Building Bridges for Federated Learning in Healthcare: Review on Approaches for Common Data Model Development.

Common data models provide a standardized way to represent data used in federated learning tasks. The aim of this review was to explore the development and use of common data models to harmonize electronic health record data in health research. The data search yielded 724 records, of which 19 were included for this study. None of the research focused on nursing specific topics. All studies either utilized the Observational Medical Outcomes Partnership (OMOP) common data model, or developed a model partly based on the OMOP. A roadmap to guide research for the development of common data models for federated learning are warranted.

Improving Quality of ICD-10 (International Statistical Classification of Diseases, Tenth Revision) Coding Using AI: Protocol for a Crossover Randomized Controlled Trial.

BACKGROUND: Computer-assisted clinical coding (CAC) tools are designed to help clinical coders assign standardized codes, such as the ICD-10 (International Statistical Classification of Diseases, Tenth Revision), to clinical texts, such as discharge summaries. Maintaining the integrity of these standardized codes is important both for the functioning of health systems and for ensuring data used for secondary purposes are of high quality. Clinical coding is an error-prone cumbersome task, and the complexity of modern classification systems such as the ICD-11 (International Classification of Diseases, Eleventh Revision) presents significant barriers to implementation. To date, there have only been a few user studies; therefore, our understanding is still limited regarding the role CAC systems can play in reducing the burden of coding and improving the overall quality of coding. OBJECTIVE: The objective of the user study is to generate both qualitative and quantitative data for measuring the usefulness of a CAC system, Easy-ICD, that was developed for recommending ICD-10 codes. Specifically, our goal is to assess whether our tool can reduce the burden on clinical coders and also improve coding quality. METHODS: The user study is based on a crossover randomized controlled trial study design, where we measure the performance of clinical coders when they use our CAC tool versus when they do not. Performance is measured by the time it takes them to assign codes to both simple and complex clinical texts as well as the coding quality, that is, the accuracy of code assignment. RESULTS: We expect the study to provide us with a measurement of the effectiveness of the CAC system compared to manual coding processes, both in terms of time use and coding quality. Positive outcomes from this study will imply that CAC tools hold the potential to reduce the burden on health care staff and will have major implications for the adoption of artificial intelligence-based CAC innovations to improve coding practice. Expected results to be published summer 2024. CONCLUSIONS: The planned user study promises a greater understanding of the impact CAC systems might have on clinical coding in real-life settings, especially with regard to coding time and quality. Further, the study may add new insights on how to meaningfully exploit current clinical text mining capabilities, with a view to reducing the burden on clinical coders, thus lowering the barriers and paving a more sustainable path to the adoption of modern coding systems, such as the new ICD-11. TRIAL REGISTRATION: clinicaltrials.gov NCT06286865; https://clinicaltrials.gov/study/NCT06286865. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/54593.

A multinational study on artificial intelligence adoption: Clinical implementers' perspectives.

BACKGROUND: Despite substantial progress in AI research for healthcare, translating research achievements to AI systems in clinical settings is challenging and, in many cases, unsatisfactory. As a result, many AI investments have stalled at the prototype level, never reaching clinical settings. OBJECTIVE: To improve the chances of future AI implementation projects succeeding, we analyzed the experiences of clinical AI system implementers to better understand the challenges and success factors in their implementations. METHODS: Thirty-seven implementers of clinical AI from European and North and South American countries were interviewed. Semi-structured interviews were transcribed and analyzed qualitatively with the framework method, identifying the success factors and the reasons for challenges as well as documenting proposals from implementers to improve AI adoption in clinical settings. RESULTS: We gathered the implementers' requirements for facilitating AI adoption in the clinical setting. The main findings include 1) the lesser importance of AI explainability in favor of proper clinical validation studies, 2) the need to actively involve clinical practitioners, and not only clinical researchers, in the inception of AI research projects, 3) the need for better information structures and processes to manage data access and the ethical approval of AI projects, 4) the need for better support for regulatory compliance and avoidance of duplications in data management approval bodies, 5) the need to increase both clinicians' and citizens' literacy as respects the benefits and limitations of AI, and 6) the need for better funding schemes to support the implementation, embedding, and validation of AI in the clinical workflow, beyond pilots. CONCLUSION: Participants in the interviews are positive about the future of AI in clinical settings. At the same time, they proposenumerous measures to transfer research advancesinto implementations that will benefit healthcare personnel. Transferring AI research into benefits for healthcare workers and patients requires adjustments in regulations, data access procedures, education, funding schemes, and validation of AI systems.

