Querying a Clinical Data Warehouse for Combinations of Clinical and Imaging Data.

This study aims to show the feasibility and benefit of single queries in a research data warehouse combining data from a hospital's clinical and imaging systems. We used a comprehensive integration of a production picture archiving and communication system (PACS) with a clinical data warehouse (CDW) for research to create a system that allows data from both domains to be queried jointly with a single query. To achieve this, we mapped the DICOM information model to the extended entity-attribute-value (EAV) data model of a CDW, which allows data linkage and query constraints on multiple levels: the patient, the encounter, a document, and a group level. Accordingly, we have integrated DICOM metadata directly into CDW and linked it to existing clinical data. We included data collected in 2016 and 2017 from the Department of Internal Medicine in this analysis for two query inquiries from researchers targeting research about a disease and in radiology. We obtained quantitative information about the current availability of combinations of clinical and imaging data using a single multilevel query compiled for each query inquiry. We compared these multilevel query results to results that linked data at a single level, resulting in a quantitative representation of results that was up to 112% and 573% higher. An EAV data model can be extended to store data from clinical systems and PACS on multiple levels to enable combined querying with a single query to quickly display actual frequency data.

Improved Mounting of Strain Sensors by Reactive Bonding.

Aim of this work is to improve the bond between a strain sensor and a device on which the strain shall be determined. As strain sensor, a CMOS-integrated chip featuring piezoresistive sensor elements was used which is capable of wireless energy and data transmission. The sensor chip was mounted on a standardized tensile test specimen of stainless steel by a bonding process using reactive multilayer systems (RMS). RMS provide a well-defined amount of heat within a very short reaction time of a few milliseconds and are placed in-between two bonding partners. RMS were combined with layers of solder which melt during the bonding process. Epoxy adhesive films were used as a reference bonding process. Under mechanical tensile loading, the sensor bonded with RMS shows a linear strain sensitivity in the whole range of tested forces whereas the adhesive-bonded sensor has slightly nonlinear behavior for low forces. Compared to the adhesive-bonded chips, the sensitivity of the reactively bonded chips is increased by a factor of about 2.5. This indicates a stronger mechanical coupling by reactive bonding as compared to adhesive bonding.

Query Translation Between AQL and CQL.

Secondary use of electronic health records using data aggregation systems (DAS) with standardized access interfaces (e.g. openEHR, i2b2, FHIR) have become an attractive approach to support clinical research. In order to increase the volume of underlying patient data, multiple DASs at different institutions can be connected to research networks. Two obstacles to connect a DAS to such a network are the syntactical differences between the involved DAS query interfaces and differences in the data models the DASs operate on. The current work presents an approach to tackle both problems by translating queries from a DAS using openEHR's query language AQL (Archetype Query Language) into queries using the query language CQL (Clinical Quality Language) and vice versa. For the subset of queries which are expressible in both query languages the presented approach is well feasible.

Query Translation Between openEHR and i2b2.

Secondary use of electronic health records using data warehouses (DW) has become an attractive approach to support clinical research. In order to increase the volume of underlying patient data DWs at different institutions can be connected to research networks. Two obstacles to connect a DW to such a network are the syntactical differences between the involved DW technologies and differences in the data models of the connected DWs. The current work presents an approach to tackle both problems by translating queries from the DW system openEHR into queries from the DW system i2b2 and vice versa. For the subset of queries expressible in the query languages of both systems, the presented approach is well feasible.

Replicating medication trend studies using ad hoc information extraction in a clinical data warehouse.

