Stöbern Sie in den Artikeln und gewinnen Sie neue Einsichten!
Infomed: first feedbacks The first phase of the project Infomed is now in production. At this point, medical documents from the hospitals are being sent to private doctors. A pilot with 10 doctors gave us 2 main feedback subjects: Firstly, all the types of documents shared are judged “Helpful” to “Very helpful”. Secondly, it is essential to have a strong integration of the eHealth platform directly in the local patient record in order to save time and have added values. Also, during this step some technical problems arose regarding the authentication and the use of the X.509 certificate of the Health Professional Card (HPC).
Development and implementation of a new approach for therapeutic drug monitoring and dosage individualization for the patients will significantly improve modern therapeutics. To make it possible we aim at tackling the following questions: how to achieve (semantic) interoperability and medical data integration in a distributed environment; and how to ensure privacy of the patients in case of sharing and aggregation health data for research purposes.
The fatal consequences of implementing a HL7 ADT interface One of our sites implemented a HL7 ADT interface for demographic patient data between the hospital information system (HIS) and the radiology information system (RIS). In the RIS, they implemented a rule that merges patient records if these have an identical patient identification number (PID). They didn’t realise that the two systems had overlapping number ranges. As a result, they performed many merges on different patients. Following the RIS, the picture archiving and communication system (PACS) also merged the patient dossiers and changed the corresponding DICOM-header. When the mistake was discovered, we managed to separate the patients in the RIS but not in the PACS. This implicates some decisions and workarounds for the daily tasks of our radiologists and for the future migration to a new RIS/PACS.
To help manage the large amount of biomedical images produced, image information retrieval tools have been developed to help access the right information at the right moment. To provide a test bed for image retrieval evaluation, the ImageCLEFmed benchmark proposes a biomedical classification task that automatically focuses on determining the image modality of figures from biomedical journal articles. In the training data for this machine learning task, some classes have many more images than others and thus a few classes are not well represented, which is a challenge for automatic image classification. To address this problem, an automatic training set expansion was first proposed. To improve the accuracy of the automatic training set expansion, a manual verification of the training set is done using the crowdsourcing platform Crowdflower. This platform allows the use of external persons to pay for the crowdsourcing or to use personal contacts free of charge. Crowdsourcing requires strict quality control or using trusted persons but it can quickly give access to a large number of judges and thus improve many machine learning tasks. Results show that the manual annotation of a large amount of biomedical images carried out in this project can help with image classification.
Physiological sensors network data aggregation Taken independently, the data produced by physiological sensors have many applications. However, only a tight integration of the signals will provide a global comprehension of the health status of an individual. In this article, we share our experience acquired when building a network of physiological sensors connected to the Arduino platform. We have conceived an architecture integrating captors, going from the connection to the manipulation of the produced data. Four key requirements have been identified: On the one hand, to provide a plug and play connectivity, and to communicate the data stream in a coherent way. On the other hand, to aggregate several data streams and to present these streams in a coherent manner. We propose to deal with the connection issue by delegating the responsibility to a remote platform. The data transfer itself will be done based on the existing standards. The automatic data aggregation is possible through a set of rules and algorithms. Finally, the visualisation can be done by associating every signal to a specific visualisation widget. The connection of heterogeneous sensors is a complex task. In this article, we propose a concrete end to end solutions to the main challenges.
This article describes an application using Google Glass to allow emergency paramedics to access information from a hospital and send information from an accident side. This can potentially enhance pre-hospital care and also reduce costs. This article describes the goals of the application, its challenges and presents preliminary results of tests using Google Glass to prepare the final application. Results will be finalised in a bachelor thesis at the HES-SO Valais in Sierre.
Dans le cadre de ce papier nous rapportons un système de télémédecine pour le suivi de patients diabétiques à risque de développer des comorbidités. L’utilisation de capteurs multiparamétriques et ergonomiques vise à un suivi continu de l’état du patient tout en préservant sa qualité de vie. L’architecture système mise en place selon les spécifications Health Level 7 (HL7) assure l’interopérabilité entre les composants hétérogènes de la plateforme ainsi que son interconnexion avec d’éventuels systèmes tiers. Un système de raisonnement, basé sur des règles définies en accord avec des médecins, a pour but de lancer des alertes à destination du personnel médical en cas de la survenue de complications. Un essai clinique permettra d’évaluer la faisabilité de ce système de télémédecine en situation réelle en comparaison à l’approche conventionnelle.
Critical incidents happen in 2%‒8% of hospitalisations. The number of deaths by preventable adverse events is greater than those of flu, AIDS and traffic accidents combined. To reduce those preventable adverse events and to increase the patient’s safety, it is important to not only react when a preventable adverse event happened but also to be aware of the near misses. One possible way to detect near misses is to introduce a Critical Incident Reporting System (CIRS). The goal of this project was to develop a CIRS for mobile devices. Although mobile devices are not suitable for text input, they could possibly close a gap in the recording of critical incidents due to the fact that you can take these mobile devices with you as easily as a notepad and record your CI’s where you want (e.g. in a cafeteria). Therefore, the usability was one of the key elements in this project. To reach this goal the authors created a clickable mock-up, which was used to test the functionality with future end-users. On the basis of the results of the mock-up and the usability tests, the authors developed a final prototype based on the web technology HTML5.
With the eHealth Prescription App we created a prototype for Smartphones with the aim to remind patients to take their medicine and therefore to improve their adherence. Furthermore a concept for the eHealth Prescription App was created to describe how the App can be integrated into a prescription workflow involving doctors, pharmacies and Vivates – the eHealth platform from Swiss Post Solution AG.
Introduction of CPOE resulted in a high percentage of orders without defined durations of IV administrations. More importantly, the durations of injections and infusions specified were too short in 1.4% of the orders, documenting an important potential for improving prescriptions by CDS.
