Studies in health technology and informatics最新文献

Natural Language Processing (NLP) has shown promise in fields like radiology for converting unstructured into structured data, but acquiring suitable datasets poses several challenges, including privacy concerns. Specifically, we aim to utilize Large Language Models (LLMs) to extract medical information from dialogues between ambulance staff and patients to populate emergency protocol forms. However, we currently lack dialogues with known content that can serve as a gold standard for an evaluation. We designed a pipeline using the quantized LLM "Zephyr-7b-beta" for initial dialogue generation, followed by refinement and translation using OpenAI's GPT-4 Turbo. The MIMIC-IV database provided relevant medical data. The evaluation involved accuracy assessment via Retrieval-Augmented Generation (RAG) and sentiment analysis using multilingual models. Initial results showed a high accuracy of 94% with "Zephyr-7b-beta," slightly decreasing to 87% after refinement with GPT-4 Turbo. Sentiment analysis indicated a qualitative shift towards more positive sentiment post-refinement. These findings highlight the potential and challenges of using LLMs for generating synthetic medical dialogues, informing future NLP system development in healthcare.

Incorporating digital medicine into medical education equips students for the evolving landscape of healthcare. This study aimed to assess a digital medicine course developed at Bielefeld University by evaluating student attainment of learning outcomes outlined by Foadi et al. In the course, the students designed a digital application for various medical conditions, taking into account interdisciplinary factors. The course took place in 2023 with medical students who attended the course due to the focus of their studies. In a pilot study, the progress of ten participants was assessed using a pre-post survey design. Results revealed substantial improvement in students' achievement of learning outcomes post-course (median = 2, IQR 1-2) compared to pre-course (median = 3, IQR 3-4), suggesting the course's efficacy in effectively teaching digital medicine.

GNU Health, an open-source clinical information system, offers a comprehensive solution for managing health records, hospital information, and laboratory data. Despite its robust functionality and cost-effective nature, GNU Health remains underutilized in the European healthcare context. This paper explores the potential benefits of implementing GNU Health in European healthcare systems, emphasizing its capacity for customization, integration, and scalability. We also examine the barriers to its widespread adoption, including regulatory challenges, interoperability issues, and resistance to change from established proprietary systems. Through one case study and expert interviews, we provide insights into why these obstacles can hardly be overcome.

Medical informatics is a multidisciplinary field combining clinical and technical expertise. Addressing the challenge of aligning software design with clinicians' real-world needs, Dedalus established the Medical Office, a dedicated department designed to integrate clinical expertise directly into the software development process, in 2022. This paper details the approach and impact of the Medical Office. An international team of 15 healthcare professionals with experience in medical informatics was assembled. The team employed a multifaceted approach, incorporating global communication sessions and a ticketing system to track and analyze service requests. Over two years, 398 tickets were received, categorized into nine areas: clinical content curation, medical terminologies, clinical safety, clinical evaluation, design support, clinical UX, research & publication, real-world medical cases, and pre-sales support. The average duration of ticket resolution decreased over time, attributed to process fine-tuning and the formation of a relevant expert group. A preliminary satisfaction survey indicated positive feedback from technical teams. The collaborative model improved software design, usability, and clinical safety, demonstrating the value of clinician involvement. While preliminary results are promising, ongoing evaluation and adaptation are essential. The study emphasizes the importance of interdisciplinary collaboration in medical informatics and the benefits of clinician involvement in healthcare technology development. Future studies should explore this model's long-term impacts and scalability in other organizations and healthcare systems.

Animal owners may increasingly rely on large language models for gathering animal health information alongside internet sources in the future. This study therefore aims to provide initial results on the accuracy of ChatGPT-4o in triage and tentative diagnostics, using horses as a case study. Ten test vignettes were used to prompt situation assessments from the tool, which were then compared to original assessments made by a veterinary specialist for horses. The most probable diagnosis suggested by ChatGPT-4o was found to be quite accurate in most cases, with the urgency to contact a veterinarian sometimes assessed as higher than necessary. When provided with all relevant information, the tool does not seem to compromise horse health by recommending excessively long waiting times, although there is still potential for improving the relief of veterinarians' workload.

