Prediction and simulation of PEEP setting effects with machine learning models / Predicción y simulación de los efectos de la configuración de la PEEP con modelos de aprendizaje automático

Objective To establish a new machine learning-based method to adjust positive end-expiratory pressure (PEEP) using only already routinely measured data. Design Retrospective observational study. Setting Intensive care unit (ICU). Patients or participants 51811 mechanically ventilated patients in multiple ICUs in the USA (data from MIMIC-III and eICU databases). Interventions No interventions. Main variables of interest Success parameters of ventilation (arterial partial pressures of oxygen and carbon dioxide and respiratory system compliance). Results The multi-tasking neural network model performed significantly best for all target tasks in the primary test set. The model predicts arterial partial pressures of oxygen and carbon dioxide and respiratory system compliance about 45 min into the future with mean absolute percentage errors of about 21.7%, 10.0% and 15.8%, respectively. The proposed use of the model was demonstrated in case scenarios, where we simulated possible effects of PEEP adjustments for individual cases. Conclusions Our study implies that machine learning approach to PEEP titration is a promising new method which comes with no extra cost once the infrastructure is in place. Availability of databases with most recent ICU patient data is crucial for the refinement of prediction performance. (AU)

Objetivo Establecer un nuevo método basado en el aprendizaje automático para ajustar la presión positiva al final de la espiración (PEEP según sus siglas en inglés) utilizando únicamente datos ya obtenidos de forma rutinaria. Diseño Estudio retrospectivo de observación. Ámbito Unidad de cuidados intesivos (UCI) Pacientes o participantes 51811 pacientes ventilados mecánicamente en múltiples UCIs de EE.UU. (tomados de las bases de datos MIMIC-III y eICU). Intervenciones Sin intervenciones. Variables de interés principales Parametros de éxito de la ventilación (presiones parciales arteriales de oxígeno y dióxido de carbono y distensibilidad del sistema respiratorio). Resultados El modelo de red neuronal multitarea obtuvo los mejores resultados en todos los objetivos del conjunto de pruebas primario. El modelo predice las presiones parciales arteriales de oxígeno y dióxido de carbono así como la distensibilidad del sistema respiratorio con aproximadamente 45 minutos de anticipación, mostrando errores porcentuales absolutos medios de aproximadamente 21.7%, 10.0% y 15.8%, respectivamente. El uso propuesto del modelo se demostró en situaciones hipotéticas en las que se simularon los posibles efectos de los ajustes de PEEP para casos individuales. Conclusiones Nuestro estudio implica que el enfoque de aprendizaje automático para el ajuste de la PEEP es un método nuevo y prometedor que no supone ningún coste adicional una vez que se dispone de la infraestructura necesaria. La disponibilidad de bases de datos con información de pacientes de UCI más recientes es crucial para perfeccionar el rendimiento de la predicción. (AU)

An Image Retrieval Pipeline in a Medical Data Integration Center.

Medical images need annotations with high-level semantic descriptors, so that domain experts can search for the desired dataset among an enormous volume of visual media within a Medical Data Integration Center. This article introduces a processing pipeline for storing and annotating DICOM and PNG imaging data by applying Elasticsearch, S3 and Deep Learning technologies. The proposed method processes both DICOM and PNG images to generate annotations. These image annotations are indexed in Elasticsearch with the corresponding raw data paths, where they can be retrieved and analyzed.

Structured Data Retrieval and Analysis of HL7 v2 Messages with Elasticsearch.

The era of the electronic health record (EHR) requires lots of semantic interoperability for data sharing and reusability. We select HL7 v2 messages as the most common structured data type in hospital information systems, to investigate the plausibility of using Elasticsearch (ES) as a healthcare search engine and data analytics tool. Due to the facts, Elasticsearch can be integrated as a powerful searchable database for practical healthcare applications, to analyze structured healthcare data from various locations. It allows easy and efficient searching for complex query tasks.

Prediction and simulation of PEEP setting effects with machine learning models.

