Analyzing complex patients' temporal histories: new frontiers in temporal data mining

Workflow Differences Affect Data Accuracy in Oncologic EHRs: A First Step Toward Detangling the Diagnosis Data Babel

PURPOSE

Diagnosis (DX) information is key to clinical data reuse, yet accessible structured DX data often lack accuracy. Previous research hints at workflow differences in cancer DX entry, but their link to clinical data quality is unclear. We hypothesized that there is a statistically significant relationship between workflow-describing variables and DX data quality.

METHODS

We extracted DX data from encounter and order tables within our electronic health records (EHRs) for a cohort of patients with confirmed brain neoplasms. We built and optimized logistic regressions to predict the odds of fully accurate (ie, correct neoplasm type and anatomic site), inaccurate, and suboptimal (ie, vague) DX entry across clinical workflows. We selected our variables based on correlation strength of each outcome variable.

RESULTS

Both workflow and personnel variables were predictive of DX data quality. For example, a DX entered in departments other than oncology had up to 2.89 times higher odds of being accurate (P < .0001) compared with an oncology department; an outpatient care location had up to 98% fewer odds of being inaccurate (P < .0001), but had 458 times higher odds of being suboptimal (P < .0001) compared with main campus, including the cancer center; and a DX recoded by a physician assistant had 85% fewer odds of being suboptimal (P = .005) compared with those entered by physicians.

CONCLUSION

These results suggest that differences across clinical workflows and the clinical personnel producing EHR data affect clinical data quality. They also suggest that the need for specific structured DX data recording varies across clinical workflows and may be dependent on clinical information needs. Clinicians and researchers reusing oncologic data should consider such heterogeneity when conducting secondary analyses of EHR data.

A method for the graphical modeling of relative temporal constraints

Searching for patient cohorts in electronic patient data often requires the definition of temporal constraints between the selection criteria. However, beyond a certain degree of temporal complexity, the non-graphical, form-based approaches implemented in current translational research platforms may be limited when modeling such constraints. In our opinion, there is a need for an easily accessible and implementable, fully graphical method for creating temporal queries. We aim to respond to this challenge with a new graphical notation. Based on Allen's time interval algebra, it allows for modeling temporal queries by arranging simple horizontal bars depicting symbolic time intervals. To make our approach applicable to complex temporal patterns, we apply two extensions: with duration intervals, we enable the inference about relative temporal distances between patient events, and with time interval modifiers, we support counting and excluding patient events, as well as constraining numeric values. We describe how to generate database queries from this notation. We provide a prototypical implementation, consisting of a temporal query modeling frontend and an experimental backend that connects to an i2b2 system. We evaluate our modeling approach on the MIMIC-III database to demonstrate that it can be used for modeling typical temporal phenotyping queries.

A tale of three subspecialties: Diagnosis recording patterns are internally consistent but Specialty-Dependent

BACKGROUND

Structured diagnosis (DX) are crucial for secondary use of electronic health record (EHR) data. However, they are often suboptimally recorded. Our previous work showed initial evidence of variable DX recording patterns in oncology charts even after biopsy records are available.

OBJECTIVE

We verified this finding's internal and external validity. We hypothesized that this recording pattern would be preserved in a larger cohort of patients for the same disease. We also hypothesized that this effect would vary across subspecialties.

METHODS

We extracted DX data from EHRs of patients treated for brain, lung, and pancreatic neoplasms, identified through clinician-led chart reviews. We used statistical methods (i.e., binomial and mixed model regressions) to test our hypotheses.

RESULTS

We found variable recording patterns in brain neoplasm DX (i.e., larger number of distinct DX-OR = 2.2, P < 0.0001, higher descriptive specificity scores-OR = 1.4, P < 0.0001-and much higher entropy after the BX-OR = 3.8 P = 0.004 and OR = 8.0, P < 0.0001), confirming our initial findings. We also found strikingly different patterns for lung and pancreas DX. Although both seemed to have much lower DX sequence entropy after the BX-OR = 0.198, P = 0.015 and OR = 0.099, P = 0.015, respectively compared to OR = 3.8 P = 0.004). We also found statistically significant differences between the brain dataset and both the lung (P < 0.0001) and pancreas (0.009

CONCLUSION

Our results suggest that disease-specific DX entry patterns exist and are established differently by clinical subspecialty. These differences should be accounted for during clinical data reuse and data quality assessments but also during EHR entry system design to maximize accurate, precise and consistent data entry likelihood.