BACKGROUND: Medication trend studies show the changes of medication over the years and may be replicated using a clinical Data Warehouse (CDW). Even nowadays, a lot of the patient information, like medication data, in the EHR is stored in the format of free text. As the conventional approach of information extraction (IE) demands a high developmental effort, we used ad hoc IE instead. This technique queries information and extracts it on the fly from texts contained in the CDW. METHODS: We present a generalizable approach of ad hoc IE for pharmacotherapy (medications and their daily dosage) presented in hospital discharge letters. We added import and query features to the CDW system, like error tolerant queries to deal with misspellings and proximity search for the extraction of the daily dosage. During the data integration process in the CDW, negated, historical and non-patient context data are filtered. For the replication studies, we used a drug list grouped by ATC (Anatomical Therapeutic Chemical Classification System) codes as input for queries to the CDW. RESULTS: We achieve an F1 score of 0.983 (precision 0.997, recall 0.970) for extracting medication from discharge letters and an F1 score of 0.974 (precision 0.977, recall 0.972) for extracting the dosage. We replicated three published medical trend studies for hypertension, atrial fibrillation and chronic kidney disease. Overall, 93% of the main findings could be replicated, 68% of sub-findings, and 75% of all findings. One study could be completely replicated with all main and sub-findings. CONCLUSION: A novel approach for ad hoc IE is presented. It is very suitable for basic medical texts like discharge letters and finding reports. Ad hoc IE is by definition more limited than conventional IE and does not claim to replace it, but it substantially exceeds the search capabilities of many CDWs and it is convenient to conduct replication studies fast and with high quality.

Information Extraction from Echocardiography Reports for a Clinical Follow-up Study-Comparison of Extracted Variables Intended for General Use in a Data Warehouse with Those Intended Specifically for the Study.

BACKGROUND: The interest in information extraction from clinical reports for secondary data use is increasing. But experience with the productive use of information extraction processes over time is scarce. A clinical data warehouse has been in use at our university hospital for several years, which also provides an information extraction of echocardiography reports developed for general use. OBJECTIVES: This study aims to illustrate the difficulties encountered, while using data from a preexisting information extraction process for a large clinical study. To compare the data from the preexisting process with the data obtained from a specially developed process designed to improve the quality and completeness of the study data. METHODS: We extracted the echocardiography variables for 440 patients from the general-use information extraction of the data warehouse (678 reports). Then we developed an information extraction process for the same variables but specifically for this study, with the aim to extract as much information as possible from the text. The extracted data of both processes were compared with a newly created gold standard defined by a cardiologist with long-standing experience in heart failure. RESULTS: Among 57 echocardiography variables considered relevant for the study, 50 were documented in the routine text reports and could be extracted. Twenty of the required variables were not provided by the general-use extraction process, some others were not provided correctly. The median macro F1-score (precision, recall) across the 30 variables for which values were extracted was 0.81 (0.94, 0.77). Across all 50 variables, as relevant for the study, median macro F1-score was only 0.49 (0.56, 0.46). Employing the study-specific approach considerably improved the quality and completeness of the variables, resulting in F1-scores of 0.97 (0.98, 0.96) across all variables. CONCLUSION: Data from information extractions can be used for large clinical studies. However, preexisting information extraction processes should be treated with caution, as the time and effort spent defining each variable in the information extraction process may not be clear.

Finding Needles in the Haystack: Identifying Patients with Rare Subtype of Multiple Myeloma Supported by a Data Warehouse and Information Extraction.

Finding patient cases with extremely rare pathologies is a laborious task. To decrease time spent on manually searching through thousands of discharge letters and reports, a data warehouse with a fast fulltext search index was queried. Our use case is to find "macrofocal myeloma", i.e. Multiple Myeloma patients with few large lesions. We guessed the number of those patients in the University Hospital Würzburg at about 20. Most criteria were available in the data warehouse in an unstructured form requiring information extraction. 8 patient cases were found by searching for different spellings of "macrofocal myeloma" in discharge letters directly. With an indirect search combining several criteria, we found additional 23 candidate patient cases, from which 10 were classified by a domain expert as correct. The most difficult criteria were determining the degree of bone marrow infiltration. We achieved an F1 score of 93.2 % for this task. The number of patient cases to be screened manually for this disease decreased from about 25000 to 23.

Impact of primary tumour location and RAS/BRAF mutational status in metastatic colorectal cancer treated with first-line regimens containing oxaliplatin and bevacizumab: Prognostic factors from the AIO KRK0207 first-line and maintenance therapy trial.