OBJECTIVE: To establish a new machine learning-based method to adjust positive end-expiratory pressure (PEEP) using only already routinely measured data. DESIGN: Retrospective observational study. SETTING: Intensive care unit (ICU). PATIENTS OR PARTICIPANTS: 51811 mechanically ventilated patients in multiple ICUs in the USA (data from MIMIC-III and eICU databases). INTERVENTIONS: No interventions. MAIN VARIABLES OF INTEREST: Success parameters of ventilation (arterial partial pressures of oxygen and carbon dioxide and respiratory system compliance) RESULTS: The multi-tasking neural network model performed significantly best for all target tasks in the primary test set. The model predicts arterial partial pressures of oxygen and carbon dioxide and respiratory system compliance about 45 min into the future with mean absolute percentage errors of about 21.7%, 10.0% and 15.8%, respectively. The proposed use of the model was demonstrated in case scenarios, where we simulated possible effects of PEEP adjustments for individual cases. CONCLUSIONS: Our study implies that machine learning approach to PEEP titration is a promising new method which comes with no extra cost once the infrastructure is in place. Availability of databases with most recent ICU patient data is crucial for the refinement of prediction performance.

Functional Requirements for Medical Data Integration into Knowledge Management Environments: Requirements Elicitation Approach Based on Systematic Literature Analysis.

BACKGROUND: In patient care, data are historically generated and stored in heterogeneous databases that are domain specific and often noninteroperable or isolated. As the amount of health data increases, the number of isolated data silos is also expected to grow, limiting the accessibility of the collected data. Medical informatics is developing ways to move from siloed data to a more harmonized arrangement in information architectures. This paradigm shift will allow future research to integrate medical data at various levels and from various sources. Currently, comprehensive requirements engineering is working on data integration projects in both patient care- and research-oriented contexts, and it is significantly contributing to the success of such projects. In addition to various stakeholder-based methods, document-based requirement elicitation is a valid method for improving the scope and quality of requirements. OBJECTIVE: Our main objective was to provide a general catalog of functional requirements for integrating medical data into knowledge management environments. We aimed to identify where integration projects intersect to derive consistent and representative functional requirements from the literature. On the basis of these findings, we identified which functional requirements for data integration exist in the literature and thus provide a general catalog of requirements. METHODS: This work began by conducting a literature-based requirement elicitation based on a broad requirement engineering approach. Thus, in the first step, we performed a web-based systematic literature review to identify published articles that dealt with the requirements for medical data integration. We identified and analyzed the available literature by applying the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. In the second step, we screened the results for functional requirements using the requirements engineering method of document analysis and derived the requirements into a uniform requirement syntax. Finally, we classified the elicited requirements into a category scheme that represents the data life cycle. RESULTS: Our 2-step requirements elicitation approach yielded 821 articles, of which 61 (7.4%) were included in the requirement elicitation process. There, we identified 220 requirements, which were covered by 314 references. We assigned the requirements to different data life cycle categories as follows: 25% (55/220) to data acquisition, 35.9% (79/220) to data processing, 12.7% (28/220) to data storage, 9.1% (20/220) to data analysis, 6.4% (14/220) to metadata management, 2.3% (5/220) to data lineage, 3.2% (7/220) to data traceability, and 5.5% (12/220) to data security. CONCLUSIONS: The aim of this study was to present a cross-section of functional data integration-related requirements defined in the literature by other researchers. The aim was achieved with 220 distinct requirements from 61 publications. We concluded that scientific publications are, in principle, a reliable source of information for functional requirements with respect to medical data integration. Finally, we provide a broad catalog to support other scientists in the requirement elicitation phase.

3LGM2IHE: Requirements for Data-Protection-Compliant Research Infrastructures-A Systematic Comparison of Theory and Practice-Oriented Implementation.

