International Journal of Medical Informatics最新文献

Background

Measuring satisfaction with telerehabilitation provides a way to evaluate and improve the effectiveness of both the technology used and the rehabilitation provided. On the other hand, valid and reliable tools are needed to evaluate satisfaction of patients receiving physiotherapy via telerehabilitation.

Aims

The purpose of the current study was to develop Telerehabilitation Satisfaction Questionnaire (TrSQ) and evaluate its validity and reliability.

Methods

Sixty-three patients with stroke, Multiple Sclerosis, or Parkinson’s disease participated in this study. Content validity was reviewed by a panel experienced in telerehabilitation. Construct validity of the model was investigated using and Confirmatory Factor Analysis (CFA) and Explanatory Factor Analysis (EFA). Test-retest reliability and Internal consistency were used to evaluate the reliability of the TrSQ.

Results

A one-factor structure was determined based on EFA. The structure fitted well in terms of the fit indices according to the confirmatory factor analysis results (x2/df = 1.016, p = 0.442, IFI=0.997, CFI=0.997, and RMSEA=0.016). The questionnaire was proven to have an acceptable reliability level (Cronbach’s alpha = 0.858) and it was found that all items were necessary. Finally, an 11-item version was obtained and tested twice on 30 patients. The questionnaire was shown to have acceptable test–retest reliability (ICC=0.753).

Conclusions

TrSQ can be used as a valid and reliable questionnaire in evaluating patient satisfaction with telerehabilitation in neurological diseases. However, in order for it to be widely applicable, adaptation to different languages is needed.

Purpose

Parametric regression models have been the main statistical method for identifying average treatment effects. Causal machine learning models showed promising results in estimating heterogeneous treatment effects in causal inference. Here we aimed to compare the application of causal random forest (CRF) and linear regression modelling (LRM) to estimate the effects of organisational factors on ICU efficiency.

Methods

A retrospective analysis of 277,459 patients admitted to 128 Brazilian and Uruguayan ICUs over three years. ICU efficiency was assessed using the average standardised efficiency ratio (ASER), measured as the average of the standardised mortality ratio (SMR) and the standardised resource use (SRU) according to the SAPS-3 score. Using a causal inference framework, we estimated and compared the conditional average treatment effect (CATE) of seven common structural and organisational factors on ICU efficiency using LRM with interaction terms and CRF.

Results

The hospital mortality was 14 %; median ICU and hospital lengths of stay were 2 and 7 days, respectively. Overall median SMR was 0.97 [IQR: 0.76,1.21], median SRU was 1.06 [IQR: 0.79,1.30] and median ASER was 0.99 [IQR: 0.82,1.21]. Both CRF and LRM showed that the average number of nurses per ten beds was independently associated with ICU efficiency (CATE [95 %CI]: −0.13 [-0.24, −0.01] and −0.09 [-0.17,-0.01], respectively). Finally, CRF identified some specific ICUs with a significant CATE in exposures that did not present a significant average effect.

Conclusion

In general, both methods were comparable to identify organisational factors significantly associated with CATE on ICU efficiency. CRF however identified specific ICUs with significant effects, even when the average effect was nonsignificant. This can assist healthcare managers in further in-dept evaluation of process interventions to improve ICU efficiency.

Extensive research has been devoted to predicting ICU mortality, to assist clinical teams managing critical patients. Electronic health records (EHR) contain both static and dynamic medical data, with the latter accumulating during ICU stays. Existing models often rely on a fixed time window (e.g., first 24 h) for prediction, potentially missing vital post-24-hour data. The present study aims to improve mortality prediction for ICU patients following Cardiac Arrest (CA) using a dynamic sliding window approach that accommodates evolving data characteristics.

Our cohort included 2331 CA patients, of whom 684 died in the ICU and 1647 survived. Applying the sliding window technique, we created six different time windows and used each separately for model training and validation. We compared our results to a baseline accumulative window.

The different time windows created by the sliding window technique differed in their prediction performance and outperformed the baseline 24-hour window significantly. The XGBoost model outperformed all other models, with the 30–42 h time window achieving the best results (AUC = 0.8, accuracy = 0.77).

Our work shows that the sliding window technique is effective in improving mortality prediction. We demonstrated how important time-window selection is and showed that enhancing it can save time and thus improve mortality prediction. These findings promise to improve the clinical team’s efficiency in prioritizing patients and giving greater attention to higher-risk patients.

To conclude, mortality prediction in the ICU can be improved if we consider alternative time windows instead of the 24-hour window, which is currently the most widely accepted among scoring systems today.

Background

The conduct of virtual physical examination has provided significant information for the diagnosis during a teleconsultation session, especially during the COVID-19 pandemic, where in-person physical examinations have been greatly compromised.

Objective

The aim of this scoping review was to provide a comprehensive overview of the available evidence concerning virtual physical examination (VPE) in all healthcare settings during the COVID-19 pandemic. The review focuses on types of VPE, technological and non-technological approaches, patient and clinician experiences, as well as barriers and facilitators of VPE.