BACKGROUND: The major prognostic relevance of primary tumour location (LPT) in advanced colorectal cancer was shown in large retrospective studies, but quantitative estimates are highly heterogeneous, and there is still limited information about its impact within the framework of biomarker-guided treatment strategies. Therefore, we analysed LPT in relation to other clinical and molecular parameters, based on mature survival data from the recent randomised AIO KRK0207 trial. METHODS: Patients uniformly received first-line induction treatment with a combination of bevacizumab, oxaliplatin and fluoropyrimidine. LPT was retrospectively determined using surgical reports, pathology reports and endoscopy reports. The prognostic analyses were performed using Kaplan-Meier estimations and log-rank tests, while hazard ratios (HRs) and multivariable results were derived from Cox models. RESULTS: Among 754 patients with unequivocal information on LPT, patients with left-sided tumours showed a median overall survival of 24.8 months compared with the right-sided cohort with 18.4 months (HR: 1.54, 95% confidence interval: 1.30-1.81, P < 0.0001). In a multivariable model, LPT proved to be the strongest prognosticator (HR 1.60), with performance status, number of metastatic sites, baseline carcinoembryonic antigen (CEA) and platelets independently retaining prognostic significance. In the subgroup of patients with known RAS/BRAF status (n = 567, 75%), a BRAF mutation showed the greatest unfavourable impact (HR 3.16). Although BRAF is strongly correlated to LPT, the latter remained a significant prognosticator in the BRAF wild-type subgroup. In contrast, no major impact of LPT was seen on tumours carrying RAS mutations. CONCLUSIONS: Within the framework of a uniform treatment strategy according to the current standards, LPT proved to have an important, although not solely dominating, relevance for survival prognosis. Its impact seems to be low in tumours with a RAS mutation. REGISTRATION: ClinicalTrials.govNCT00973609.

Ad Hoc Information Extraction for Clinical Data Warehouses.

BACKGROUND: Clinical Data Warehouses (CDW) reuse Electronic health records (EHR) to make their data retrievable for research purposes or patient recruitment for clinical trials. However, much information are hidden in unstructured data like discharge letters. They can be preprocessed and converted to structured data via information extraction (IE), which is unfortunately a laborious task and therefore usually not available for most of the text data in CDW. OBJECTIVES: The goal of our work is to provide an ad hoc IE service that allows users to query text data ad hoc in a manner similar to querying structured data in a CDW. While search engines just return text snippets, our systems also returns frequencies (e.g. how many patients exist with "heart failure" including textual synonyms or how many patients have an LVEF < 45) based on the content of discharge letters or textual reports for special investigations like heart echo. Three subtasks are addressed: (1) To recognize and to exclude negations and their scopes, (2) to extract concepts, i.e. Boolean values and (3) to extract numerical values. METHODS: We implemented an extended version of the NegEx-algorithm for German texts that detects negations and determines their scope. Furthermore, our document oriented CDW PaDaWaN was extended with query functions, e.g. context sensitive queries and regex queries, and an extraction mode for computing the frequencies for Boolean and numerical values. RESULTS: Evaluations in chest X-ray reports and in discharge letters showed high F1-scores for the three subtasks: Detection of negated concepts in chest X-ray reports with an F1-score of 0.99 and in discharge letters with 0.97; of Boolean values in chest X-ray reports about 0.99, and of numerical values in chest X-ray reports and discharge letters also around 0.99 with the exception of the concept age. DISCUSSION: The advantages of an ad hoc IE over a standard IE are the low development effort (just entering the concept with its variants), the promptness of the results and the adaptability by the user to his or her particular question. Disadvantage are usually lower accuracy and confidence.This ad hoc information extraction approach is novel and exceeds existing systems: Roogle [1] extracts predefined concepts from texts at preprocessing and makes them retrievable at runtime. Dr. Warehouse [2] applies negation detection and indexes the produced subtexts which include affirmed findings. Our approach combines negation detection and the extraction of concepts. But the extraction does not take place during preprocessing, but at runtime. That provides an ad hoc, dynamic, interactive and adjustable information extraction of random concepts and even their values on the fly at runtime. CONCLUSIONS: We developed an ad hoc information extraction query feature for Boolean and numerical values within a CDW with high recall and precision based on a pipeline that detects and removes negations and their scope in clinical texts.