OBJECTIVES: The TMF (Technology, Methods, and Infrastructure for Networked Medical Research) Data Protection Guide (TMF-DP) makes path-breaking recommendations on the subject of data protection in research projects. It includes comprehensive requirements for applications such as patient lists, pseudonymization services, and consent management services. Nevertheless, it lacks a structured, categorized list of requirements for simplified application in research projects and systematic evaluation. The 3LGM2IHE ("Three-layer Graphbased meta model - Integrating the Healthcare Enterprise [IHE] " ) project is funded by the German Research Foundation (DFG). 3LGM2IHE aims to define modeling paradigms and implement modeling tools for planning health care information systems. In addition, one of the goals is to create and publish 3LGM2 information system architecture design patterns (short "design patterns") for the community as design models in terms of a framework. A structured list of data protection-related requirements based on the TMF-DP is a precondition to integrate functions (3LGM2 Domain Layer) and building blocks (3LGM2 Logical Tool Layer) in 3LGM2 design patterns. METHODS: In order to structure the continuous text of the TMF-DP, requirement types were defined in a first step. In a second step, dependencies and delineations of the definitions were identified. In a third step, the requirements from the TMF-DP were systematically extracted. Based on the identified lists of requirements, a fourth step included the comparison of the identified requirements with exemplary open source tools as provided by the "Independent Trusted Third Party of the University Medicine Greifswald" (TTP tools). RESULTS: As a result, four lists of requirements were created, which contain requirements for the "patient list", the "pseudonymization service", and the "consent management", as well as cross-component requirements from the TMF-DP chapter 6 in a structured form. Further to requirements (1), possible variants (2) of implementations (to fulfill a single requirement) and recommendations (3) were identified. A comparison of the requirements lists with the functional scopes of the open source tools E-PIX (record linkage), gPAS (pseudonym management), and gICS (consent management) has shown that these fulfill more than 80% of the requirements. CONCLUSIONS: A structured set of data protection-related requirements facilitates a systematic evaluation of implementations with respect to the fulfillment of the TMF-DP guidelines. These re-usable lists provide a decision aid for the selection of suitable tools for new research projects. As a result, these lists form the basis for the development of data protection-related 3LGM2 design patterns as part of the 3LGM2IHE project.

Cofactors of drug hypersensitivity-A monocenter retrospective analysis.

Background: Drug hypersensitivity reactions (DHRs) are major medical problems that influence the treatment of patients by both under- and overdiagnosis. Still, little is known about the role of predisposing or protecting cofactors of DHR. Objective: This study aims to determine drug-specific cofactors in patients with DHR. Methods: Retrospective file chart analysis of inpatients with suspected DHR in our department between 2015 and 2020 was performed. Descriptive statistics and multiple logistic regression were conducted for the estimation and statistical interference. Results: DHRs were suspected in 393 patients with 678 culprit drugs. In 183 cases, drug hypersensitivities were confirmed, mostly against nonopioid analgesic drugs and antibiotics. Multiple logistic regression analysis identified a positive association of antibiotic hypersensitivity with obesity [odds ratio (OR) 5.75, average marginal effect (AME) +24.4%] and age and a negative association with arterial hypertension, female sex, elevated immunoglobulin E (IgE), and allergic rhinitis. Hypersensitivity to nonopioid analgesics was associated with atopic dermatitis (OR 10.28, AME +28.5%), elevated IgE, and arterial hypertension. Conclusions: Drug-specific cofactors of DHR include obesity for antibiotics and atopic dermatitis for nonopioid analgesics, the knowledge of which may improve the risk calculation for drug provocation tests.

Applicability of Different Electronic Record Types for Use in Patient Recruitment Support Systems: Comparative Analysis.

BACKGROUND: Clinical trials constitute an important pillar in medical research. It is beneficial to support recruitment for clinical trials using software tools, so-called patient recruitment support systems; however, such information technology systems have not been frequently used to date. Because medical information systems' underlying data collection methods strongly influence the benefits of implementing patient recruitment support systems, we investigated patient recruitment support system requirements and corresponding electronic record types such as electronic medical record, electronic health record, electronic medical case record, personal health record, and personal cross-enterprise health record. OBJECTIVE: The aim of this study was to (1) define requirements for successful patient recruitment support system deployment and (2) differentiate and compare patient recruitment support system-relevant properties of different electronic record types. METHODS: In a previous study, we gathered requirements for patient recruitment support systems from literature and unstructured interviews with stakeholders (15 patients, 3 physicians, 5 data privacy experts, 4 researchers, and 5 staff members of hospital administration). For this investigation, the requirements were amended and categorized based on input from scientific sessions. Based on literature with a focus on patient recruitment support system-relevant properties, different electronic record types (electronic medical record, electronic health record, electronic medical case record, personal health record and personal cross-enterprise health record) were described in detail. We also evaluated which patient recruitment support system requirements can be achieved for each electronic record type. RESULTS: Patient recruitment support system requirements (n=16) were grouped into 4 categories (consent management, patient recruitment management, trial management, and general requirements). All 16 requirements could be partially met by at least 1 type of electronic record. Only 1 requirement was fully met by all 5 types. According to our analysis, personal cross-enterprise health records fulfill most requirements for patient recruitment support systems. They demonstrate advantages especially in 2 domains (1) supporting patient empowerment and (2) granting access to the complete medical history of patients. CONCLUSIONS: In combination with patient recruitment support systems, personal cross-enterprise health records prove superior to other electronic record types, and therefore, this integration approach should be further investigated.