Using a Large Open Clinical Corpus for Improved ICD-10 Diagnosis Coding.

With the recent advances in natural language processing and deep learning, the development of tools that can assist medical coders in ICD-10 diagnosis coding and increase their efficiency in coding discharge summaries is significantly more viable than before. To that end, one important component in the development of these models is the datasets used to train them. In this study, such datasets are presented, and it is shown that one of them can be used to develop a BERT-based language model that can consistently perform well in assigning ICD-10 codes to discharge summaries written in Swedish. Most importantly, it can be used in a coding support setup where a tool can recommend potential codes to the coders. This reduces the range of potential codes to consider and, in turn, reduces the workload of the coder. Moreover, the de-identified and pseudonymised dataset is open to use for academic users.

De-identifying Norwegian Clinical Text using Resources from Swedish and Danish.

The lack of relevant annotated datasets represents one key limitation in the application of Natural Language Processing techniques in a broad number of tasks, among them Protected Health Information (PHI) identification in Norwegian clinical text. In this work, the possibility of exploiting resources from Swedish, a very closely related language, to Norwegian is explored. The Swedish dataset is annotated with PHI information. Different processing and text augmentation techniques are evaluated, along with their impact in the final performance of the model. The augmentation techniques, such as injection and generation of both Norwegian and Scandinavian Named Entities into the Swedish training corpus, showed to increase the performance in the de-identification task for both Danish and Norwegian text. This trend was also confirmed by the evaluation of model performance on a sample Norwegian gastro surgical clinical text.

Artificial Intelligence Implementation in Healthcare: A Theory-Based Scoping Review of Barriers and Facilitators.

There is a large proliferation of complex data-driven artificial intelligence (AI) applications in many aspects of our daily lives, but their implementation in healthcare is still limited. This scoping review takes a theoretical approach to examine the barriers and facilitators based on empirical data from existing implementations. We searched the major databases of relevant scientific publications for articles related to AI in clinical settings, published between 2015 and 2021. Based on the theoretical constructs of the Consolidated Framework for Implementation Research (CFIR), we used a deductive, followed by an inductive, approach to extract facilitators and barriers. After screening 2784 studies, 19 studies were included in this review. Most of the cited facilitators were related to engagement with and management of the implementation process, while the most cited barriers dealt with the intervention's generalizability and interoperability with existing systems, as well as the inner settings' data quality and availability. We noted per-study imbalances related to the reporting of the theoretic domains. Our findings suggest a greater need for implementation science expertise in AI implementation projects, to improve both the implementation process and the quality of scientific reporting.

A Privacy-Preserving Audit and Feedback System for the Antibiotic Prescribing of General Practitioners: Survey Study.

BACKGROUND: Antibiotic resistance is a worldwide public health problem that is accelerated by the misuse and overuse of antibiotics. Studies have shown that audits and feedback enable clinicians to compare their personal clinical performance with that of their peers and are effective in reducing the inappropriate prescribing of antibiotics. However, privacy concerns make audits and feedback hard to implement in clinical settings. To solve this problem, we developed a privacy-preserving audit and feedback (A&F) system. OBJECTIVE: This study aims to evaluate a privacy-preserving A&F system in clinical settings. METHODS: A privacy-preserving A&F system was deployed at three primary care practices in Norway to generate feedback for 20 general practitioners (GPs) on their prescribing of antibiotics for selected respiratory tract infections. The GPs were asked to participate in a survey shortly after using the system. RESULTS: A total of 14 GPs responded to the questionnaire, representing a 70% (14/20) response rate. The participants were generally satisfied with the usefulness of the feedback and the comparisons with peers, as well as the protection of privacy. The majority of the GPs (9/14, 64%) valued the protection of their own privacy as well as that of their patients. CONCLUSIONS: The system overcomes important privacy and scaling challenges that are commonly associated with the secondary use of electronic health record data and has the potential to improve antibiotic prescribing behavior; however, further study is required to assess its actual effect.