Collapse

Key Words

• clinical data management
• data quality
• electronic health records
• learning healthcare system
• secondary use of clinical data

Collapse

MESH Headings Collapse

Grants

• K12 GM102773 NIGMS NIH HHS
• UL1 TR001420 NCATS NIH HHS

Collapse

Affiliation(s)

Jose-Franck Diaz-Garelli Wake Forest Baptist Medical Center, Winston Salem, North Carolina, USA
Roy Strowd Wake Forest Baptist Medical Center, Winston Salem, North Carolina, USA
Tamjeed Ahmed Wake Forest Baptist Medical Center, Winston Salem, North Carolina, USA
Brian J Wells Wake Forest Baptist Medical Center, Winston Salem, North Carolina, USA
Rebecca Merrill Wake Forest Baptist Medical Center, Winston Salem, North Carolina, USA
Javier Laurini Wake Forest Baptist Medical Center, Winston Salem, North Carolina, USA
Boris Pasche Wake Forest Baptist Medical Center, Winston Salem, North Carolina, USA
Umit Topaloglu Wake Forest Baptist Medical Center, Winston Salem, North Carolina, USA

Collapse

Predicting ICU readmission using grouped physiological and medication trends

BACKGROUND

Patients who are readmitted to an intensive care unit (ICU) usually have a high risk of mortality and an increased length of stay. ICU readmission risk prediction may help physicians to re-evaluate the patient's physical conditions before patients are discharged and avoid preventable readmissions. ICU readmission prediction models are often built based on physiological variables. Intuitively, snapshot measurements, especially the last measurements, are effective predictors that are widely used by researchers. However, methods that only use snapshot measurements neglect predictive information contained in the trends of physiological and medication variables. Mean, maximum or minimum values take multiple time points into account and capture their summary statistics, however, these statistics are not able to catch the detailed picture of temporal trends. In this study, we find strong predictors with ability of capturing detailed temporal trends of variables for 30-day readmission risk and build prediction models with high accuracy.

METHODS

We study physiological measurements and medications from the Multiparameter Intelligent Monitoring in Intensive Care II (MIMIC-II) clinical dataset. Time series of each variable are converted into trend graphs with nodes being discretized measurements of each variable. Then we extract important temporal trends by applying frequent subgraph mining on the trend graphs. The frequency of a subgraph is a good cue to find important temporal trends since similar patients often share similar trends regarding their pathophysiological evolution under medical interventions. Important temporal trends are then grouped automatically by non-negative matrix factorization. The grouped trends could be considered as an approximate representation of patients' pathophysiological states and medication profiles. We train a logistic regression model to predict 30-day ICU readmission risk based on snapshot measurements, grouped physiological trends and medication trends.

RESULTS

Our dataset consists of 1170 patients who are alive 30 days after discharge from ICU and have at least 12 h of data. In the dataset, 860 patients were not readmitted and 310 were readmitted, within 30 days after discharge. Our model outperforms all comparison models, and shows an improvement in the area under the receiver operating characteristic curve (AUC) of almost 4% from the best comparison model.

CONCLUSIONS

Grouped physiological and medication trends carry predictive information for ICU readmission risk. In order to build predictive models with higher accuracy, we should add grouped physiological and medication trends as complementary features to snapshot measurements.

Predicting delayed cerebral ischemia after subarachnoid hemorrhage using physiological time series data

To develop and validate a prediction model for delayed cerebral ischemia (DCI) after subarachnoid hemorrhage (SAH) using a temporal unsupervised feature engineering approach, demonstrating improved precision over standard features. 488 consecutive SAH admissions from 2006 to 2014 to a tertiary care hospital were included. Models were trained on 80%, while 20% were set aside for validation testing. Baseline information and standard grading scales were evaluated: age, sex, Hunt Hess grade, modified Fisher Scale (mFS), and Glasgow Coma Scale (GCS). An unsupervised approach applying random kernels was used to extract features from physiological time series (systolic and diastolic blood pressure, heart rate, respiratory rate, and oxygen saturation). Classifiers (Partial Least Squares, linear and kernel Support Vector Machines) were trained on feature subsets of the derivation dataset. Models were applied to the validation dataset. The performances of the best classifiers on the validation dataset are reported by feature subset. Standard grading scale (mFS): AUC 0.58. Combined demographics and grading scales: AUC 0.60. Random kernel derived physiologic features: AUC 0.74. Combined baseline and physiologic features with redundant feature reduction: AUC 0.77. Current DCI prediction tools rely on admission imaging and are advantageously simple to employ. However, using an agnostic and computationally inexpensive learning approach for high-frequency physiologic time series data, we demonstrated that our models achieve higher classification accuracy.