The Process of Modeling Information System Architectures with IHE.

IHE has defined more than 200 integration profiles in order to improve the interoperability of application systems in healthcare. These profiles describe how standards should be used in particular use cases. These profiles are very helpful but their correct use is challenging, if the user is not familiar to the specifications. Therefore, inexperienced modelers of information systems quickly lose track of existing IHE profiles. In addition, the users of these profiles are often not aware of rules that are defined within these profiles and of dependencies that exist between the profiles. There are also modelers that do not notice the differences between the implemented actors, because they do not know the optional capabilities of some actors. The aim of this paper is therefore to describe a concept how modelers of information systems can be supported in the selection and use of IHE profiles and how this concept was prototypically implemented in the "Three-layer Graph-based meta model" modeling tool (3LGM2 Tool). The described modeling process consists of the following steps that can be looped: defining the use case, choosing suitable integration profiles, choosing actors and their options and assigning them to application systems, checking for required actor groupings and modeling transactions. Most of these steps were implemented in the 3LGM2 Tool. Further implementation effort and evaluation of our approach by inexperienced users is needed. But after that our tool should be a valuable tool for modelers planning healthcare information system architectures, in particular those based on IHE.

Barriers and Facilitators to the Implementation of eHealth Services: Systematic Literature Analysis.

BACKGROUND: The field of eHealth has a history of more than 20 years. During that time, many different eHealth services were developed. However, factors influencing the adoption of such services were seldom the main focus of analyses. For this reason, organizations adopting and implementing eHealth services seem not to be fully aware of the barriers and facilitators influencing the integration of eHealth services into routine care. OBJECTIVE: The objective of this work is to provide (1) a comprehensive list of relevant barriers to be considered and (2) a list of facilitators or success factors to help in planning and implementing successful eHealth services. METHODS: For this study, a twofold approach was applied. First, we gathered experts' current opinions on facilitators and barriers in implementing eHealth services via expert discussions at two health informatics conferences held in Europe. Second, we conducted a systematic literature analysis concerning the barriers and facilitators for the implementation of eHealth services. Finally, we merged the results of the expert discussions with those of the systematic literature analysis. RESULTS: Both expert discussions (23 and 10 experts, respectively) identified 15 barriers and 31 facilitators, whereas 76 barriers and 268 facilitators were found in 38 of the initial 56 articles published from 12 different countries. For the analyzed publications, the count of distinct barriers reported ranged from 0 to 40 (mean 10.24, SD 8.87, median 8). Likewise, between 0 and 48 facilitators were mentioned in the literature (mean 9.18, SD 9.33, median 6). The combination of both sources resulted in 77 barriers and 292 facilitators for the adoption and implementation of eHealth services. CONCLUSIONS: This work contributes a comprehensive list of barriers and facilitators for the implementation and adoption of eHealth services. Addressing barriers early, and leveraging facilitators during the implementation, can help create eHealth services that better meet the needs of users and provide higher benefits for patients and caregivers.

Mapping Medication Metadata from the ABDA Data Model to an OpenEHR Medication Archetype: A Qualitative Analysis.

Integrating data from various source systems to gain knowledge and meaningful data about patients for care and research is challenging. This work demonstrates how medication knowledge data from the database of the Federal Union of German Associations of Pharmacists (ABDA) can be used for storing and annotating medicinal products in an openEHR medication archetype.

Towards a Software Tool for Planning IHE-Compliant Information Systems.

As hospital information systems are complex and the requirements for interoperability grow with the increasing networking in healthcare, careful planning becomes more and more necessary. The use of standards as described in IHE profiles, for example, are an important prerequisite for enabling interoperability. Enterprise Architecture Planning (EAP) methods should support this, but none of the currently available EAP methods offers the option of using IHE profiles. The 3LGM2IHE project wants to close this gap and implement the support of IHE profiles in the 3LGM2 tool. This paper describes how requirements for this tool were determined and presents the results.