Clinical Notes De-Identification: Scoping Recent Benchmarks for n2c2 Datasets.

Publicly shared repositories play an important role in advancing performance benchmarks for some of the most important tasks in natural language processing (NLP) and healthcare in general. This study reviews most recent benchmarks based on the 2014 n2c2 de-identification dataset. Pre-processing challenges were uncovered, and attention brought to the discrepancies in reported number of Protected Health Information (PHI) entities among the studies. Improved reporting is required for greater transparency and reproducibility.

De-Identifying Swedish EHR Text Using Public Resources in the General Domain.

Sensitive data is normally required to develop rule-based or train machine learning-based models for de-identifying electronic health record (EHR) clinical notes; and this presents important problems for patient privacy. In this study, we add non-sensitive public datasets to EHR training data; (i) scientific medical text and (ii) Wikipedia word vectors. The data, all in Swedish, is used to train a deep learning model using recurrent neural networks. Tests on pseudonymized Swedish EHR clinical notes showed improved precision and recall from 55.62% and 80.02% with the base EHR embedding layer, to 85.01% and 87.15% when Wikipedia word vectors are added. These results suggest that non-sensitive text from the general domain can be used to train robust models for de-identifying Swedish clinical text; and this could be useful in cases where the data is both sensitive and in low-resource languages.

Play and Learn: Developing a Social Game for Children with Diabetes.

While serious games in healthcare have gained much attention in recent years, the pedagogical, social or behavioural frameworks tied to the game elements are still poorly understood. We report the prototyping effort as work-in-progress for a serious social gaming framework for children with diabetes. Motivation theories were combined with child education literature to evaluate potential elements of the framework. Based on the evaluation, we designed cooperation, social comparison and focus on positive achievements as core game elements, and limited the extent of competition. Examining the theoretical foundations that underpin different elements of serious games promises a greater understanding of effective gaming techniques for healthcare.

Serious Game Co-Design for Children with Type 1 Diabetes.

Co-design or participatory design has emerged as a useful concept where stakeholders and end-users have a greater stake in designing the end product. To date, few accounts exist of the use of the concept in serious game design, especially for children with chronic diseases. We report initial steps in serious game co-design for children with type 1 diabetes. Participants included 14 children (mean age 8.6 years, range of 4-13) who were invited to sketch a diabetes game. The most prevalent themes that emerged from the sketches (N=17) include blood glucose monitoring (n=12), nutrition (n=8) and insulin (n=8); all of which are consistent with diabetes education guidelines. Co-design is a promising concept for understanding children's world-view when designing healthcare games.

Effectiveness of an Internet Community for Severely Obese Women.

While Internet communities have become thriving sources of support, little is yet known about their effectiveness. We retrospectively sampled morbidly obese (Body Mass Index, BMI > 40) women who were active for at least a year in an Internet community. We compared self-reported weight changes between women who had high online participation levels (n = 71) versus those with low participation levels as control (n = 69). Women who actively participated online lost on average 7.52%, while those who were passive lost 5.39% of their original body weight. For active women, there was positive, albeit weak, correlation (r = 0.22, p < 0.05) between online participation levels and weight loss, while no significant correlation was noted for the control. Current results indicate modest evidence supporting active participation in Internet groups as an effective weight loss strategy for the target group.

Persuasive attributes of medication adherence interventions for older adults: a systematic review.