Exporting Data from a Clinical Data Warehouse.

Data Warehouses (DW) are useful tools to support clinical studies as they can provide exports of routine care data for scientific reuse. Exported DW data is usually post-processed and integrated into study databases by study staff that is reasonably trained in specific tools like SPSS and Excel but which are no programmers or computer scientists. DW systems should therefore be configurable to satisfy export format desiderata as much as possible so that exports contain no unnecessary post-processing obstacles. In the presented work the authors analyze various existing DW systems in respect to a list of potential export formats.

Underestimated prevalence of heart failure in hospital inpatients: a comparison of ICD codes and discharge letter information.

BACKGROUND: Heart failure is the predominant cause of hospitalization and amongst the leading causes of death in Germany. However, accurate estimates of prevalence and incidence are lacking. Reported figures originating from different information sources are compromised by factors like economic reasons or documentation quality. METHODS: We implemented a clinical data warehouse that integrates various information sources (structured parameters, plain text, data extracted by natural language processing) and enables reliable approximations to the real number of heart failure patients. Performance of ICD-based diagnosis in detecting heart failure was compared across the years 2000-2015 with (a) advanced definitions based on algorithms that integrate various sources of the hospital information system, and (b) a physician-based reference standard. RESULTS: Applying these methods for detecting heart failure in inpatients revealed that relying on ICD codes resulted in a marked underestimation of the true prevalence of heart failure, ranging from 44% in the validation dataset to 55% (single year) and 31% (all years) in the overall analysis. Percentages changed over the years, indicating secular changes in coding practice and efficiency. Performance was markedly improved using search and permutation algorithms from the initial expert-specified query (F1 score of 81%) to the computer-optimized query (F1 score of 86%) or, alternatively, optimizing precision or sensitivity depending on the search objective. CONCLUSIONS: Estimating prevalence of heart failure using ICD codes as the sole data source yielded unreliable results. Diagnostic accuracy was markedly improved using dedicated search algorithms. Our approach may be transferred to other hospital information systems.

Estimating a Bias in ICD Encodings for Billing Purposes.

ICD encoded diagnoses are a popular criterion for eligibility algorithms for study cohort recruitment. However, "official" ICD encoded diagnoses used for billing purposes are afflicted with a bias originating from legal issues. This work presents an approach to estimate the degree of the encoding bias for the complete ICD catalogue at a German university hospital. The free text diagnoses sections of discharge letters are automatically classified using a supervised machine learning algorithm. The automatic classifications are compared with the official, manually classified codes. For selected ICD codes the approach works sufficiently well.

Semi-Automatic Terminology Generation for Information Extraction from German Chest X-Ray Reports.

Extraction of structured data from textual reports is an important subtask for building medical data warehouses for research and care. Many medical and most radiology reports are written in a telegraphic style with a concatenation of noun phrases describing the presence or absence of findings. Therefore a lexico-syntactical approach is promising, where key terms and their relations are recognized and mapped on a predefined standard terminology (ontology). We propose a two-phase algorithm for terminology matching: In the first pass, a local terminology for recognition is derived as close as possible to the terms used in the radiology reports. In the second pass, the local terminology is mapped to a standard terminology. In this paper, we report on an algorithm for the first step of semi-automatic generation of the local terminology and evaluate the algorithm with radiology reports of chest X-ray examinations from Würzburg university hospital. With an effort of about 20 hours work of a radiologist as domain expert and 10 hours for meetings, a local terminology with about 250 attributes and various value patterns was built. In an evaluation with 100 randomly chosen reports it achieved an F1-Score of about 95% for information extraction.

A Customizable Importer for the Clinical Data Warehouses PaDaWaN and I2B2.

In recent years, clinical data warehouses (CDW) storing routine patient data have become more and more popular to support scientific work in the medical domain. Although CDW systems provide interfaces to import new data, these interfaces have to be used by processing tools that are often not included in the systems themselves. In order to establish an extraction-transformation-load (ETL) workflow, already existing components have to be taken or new components have to be developed to perform the load part of the ETL. We present a customizable importer for the two CDW systems PaDaWaN and I2B2, which is able to import the most common import formats (plain text, CSV and XML files). In order to be run, the importer only needs a configuration file with the user credentials for the target CDW and a list of XML import configuration files, which determine how already exported data is indented to be imported. The importer is provided as a Java program, which has no further software requirements.