Sentinel surveillance networks are sophisticated health information systems that warn about outbreaks and spread of infectious diseases with epidemic or pandemic potential, the effectiveness of countermeasures and pressures on health systems. They are underpinned by their ability to turn data into information and knowledge in a timely way. The Royal College of General Practitioners (RCGP) Research and Surveillance Centre (RSC) is one of Europe's oldest. We report its progressive use of technology to improve the scope of sentinel surveillance, with a focus on genomic surveillance. The technologies include terminologies, phenotypes, compute capability, virology including virial genome sequencing, and serology. The RSC's data collection developed from partial, then full extraction of computerised medical record (CMR) data. with increasing sophistication in its creation of phenotypes. The scope of surveillance in 1967 was clinical diagnosis, influenza-like-illness (ILI) was its focus. In the 1992-1993 winter virology sampling started, with progressively more sophisticated sequencing of the viral genome. From 2008 viral sequencing was comprehensive with the Global Initiative on Sharing All Influenza Data (GISAID) the primary repository, supplemented by the COVID-19 Genomics UK (COG-UK) consortium in-pandemic. High quality primary care data captures sociodemographic features, risk group status, and vaccine exposure; linked hospital and death data informs about severe outcomes; virology identified the causative organism and genomic surveillance the variant. Timely data access and analysis will enable identification of new variants resistant to vaccination or other countermeasures and enable new interventions to be developed.

An increasing number of explainability methods began to emerge as a response for the black-box methods used to make decisions that could not be easily explained. This created the need for a better evaluation for these methods. In this paper we propose a new method for evaluation based on features. The main advantage of applying the proposed method to CNNs explanations are: a fully automated way to measure the quality of an explanation and the fact that the score uses the same information as the CNN, in this way being able to offer a measure of the quality of explanation that can be obtained automatically, ensuring that the human bias will not be present in the measurement of the explanation.

Usability is understood as a critical component to the success of electronic health records and other related healthcare technologies. Usability testing methods routinely employ scripts that help researchers understand how a particular tool works under real world conditions. This scoping review sought to better understand the guiding frameworks, principles, and methodologies employed when generating usability testing scripts to better understand how script generation occurs. Three main themes emerged through qualitative analysis: researchers sought to observe the baseline functionality being tested, the most representative tasks, or the most complex tasks. This scoping review highlights a lack of consistent processes in usability test script generation. There is a need to create standardized usability testing scripts for usability testing.

Access to the internet and online resources changes the concept of health and increases people's autonomy. In this context, Health Literacy (HL) is a critical determinant of health-related choices. At World Health Organization (WHO) level, M-POHL (Action Network on Measuring Population and Organizational Health Literacy of WHO-Europe) created and validated on European population four questionnaires: digital HL (HLS19-DIGI), communication HL (with doctors from health care services - HLS19-COM-P-Q11 long version and HLS19-COM-P-Q6 short version), online navigation HL (HLS19-NAV), and vaccination HL (HLS19-VAC). Based on the expertise of the team, the present study aimed to report the study protocol for Romanian translation, culturally adapting and psychometric testing the following three M-POHL health literacy tools: HLS19-DIGI, HLS19-NAV, and HLS19-COM-P-Q11, HLS19-COM-P-Q6. We will conduct a qualitative descriptive study design in seven steps to translate and adapt the HLS19-DIGI, HLS19-NAV, and HLS19-COM-P-Q11, HLS19-COM-P-Q6 to the Romanian speakers. The study will begin with the translation of English (En)-Romanian (Ro) (2 researchers involved) (step 1), followed by the evaluation of the translation by a bilingual researcher independent of the two researchers who did the En-Ro translation (step 2), the translation of Ro-En (2 researchers but not those in step 1; step 3), the evaluation of the translation by a bilingual researcher independent of the two researchers who did the Ro-En translation (step 4), evaluation of the translation of the tool in an expert group (step 5), pilot testing on a sample of the target population (step 6) and full psychometric testing of the version resulting from step 6 (step 7).

Prostate cancer (PC) is one of the leading causes of cancer-related mortality among men. Androgen Deprivation Therapy (ADT) has been shown to increase survival in men with metastatic PC. Despite its efficacy, ADT's long-term use adversely impacts quality of life (QoL) due to multiple side effects. Multimodal cancer rehabilitation (CR) programs improve QoL but face limited uptake. This study explores perspectives of patients with metastatic PC on a Home Automated Telemanagement (HAT) system using semi-structured interviews. Findings indicate that patients appreciated the ease of use, guided exercise content, and progress tracking features of the HAT system. Significant benefits in symptom management, cognitive support, and QoL improvements were reported, suggesting the program's potential for long-term health management. In summary, the HAT system is a promising tool for supporting CR in patients with metastatic PC. Future developments should address technical issues, incorporate motivational elements, and enhance user feedback to optimize patient engagement and satisfaction.