BACKGROUND: Low adherence to prescribed medications leads to serious negative health consequences in older adults. Effective interventions that improve adherence are often labor-intensive and complex. However, most studies do not analyze the separate effects of the components. OBJECTIVE: Persuasive System Design (PSD) is framework that analyzes the motivations that change behavior. In this paper, we aim to apply the model to changing the pill-taking behaviors of the aging population and determine which persuasive elements in interventions drive improvement in medication adherence. METHODS: Systematic review using the databases Medline (1977 to February 2012), Cochrane library (2000 to June 2013); Cinahl (1975 to June 2013), and Psycinfo (2002 to June 2012). Inclusion criteria were experimental trials with participants' mean age ⩾ 60 years and had medication adherence as a primary or secondary measure. RESULTS: Meta-analysis (40 studies) demonstrated a significant association of tailoring, or one-on-one counseling, with medication adherence. Interventions with simulation (showing the causal relationship between non-adherence and negative effects) and rehearsal (miming medication-taking behavior) also showed evidence for improved adherence. CONCLUSIONS: Future medication adherence interventions might be more effective if they were based on persuasive technology.

Designing a diabetes mobile application with social network support.

Although mobile applications and social media have emerged as important facets of the Internet, their role in healthcare is still not well-understood. We present design artefacts, inspired by persuasive technology concepts, from a study of social media as part of a diabetes mHealth application. We used the design science approach for mobile application design, and real-life user testing and focus group meetings to test the application over a 12-week period with 7 participants. Based on the System Usability Score (SUS), the mobile application scored an average of 84.6 (SD=13.2), which represents a fairly high usability score compared to the literature. Regression analysis on the daily blood glucose levels showed significant decreases for some patients, and although the study is not powered, the HbA1c showed a promising trend, and self-efficacy marginally increased. Incorporating persuasive elements such as blood glucose tracking and visualisation, and social media access directly from the mobile application produced promising results that warrant a larger study of behaviour change for people with diabetes.

Surgical telementoring in knowledge translation--clinical outcomes and educational benefits: a comprehensive review.

BACKGROUND: Surgical telementoring has been reported for decades. However, there exists limited evidence of clinical outcome and educational benefits. OBJECTIVE: To perform a comprehensive review of surgical telementoring surveys published in the past 2 decades. RESULTS: Of 624 primary identified articles, 34 articles were reviewed. A total of 433 surgical procedures were performed by 180 surgeons. Most common telementored procedures were laparoscopic cholecystectomy (57 cases, 13%), endovascular treatment of aortic aneurysm (48 cases, 11%), laparoscopic colectomy (32 cases, 7%), and nefrectomies (41 cases, 9%). In all, 167 (38%) cases had a laparoscopic approach, and 8 cases (5%) were converted to open surgery. Overall, 20 complications (5%) were reported (liver bleeding, trocar port bleeding, bile collection, postoperative ileus, wound infection, serosa tears, iliac artery rupture, conversion open surgery). Eight surveys (23%) have structured assessment of educational outcomes. Telementoring was combined with simulators (n = 2) and robotics (n = 3). Twelve surveys (35%) were intercontinental. Technology satisfaction was high among 83% of surgeons. CONCLUSION: Few surveys have a structured assessment of educational outcome. Telementoring has improved impact on surgical education. Reported complication rate was 5%.

Mobile health applications to assist patients with diabetes: lessons learned and design implications.

Self-management is critical to achieving diabetes treatment goals. Mobile phones and Bluetooth® can supportself-management and lifestyle changes for chronic diseases such as diabetes. A mobile health (mHealth) research platform--the Few Touch Application (FTA)--is a tool designed to support the self-management of diabetes. The FTA consists of a mobile phone-based diabetes diary, which can be updated both manually from user input and automatically by wireless data transfer, and which provides personalized decision support for the achievement of personal health goals. Studies and applications (apps) based on FTAs have included: (1) automatic transfer of blood glucose (BG) data; (2) short message service (SMS)-based education for type 1diabetes (T1DM); (3) a diabetes diary for type 2 diabetes (T2DM); (4) integrating a patient diabetes diary with health care (HC) providers; (5) a diabetes diary for T1DM; (6) a food picture diary for T1DM; (7) physical activity monitoring for T2DM; (8) nutrition information for T2DM; (9) context sensitivity in mobile self-help tools; and (10) modeling of BG using mobile phones. We have analyzed the performance of these 10 FTA-based apps to identify lessons for designing the most effective mHealth apps. From each of the 10 apps of FTA, respectively, we conclude: (1) automatic BG data transfer is easy to use and provides reassurance; (2) SMS-based education facilitates parent-child communication in T1DM; (3) the T2DM mobile phone diary encourages reflection; (4) the mobile phone diary enhances discussion between patients and HC professionals; (5) the T1DM mobile phone diary is useful and motivational; (6) the T1DM mobile phone picture diary is useful in identifying treatment obstacles; (7) the step counter with automatic data transfer promotes motivation and increases physical activity in T2DM; (8) food information on a phone for T2DM should not be at a detailed level; (9) context sensitivity has good prospects and is possible to implement on today's phones; and (10) BG modeling on mobile phones is promising for motivated T1DM users. We expect that the following elements will be important in future FTA designs: (A) automatic data transfer when possible; (B) motivational and visual user interfaces; (C) apps with considerable health benefits in relation to the effort required; (D) dynamic usage, e.g., both personal and together with HC personnel, long-/short-term perspective; and (E) inclusion of context sensitivity in apps. We conclude that mHealth apps will empower patients to take a more active role in managing their own health.