Extending the Query Language of a Data Warehouse for Patient Recruitment.

Patient recruitment for clinical trials is a laborious task, as many texts have to be screened. Usually, this work is done manually and takes a lot of time. We have developed a system that automates the screening process. Besides standard keyword queries, the query language supports extraction of numbers, time-spans and negations. In a feasibility study for patient recruitment from a stroke unit with 40 patients, we achieved encouraging extraction rates above 95% for numbers and negations and ca. 86% for time spans.

Effects of HIFU induced cavitation on flooded lung parenchyma.

BACKGROUND: High intensity focused ultrasound (HIFU) has gained clinical interest as a non-invasive local tumour therapy in many organs. In addition, it has been shown that lung cancer can be targeted by HIFU using One-Lung Flooding (OLF). OLF generates a gas free saline-lung compound in one lung wing and therefore acoustic access to central lung tumours. It can be assumed that lung parenchyma is exposed to ultrasound intensities in the pre-focal path and in cases of misguiding. If so, cavitation might be induced in the saline fraction of flooded lung and cause tissue damage. Therefore this study was aimed to determine the thresholds of HIFU induced cavitation and tissue erosion in flooded lung. METHODS: Resected human lung lobes were flooded ex-vivo. HIFU (1,1 MHz) was targeted under sonographic guidance into flooded lung parenchyma. Cavitation events were counted using subharmonic passive cavitation detection (PCD). B-Mode imaging was used to detect cavitation and erosion sonographically. Tissue samples out of the focal zone were analysed histologically. RESULTS: In flooded lung, a PCD and a sonographic cavitation detection threshold of 625 Wcm- 2(pr = 4, 3 MPa) and 3.600 Wcm- 2(pr = 8, 3 MPa) was found. Cavitation in flooded lung appears as blurred hyperechoic focal region, which enhances echogenity with insonation time. Lung parenchyma erosion was detected at intensities above 7.200 Wcm- 2(pr = 10, 9 MPa). CONCLUSIONS: Cavitation occurs in flooded lung parenchyma, which can be detected passively and by B-Mode imaging. Focal intensities required for lung tumour ablation are below levels where erosive events occur. Therefore focal cavitation events can be monitored and potential risk from tissue erosion in flooded lung avoided.

Data Linkage from Clinical to Study Databases via an R Data Warehouse User Interface. Experiences from a Large Clinical Follow-up Study.

BACKGROUND: Data that needs to be documented for clinical studies has often been acquired and documented in clinical routine. Usually this data is manually transferred to Case Report Forms (CRF) and/or directly into an electronic data capture (EDC) system. OBJECTIVES: To enhance the documentation process of a large clinical follow-up study targeting patients admitted for acutely decompensated heart failure by accessing the data created during routine and study visits from a hospital information system (HIS) and by transferring it via a data warehouse (DWH) into the study's EDC system. METHODS: This project is based on the clinical DWH developed at the University of Würzburg. The DWH was extended by several new data domains including data created by the study team itself. An R user interface was developed for the DWH that allows to access its source data in all its detail, to transform data as comprehensively as possible by R into study-specific variables and to support the creation of data and catalog tables. RESULTS: A data flow was established that starts with labeling patients as study patients within the HIS and proceeds with updating the DWH with this label and further data domains at a daily rate. Several study-specific variables were defined using the implemented R user interface of the DWH. This system was then used to export these variables as data tables ready for import into our EDC system. The data tables were then used to initialize the first 296 patients within the EDC system by pseudonym, visit and data values. Afterwards, these records were filled with clinical data on heart failure, vital parameters and time spent on selected wards. CONCLUSIONS: This solution focuses on the comprehensive access and transformation of data for a DWH-EDC system linkage. Using this system in a large clinical study has demonstrated the feasibility of this approach for a study with a complex visit schedule.