Methods

A literature search was conducted across three databases, namely MEDLINE, Embase, and Scopus. Only studies in the English language with primary research data collected from December 2019 to January 2023 were included. A narrative analysis, highlighting patients’ and clinicians’ experiences, was conducted on the included studies. This scoping review was reported using The PRISMA extension for scoping reviews (PRISMA-ScR) Checklist.

Results

A total of 25 articles meeting eligibility criteria were identified. Three major types of VPE included were musculoskeletal, head and neck, and chest exams. Sixteen studies involved specific technological aids, while three studies involved non-technological aids. Patients found VPE helped them to better assess their disease conditions, or aided their clinicians’ understanding of their conditions. Clinicians also reported that VPE had provided enough clinically relevant information for decision-making in 2 neurological evaluations. Barriers to conducting VPE included technological challenges, efficacy concerns, confidence level of assistants, as well as patient health conditions, health literacy, safety, and privacy.

Conclusions

Patients found virtual physical examination (VPE) helpful in understanding their own conditions, and clinicians found it useful for better assessing patient’s conditions. From the clinicians’ point of view, VPE provided sufficient clinically relevant information for decision-making in neurological evaluations. Major barriers identified for VPE included technological issues, patient’s health conditions, and their health literacy.

Background

Mental health issues among healthcare workers remain a serious problem globally. Recent surveys continue to report high levels of depression, anxiety, burnout and other conditions amongst various occupational groups. Novel approaches are needed to support clinician well-being.

Objective

This integrative literature review aims to explore the current state of research examining the use of generative artificial intelligence (GenAI) and machine learning (ML) systems to predict mental health issues and identify associated risk factors amongst healthcare professionals.

Methods

A literature search of databases was conducted in Medline then adapted as necessary to Scopus, Web of Science, Google Scholar, PubMed and CINAHL with Full Text. Eleven studies met the inclusion criteria for the review.

Results

Nine studies employed various machine learning techniques to predict different mental health outcomes among healthcare workers. Models showed good predictive performance, with AUCs ranging from 0.82 to 0.904 for outcomes such as depression, anxiety and safety perceptions. Key risk factors identified included fatigue, stress, burnout, workload, sleep issues and lack of support. Two studies explored the potential of sensor-based technologies and GenAI analysis of physiological data.

None of the included studies focused on the use of GenAI systems specifically for providing mental health support to healthcare workers.

Conclusion

Preliminary research demonstrates that AI/ML models can effectively predict mental health issues. However, more work is needed to evaluate the real-world integration and impact of these tools, including GenAI systems, in identifying clinician distress and supporting well-being over time. Further research should aim to explore how GenAI may be developed and applied to provide mental health support for healthcare workers.

Background and Objective

Real-world data encompass population diversity, enabling insights into chronic disease mortality risk among the elderly. Deep learning excels on large datasets, offering promise for real-world data. However, current models focus on single diseases, neglecting comorbidities prevalent in patients. Moreover, mortality is infrequent compared to illness, causing extreme class imbalance that impedes reliable prediction. We aim to develop a deep learning framework that accurately forecasts mortality risk from real-world data by addressing comorbidities and class imbalance.

Methods

We integrated multi-task and cost-sensitive learning, developing an enhanced deep neural network architecture that extends multi-task learning to predict mortality risk across multiple chronic diseases. Each patient cohort with a chronic disease was assigned to a separate task, with shared lower-level parameters capturing inter-disease complexities through distinct top-level networks. Cost-sensitive functions were incorporated to ensure learning of positive class characteristics for each task and achieve accurate prediction of the risk of death from multiple chronic diseases.

Results

Our study covers 15 prevalent chronic diseases and is experimented with real-world data from 482,145 patients (including 9,516 deaths) in Shenzhen, China. The proposed model is compared with six models including three machine learning models: logistic regression, XGBoost, and CatBoost, and three state-of-the-art deep learning models: 1D-CNN, TabNet, and Saint. The experimental results show that, compared with the other compared algorithms, MTL-CSDNN has better prediction results on the test set (ACC=0.99, REC=0.99, PRAUC=0.97, MCC=0.98, G-means = 0.98).

Conclusions

Our method provides valuable insights into leveraging real-world data for precise multi-disease mortality risk prediction, offering potential applications in optimizing chronic disease management, enhancing well-being, and reducing healthcare costs for the elderly population.

Background

Google Trends is a widely used tool for infodemiological surveys. However, irregularities in the random sampling and aggregation algorithms compromise the reliability of the relative search volume (RSV) and the regional online interest (ROI).

Objective

The study aims to unmask methodological criticalities commonly ignored in carrying out infodemiological surveys via Google Trends. A guide to avoiding these shortcomings is also provided.

Material and methods

The Google Topic “Coronavirus disease 2019” has been investigated using different timelapses, categories, and IP addresses. The same samples were manually collected multiple times to evaluate the RSV and ROI stability. Stability was estimated through indicators of variability (e.g., coefficient of percentage variation “CV%” and its 4-surprisal interval “4-I”). The content aggregation capacity of the algorithms relating to topics and categories was evaluated through the quantitative analysis of RSV and ROI and the qualitative examination of the related queries.