MAGIC: once upon a time in consent management-a FHIR® tale.

BACKGROUND: The use of medical data for research purposes requires an informed consent of the patient that is compliant with the EU General Data Protection Regulation. In the context of multi-centre research initiatives and a multitude of clinical and epidemiological studies scalable and automatable measures for digital consent management are required. Modular form, structure, and contents render a patient's consent reusable for varying project settings in order to effectively manage and minimise organisational and technical efforts. RESULTS: Within the DFG-funded project "MAGIC" (Grant Number HO 1937/5-1) the digital consent management service tool gICS was enhanced to comply with the recommendations published in the TMF data protection guideline for medical research. In addition, a structured exchange format for modular consent templates considering established standards and formats in the area of digital informed consent management was designed. Using the new FHIR standard and the HAPI FHIR library, the first version for an exchange format and necessary import-/export-functionalities were successfully implemented. CONCLUSIONS: The proposed exchange format is a "work in progress". It represents a starting point for current discussions concerning digital consent management. It also attempts to improve interoperability between different approaches within the wider IHE-/HL7-/FHIR community. Independent of the exchange format, providing the possibility to export, modify and import templates for consents and withdrawals to be reused in similar clinical and epidemiological studies is an essential precondition for the sustainable operation of digital consent management.

HiGHmed - An Open Platform Approach to Enhance Care and Research across Institutional Boundaries.

INTRODUCTION: This article is part of the Focus Theme of Methods of Information in Medicine on the German Medical Informatics Initiative. HiGHmed brings together 24 partners from academia and industry, aiming at improvements in care provision, biomedical research and epidemiology. By establishing a shared information governance framework, data integration centers and an open platform architecture in cooperation with independent healthcare providers, the meaningful reuse of data will be facilitated. Complementary, HiGHmed integrates a total of seven Medical Informatics curricula to develop collaborative structures and processes to train medical informatics professionals, physicians and researchers in new forms of data analytics. GOVERNANCE AND POLICIES: We describe governance structures and policies that have proven effective during the conceptual phase. These were further adapted to take into account the specific needs of the development and networking phase, such as roll-out, carerelated aspects and our focus on curricula development in Medical Inform atics. ARCHITECTURAL FRAMEWORK AND METHODOLOGY: To address the challenges of organizational, technical and semantic interoperability, a concept for a scalable platform architecture, the HiGHmed Platform, was developed. We outline the basic principles and design goals of the open platform approach as well as the roles of standards and specifications such as IHE XDS, openEHR, SNOMED CT and HL7 FHIR. A shared governance framework provides the semantic artifacts which are needed to establish semantic interoperability. USE CASES: Three use cases in the fields of oncology, cardiology and infection control will demonstrate the capabilities of the HiGHmed approach. Each of the use cases entails diverse challenges in terms of data protection, privacy and security, including clinical use of genome sequencing data (oncology), continuous longitudinal monitoring of physical activity (cardiology) and cross-site analysis of patient movement data (infection control). DISCUSSION: Besides the need for a shared governance framework and a technical infrastructure, backing from clinical leaders is a crucial factor. Moreover, firm and sustainable commitment by participating organizations to collaborate in further development of their information system architectures is needed. Other challenges including topics such as data quality, privacy regulations, and patient consent will be addressed throughout the project.

From SOMDA to application - integration strategies in the OR.NET demonstration sites.

The effective development and dissemination of the open integration for the next generation of operating rooms require a comprehensive testing environment. In this paper, we present the various challenges to be addressed in demonstration applications, and we discuss the implementation approach, the foci of the demonstration sites and the evaluation efforts. Overall, the demonstrator setups have proven the feasibility of the service-oriented medical device architecture (SOMDA) and real-time approaches with a large variety of example applications. The applications demonstrate the potentials of open device interoperability. The demonstrator implementations were technically evaluated as well as discussed with many clinicians from various disciplines. However, the evaluation is still an ongoing research at the demonstration sites. Technical evaluation focused on the properties of a network of medical devices, latencies in data transmission and stability. A careful evaluation of the SOMDA design decisions and implementations are essential to a safe and reliable interoperability of integrated medical devices and information technology (IT) system in the especially critical working environment. The clinical evaluation addressed the demands of future users and stakeholders, especially surgeons, anesthesiologists, scrub nurses and hospital operators. The opinions were carefully collected to gain further insights into the potential benefits of the technology and pitfalls in future work.