Incorporating High-Frequency Physiologic Data Using Computational Dictionary Learning Improves Prediction of Delayed Cerebral Ischemia Compared to Existing Methods

Purpose

Accurate prediction of delayed cerebral ischemia (DCI) after subarachnoid hemorrhage (SAH) can be critical for planning interventions to prevent poor neurological outcome. This paper presents a model using convolution dictionary learning to extract features from physiological data available from bedside monitors. We develop and validate a prediction model for DCI after SAH, demonstrating improved precision over standard methods alone.

Methods

488 consecutive SAH admissions from 2006 to 2014 to a tertiary care hospital were included. Models were trained on 80%, while 20% were set aside for validation testing. Modified Fisher Scale was considered the standard grading scale in clinical use; baseline features also analyzed included age, sex, Hunt-Hess, and Glasgow Coma Scales. An unsupervised approach using convolution dictionary learning was used to extract features from physiological time series (systolic blood pressure and diastolic blood pressure, heart rate, respiratory rate, and oxygen saturation). Classifiers (partial least squares and linear and kernel support vector machines) were trained on feature subsets of the derivation dataset. Models were applied to the validation dataset.

Results

The performances of the best classifiers on the validation dataset are reported by feature subset. Standard grading scale (mFS): AUC 0.54. Combined demographics and grading scales (baseline features): AUC 0.63. Kernel derived physiologic features: AUC 0.66. Combined baseline and physiologic features with redundant feature reduction: AUC 0.71 on derivation dataset and 0.78 on validation dataset.

Conclusion

Current DCI prediction tools rely on admission imaging and are advantageously simple to employ. However, using an agnostic and computationally inexpensive learning approach for high-frequency physiologic time series data, we demonstrated that we could incorporate individual physiologic data to achieve higher classification accuracy.

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.

Systems biology approaches for identifying adverse drug reactions and elucidating their underlying biological mechanisms

Small molecules are indispensable to modern medical therapy. However, their use may lead to unintended, negative medical outcomes commonly referred to as adverse drug reactions (ADRs). These effects vary widely in mechanism, severity, and populations affected, making ADR prediction and identification important public health concerns. Current methods rely on clinical trials and postmarket surveillance programs to find novel ADRs; however, clinical trials are limited by small sample size, whereas postmarket surveillance methods may be biased and inherently leave patients at risk until sufficient clinical evidence has been gathered. Systems pharmacology, an emerging interdisciplinary field combining network and chemical biology, provides important tools to uncover and understand ADRs and may mitigate the drawbacks of traditional methods. In particular, network analysis allows researchers to integrate heterogeneous data sources and quantify the interactions between biological and chemical entities. Recent work in this area has combined chemical, biological, and large-scale observational health data to predict ADRs in both individual patients and global populations. In this review, we explore the rapid expansion of systems pharmacology in the study of ADRs. We enumerate the existing methods and strategies and illustrate progress in the field with a model framework that incorporates crucial data elements, such as diet and comorbidities, known to modulate ADR risk. Using this framework, we highlight avenues of research that may currently be underexplored, representing opportunities for future work.

Big Data Technologies: New Opportunities for Diabetes Management

The so-called big data revolution provides substantial opportunities to diabetes management. At least 3 important directions are currently of great interest. First, the integration of different sources of information, from primary and secondary care to administrative information, may allow depicting a novel view of patient's care processes and of single patient's behaviors, taking into account the multifaceted nature of chronic care. Second, the availability of novel diabetes technologies, able to gather large amounts of real-time data, requires the implementation of distributed platforms for data analysis and decision support. Finally, the inclusion of geographical and environmental information into such complex IT systems may further increase the capability of interpreting the data gathered and extract new knowledge from them. This article reviews the main concepts and definitions related to big data, it presents some efforts in health care, and discusses the potential role of big data in diabetes care. Finally, as an example, it describes the research efforts carried on in the MOSAIC project, funded by the European Commission.