Results

The stability of Google Trends’ RSV and ROI is not linked exclusively to the dataset dimension or the IP address. Subregional datasets can be highly unstable (e.g., CV% = 10, 4-I: [8], [13]). Google Trends categories and topics can exclude relevant queries or include unnecessary queries. The statistical scenario is consistent with the following hypotheses: i) datasets containing too few queries are highly unstable, ii) the “interest over time” data format is generally reliable for evaluating trends and correlations, iii) Google Trends improvements have altered the RSV historical trends.

Conclusions

Google Trends can be an effective and efficient infodemiological tool as long as the reliability of web search indexes is appropriately analyzed and weighted for the scientific goal. The methodological steps discussed in this study are critical to drawing valid and relevant scientific conclusions.

In the rapidly evolving landscape of information technologies, individuals and organizations must adapt to the digital age. Given the diversity in users’ knowledge and experience with technology, their acceptance levels also vary. Over the past 30 years, various theoretical models have been introduced to provide a framework for understanding user acceptance of technology. Among these, the Technology Acceptance Model (TAM) stands out as a key theoretical framework, offering insights into why new technologies are either accepted or rejected. Analyzing user acceptance of technology has thus become a critical area of study. Healthcare organizations aim to assess the perceived efficacy and user-friendliness of a given technology. This will help health organisations design and implement HIS that meet users’ needs and preferences. In this context, how does the TAM clarify the acceptance and use of Health Information Systems (HIS)? To address this inquiry, a comprehensive literature review will be carried out. The systematic review involved 29 studies issued between 2018 and 2023 and searched the databases Pubmed, Scopus, Wos and Ulakbim TR Index. The PRISMA flowchart was used to identify the included studies. According to the results, some variables stand out in the acceptance and utilisation of HIS. Among the users of HIS, it can be said that the results relating to nurses stand out. In particular, there are studies which emphasise that ’gender’ is a crucial factor in explaining the models. Another crucial finding of the current systematic review is the need to train users in the acceptance and use of HIS.

Background

Acute kidney injury (AKI) is associated with increased mortality in critically ill patients. Due to differences in the etiology and pathophysiological mechanism, the current AKI criteria put it an embarrassment to evaluate clinical therapy and prognosis.

Objective

We aimed to identify subphenotypes based on routinely collected clinical data to expose the unique pathophysiologic patterns.

Methods

A retrospective study was conducted based on the Medical Information Mart for Intensive Care IV (MIMIC-IV) and the eICU Collaborative Research Database (eICU-CRD), and a deep clustering approach was conducted to derive subphenotypes. We conducted further analysis to uncover the underlying clinical patterns and interpret the subphenotype derivation.

Results

We studied 14,189 and 19,382 patients with AKI within 48 h of ICU admission in the two datasets, respectively. Through our approach, we identified seven distinct AKI subphenotypes with mortality heterogeneity in each cohort. These subphenotypes displayed significant variations in demographics, comorbidities, levels of laboratory measurements, and survival patterns. Notably, the subphenotypes could not be effectively characterized using the Kidney Disease: Improving Global Outcomes (KDIGO) criteria alone. Therefore, we uncovered the unique underlying characteristics of each subphenotype through model-based interpretation. To assess the usability of the subphenotypes, we conducted an evaluation, which yielded a micro-Area Under the Receiver Operating Characteristic (AUROC) of 0.81 in the single-center cohort and 0.83 in the multi-center cohort within 48-hour of admission.

Conclusion

We derived highly characteristic, interpretable, and usable AKI subphenotypes that exhibited superior prognostic values.

Background

Involving users has become a prominent principle in the development of Health Information Technologies (HIT) and has led to an uprise in agile and cocreation methods. Previous literature shows how the two can be combined in one method, but also suggest that using such a method may come with challenges, for which the solutions are unclear.

Purpose

To identify the challenges of using a method that combines agile and cocreation, provide solutions for these challenges, and evaluate its usage.

Methods and materials

The setting for this research was the Agile Cocreation of Robots for Aging (ACCRA) project. The research consisted of three phases: 1) evaluating the Agile Cocreation method to identify challenges in its usage, by analysing documents from the project (e-mails, meeting notes), 2) solving the challenges to improve the method, by designing solutions in a cocreation session; and 3) evaluating the usage of the improved version via a survey among engineers and user researchers involved in the project.

Results

We identified three main challenges and developed three solutions, which were used in the next phase of the project. First, to engage all stakeholders in cocreation, we implemented more fun and playful materials. Second, to bridge the differences between engineers and user researchers we invested in face-to-face meetings. Third, to manage knowledge in the project we intensified our meeting schedule to weekly meetings. In the quantitative evaluation of the improved cocreation method, the engineers and user researchers were positive about the agile cocreation method and about our improvements.

Conclusion

When developing HIT, a method that combines agile and cocreation is useful because it helps to identify user needs and to translate these needs into technology. To identify the needs of these users and other stakeholders it is important to involve them as active partners in cocreation using fun and playful materials. Engineers and user researchers should bridge their differences and meet face-to-face as much as possible.