Implementing the Patient Portal of the PEHR in the Metropolitan Region Rhine-Neckar - Experiences with Liferay.

A patient portal with access to a personal cross-enterprise electronic health record (PEHR) had to be developed for the project "INFormation technology for PATient-oriented healthcare in the Rhine-Neckar metropolitan region" (INFOPAT). Liferay was chosen as the platform technology in previous researches. Multiple portlets for patients, study members, security officers and administrators had been developed in Java. In this paper all portlets are presented. For three portlets the communication between them, Liferay and the PEHR are described in detail. To communicate with the PEHR a façade called IHE Connector had been implemented too. Despite the broad functionality of the Liferay Core some workarounds had to be implemented. The lack of information in the provided documentation of Liferay lead to delays in development. The Liferay IDE and Liferay itself showed an unstable state.

The Patient Portal of the Personal Cross-Enterprise Electronic Health Record (PEHR) in the Rhine-Neckar-Region.

Over the last years we stepwise implemented our vision of a personal cross-enterprise electronic health record (PEHR) in the Rhine-Neckar-Region in Germany. The patient portal is one part of the PEHR architecture with IHE connectivity. The patient is enabled to access and manage his medical record by use of the patient portal. Moreover, he can give his consent regarding which healthcare providers are allowed to send data into or read data from his medical record. Forthcoming studies will give evidence for improvements and further requirements to develop.

Patient Portal Integration - A Native IHE Connector Implementation for PEHR.

Providing patients with access to their medical data has recently evolved as a topic in several countries. Different approaches are possible. For example patient portals are used for patient access towards medical data. The University Hospital Heidelberg is engaged in a research project to develop a personal cross-enterprise electronic health record (PEHR). The objective of this work is to describe the architecture and implementation of a component called IHE Connector which represents the native IHE-based integration between the patient portal and the PEHR core components. The architecture of the PEHR is accepted based on international standards. The core components consist out of ready to use software products like a master patient index. The patient portal has been developed using Liferay framework. The IHE Connector is mainly based on the Open eHealth Integration Platform (IPF) Framework, which has been deeply integrated into the patient portal to support the needed IHE transactions. Several IHE profiles for sharing documents and patient information are supported by the IHE Connector. As IPF already provides interfaces for some IHE profiles others had to be developed from scratch. The IHE Connector can not only be used for connectivity between patient portal and PEHR core, but also provide connectivity for third party apps and healthcare providers' information systems.

Implementing PEHR: Design and Integration of a Consent Creator Service.

Giving the patient full control over his medical data electronically remains one of the most discussed topics in healthcare today. The INFOPAT project in the Rhine-Neckar region focuses on a personal cross-enterprise electronic health record (PEHR) in which the patient plays a major role. Thus, he should be provided with the possibility of granting access to his medical data which could be realized using a consent creator service. This paper presents a user interface concept for such a service as well as aspects for the technical implementation. In addition, a pattern for integrating the service into an existing IHE based infrastructure is shown. These concepts could be further adapted for improving patient empowerment in health care projects.

An Architecture for the Integration of Clinical Data from a PEHR in a Regional Research Platform.

Making clinical information available for research is not only relevant for healthcare institutions, but also for regional EHRs, as cross-sectorial information can be made accessible. In the INFOPAT (INFOrmation technology for PATient-oriented health care in the Rhine-Neckar metropolitan region) project we are thus implementing both, a regional personal cross-enterprise electronic health record (PEHR) and a regional research platform (RRP) based on information from the PEHR. IHE profiles are implemented to achieve interoperability between healthcare institutions electronic medical records (EMR) and PEHR on the one hand, as well as PEHR and RRP on the other hand. The use case for the RRP is cross-sectorial quality assessment and improvement for colorectal cancer based on a quality indicator (QI) approach including patients' perspectives. For semantic interoperability the responses are transferred in the form of HL7 CDA L2 documents. The resulting architecture for a RRP shows that implementing a PEHR in combination with a RRP based on international communication standards is possible. Also IHE XDS can be used for integration of patient care and biomedical research infrastructures.