rule Language (SWRL) rules to specify the underlying treatment behaviors and their quantified temporal structure in a CP. The established SWRL rules are integrated with the CP workflow such that a series of applicable compliance checking and evaluation can be reminded and recommended during the pathway execution. The proposed approach can, therefore, provides a comprehensive compliance checking service as a paralleling activity to the patient treatment journey of a CP rather than an afterthought. The proposed approach is illustrated with a case study on the unstable angina clinical pathway implemented in the Cardiology Department of a Chinese hospital. The results demonstrate that the approach, as a feasible solution to provide near real time conformance checking of CPs, not only enables clinicians to uncover non-compliant treatment behaviors, but also empowers clinicians with the capability to make informed decisions when dealing with treatment compliance violations in the pathway execution.

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

Discovery of clinical pathway (CP) patterns has experienced increased attention over the years due to its importance for revealing the structure, semantics and dynamics of CPs, and to its usefulness for providing clinicians with explicit knowledge which can be directly used to guide treatment activities of individual patients. Generally, discovery of CP patterns is a challenging task as treatment behaviors in CPs often have a large variability depending on factors such as time, location and patient individual. Based on the assumption that CP patterns can be derived from clinical event logs which usually record various treatment activities in CP executions, this study proposes a novel approach to CP pattern discovery by modeling CPs using mixtures of an extension to the Latent Dirichlet Allocation family that jointly models various treatment activities and their occurring time stamps in CPs. Clinical case studies are performed to evaluate the proposed approach via real-world data sets recording typical treatment behaviors in patient careflow. The obtained results demonstrate the suitability of the proposed approach for CP pattern discovery, and indicate the promise in research efforts related to CP analysis and optimization.

Anomaly detection in clinical processes

Meaningful anomalies in clinical processes may be related to caring performance or even the patient survival. It is imperative that the anomalies be timely detected such that useful and actionable knowledge of interest could be extracted to clinicians. Many previous approaches assume prior knowledge about the structure of clinical processes, using which anomalies are detected in a supervised manner. For a majority of clinical settings, however, clinical processes are complex, ad hoc, and even unknown a prior. In this paper, we investigate how to facilitate detection of anomalies in an unsupervised manner. An anomaly detection model is presented by applying a density-based clustering method on patient careflow logs. Using the learned model, it is possible to detect whether a particular patient careflow trace is anomalous with respect to normal traces in the logs. The approach has been validated over real data sets collected from a Chinese hospital.

Summarizing clinical pathways from event logs

OBJECTIVE

Clinical pathway analysis, as a pivotal issue in ensuring specialized, standardized, normalized and sophisticated therapy procedures, is receiving increasing attention in the field of medical informatics. Research in clinical pathway analysis has so far mostly focused on looking at aggregated data seen from an external perspective, and only provide very limited insight into the pathways. In some recent work, process mining techniques have been studied in discovering clinical pathway models from data. While it is interesting, discovered models may provide too much detail to give a comprehensive summary of the pathway. Moreover, the number of patterns discovered can be large. Alternatively, this article presents a new approach to build a concise and comprehensive summary that describes the entire structure of a clinical pathway, while revealing essential/critical medical behaviors in specific time intervals over the whole time period of the pathway.

METHODS

The presented approach summarizes a clinical pathway from the collected clinical event log, which regularly records all kinds of patient therapy and treatment activities in clinical workflow by various hospital information systems. The proposed approach formally defines the clinical pathway summarization problem as an optimization problem that can be solved in polynomial time by using a dynamic-programming algorithm. More specifically, given an input event log, the presented approach summarizes the pathway by segmenting the observed time period of the pathway into continuous and overlapping time intervals, and discovering frequent medical behavior patterns in each specific time interval from the log.

RESULTS

The proposed approach is evaluated via real-world data-sets, which are extracted from Zhejiang Huzhou Central hospital of China with regard to four specific diseases, i.e., bronchial lung cancer, colon cancer, gastric cancer, and cerebral infarction, in two years (2007.08-2009.09). Although the medical behaviors contained in these logs are very diverse and heterogeneous, experimental results indicates that the presented approach is feasible to construct condensed clinical pathway summaries in polynomial time from the collected logs, and have a linear scalability against the increasing size of the logs.

CONCLUSION

Experiments on real data-sets illustrate that the presented approach is efficient and discovers high-quality results: the observed time period of a clinical pathway is correctly segmented into a set of continuous and overlapping time intervals, in which essential/critical medical behaviors are well discovered from the event log to form the backbone of a clinical pathway. The experimental results indicate that the generated clinical pathway summary not only reveals the global structure of a pathway, but also provides a thorough understanding of the way in which actual medical behaviors are practiced in specific time intervals, which might be essential from the perspectives of clinical pathway analysis and improvement.