Maintenance strategies after first-line oxaliplatin plus fluoropyrimidine plus bevacizumab for patients with metastatic colorectal cancer (AIO 0207): a randomised, non-inferiority, open-label, phase 3 trial.

BACKGROUND: The definition of a best maintenance strategy following combination chemotherapy plus bevacizumab in metastatic colorectal cancer is unclear. We investigated whether no continuation of therapy or bevacizumab alone are non-inferior to fluoropyrimidine plus bevacizumab, following induction treatment with a fluoropyrimidine plus oxaliplatin plus bevacizumab. METHODS: In this open-label, non-inferiority, randomised phase 3 trial, we included patients aged 18 years or older with histologically confirmed, previously untreated metastatic colorectal cancer, Eastern Cooperative Oncology Group (ECOG) performance status of 0-2, adequate bone marrow, liver, and renal function, no pre-existing neuropathy greater than grade 1, and measurable disease, from 55 hospitals and 51 private practices in Germany. After 24 weeks of induction therapy with either fluorouracil plus leucovorin plus oxaliplatin or capecitabine plus oxaliplatin, both with bevacizumab, patients without disease progression were randomly assigned centrally by fax (1:1:1) to standard maintenance treatment with a fluoropyrimidine plus bevacizumab, bevacizumab alone, or no treatment. Both patients and investigators were aware of treatment assignment. Stratification criteria were response status, termination of oxaliplatin, previous adjuvant treatment with oxaliplatin, and ECOG performance status. At first progression, re-induction with all drugs of the induction treatment was a planned part of the protocol. Time to failure of strategy was the primary endpoint, defined as time from randomisation to second progression after maintenance (and if applicable re-induction), death, or initiation of further treatment including a new drug. Time to failure of strategy was equivalent to time to first progression for patients who did not receive re-induction (for any reason). The boundary for assessment of non-inferiority was upper limit of the one-sided 98·8% CI 1·43. Analyses were done by intention to treat. The study has completed recruitment, but follow-up of participants is ongoing. The trial is registered with ClinicalTrials.gov, number NCT00973609. FINDINGS: Between Sept 17, 2009, and Feb 21, 2013, 837 patients were enrolled and 472 randomised; 158 were randomly assigned to receive fluoropyrimidine plus bevacizumab, 156 to receive bevacizumab monotherapy, and 158 to receive no treatment. Median follow-up from randomisation is 17·0 months (IQR 9·5-25·4). Median time to failure of strategy was 6·9 months (95% CI 6·1-8·5) for the fluoropyrimidine plus bevacizumab group, 6·1 months (5·3-7·4) for the bevacizumab alone group, and 6·4 months (4·8-7·6) for the no treatment group. Bevacizumab alone was non-inferior to standard fluoropyrimidine plus bevacizumab (hazard ratio [HR] 1·08 [95% CI 0·85-1·37]; p=0·53; upper limit of the one-sided 99·8% CI 1·42), whereas no treatment was not (HR 1·26 [0·99-1·60]; p=0·056; upper limit of the one-sided 99·8% CI 1·65). The protocol-defined re-induction after first progression was rarely done (30 [19%] patients in the fluoropyrimidine plus bevacizumab group, 67 [43%] in the bevacizumab monotherapy group, and 73 [46%] in the no treatment group. The most common grade 3 adverse event was sensory neuropathy (21 [13%] of 158 patients in the fluoropyrimidine plus bevacizumab group, 22 [14%] of 156 patients in the bevacizumab alone group, and 12 [8%] of 158 patients in the no treatment group). INTERPRETATION: Although non-inferiority for bevacizumab alone was demonstrated for the primary endpoint, maintenance treatment with a fluoropyrimidine plus bevacizumab may be the preferable option for patients following an induction treatment with a fluoropyrimidine, oxaliplatin, and bevacizumab, as it allows the planned discontinuation of the initial combination without compromising time with controlled disease. Only a few patients were exposed to re-induction treatment, thus deeming the primary endpoint time to failure of strategy non-informative and clinically irrelevant. Progression-free survival and overall survival should be considered primary endpoints in future trials exploring maintenance strategies.