Exploring the community structure of a diabetes forum.

A health forum is a kind of social network where users share information for specific topics they create. The purpose of this study was the identification of the key actors and the user communities in such a network. We used the publicly available data from a diabetes forum to create the corresponding network and explore several algorithms for the detection of user communities. The degree centrality of the network followed the power law distribution demonstrating that only a few users were the key actors in the forum. It was also shown that it is feasible to infer the top communities from a forum using certain algorithms; the key actors participated in these communities. Our approach could be applied to other health forums and be extended to examine additional aspects.

Features of mobile diabetes applications: review of the literature and analysis of current applications compared against evidence-based guidelines.

BACKGROUND: Interest in mobile health (mHealth) applications for self-management of diabetes is growing. In July 2009, we found 60 diabetes applications on iTunes for iPhone; by February 2011 the number had increased by more than 400% to 260. Other mobile platforms reflect a similar trend. Despite the growth, research on both the design and the use of diabetes mHealth applications is scarce. Furthermore, the potential influence of social media on diabetes mHealth applications is largely unexplored. OBJECTIVE: Our objective was to study the salient features of mobile applications for diabetes care, in contrast to clinical guideline recommendations for diabetes self-management. These clinical guidelines are published by health authorities or associations such as the National Institute for Health and Clinical Excellence in the United Kingdom and the American Diabetes Association. METHODS: We searched online vendor markets (online stores for Apple iPhone, Google Android, BlackBerry, and Nokia Symbian), journal databases, and gray literature related to diabetes mobile applications. We included applications that featured a component for self-monitoring of blood glucose and excluded applications without English-language user interfaces, as well as those intended exclusively for health care professionals. We surveyed the following features: (1) self-monitoring: (1.1) blood glucose, (1.2) weight, (1.3) physical activity, (1.4) diet, (1.5) insulin and medication, and (1.6) blood pressure, (2) education, (3) disease-related alerts and reminders, (4) integration of social media functions, (5) disease-related data export and communication, and (6) synchronization with personal health record (PHR) systems or patient portals. We then contrasted the prevalence of these features with guideline recommendations. RESULTS: The search resulted in 973 matches, of which 137 met the selection criteria. The four most prevalent features of the applications available on the online markets (n = 101) were (1) insulin and medication recording, 63 (62%), (2) data export and communication, 61 (60%), (3) diet recording, 47 (47%), and (4) weight management, 43 (43%). From the literature search (n = 26), the most prevalent features were (1) PHR or Web server synchronization, 18 (69%), (2) insulin and medication recording, 17 (65%), (3) diet recording, 17 (65%), and (4) data export and communication, 16 (62%). Interestingly, although clinical guidelines widely refer to the importance of education, this is missing from the top functionalities in both cases. CONCLUSIONS: While a wide selection of mobile applications seems to be available for people with diabetes, this study shows there are obvious gaps between the evidence-based recommendations and the functionality used in study interventions or found in online markets. Current results confirm personalized education as an underrepresented feature in diabetes mobile applications. We found no studies evaluating social media concepts in diabetes self-management on mobile devices, and its potential remains largely unexplored.

Mobile peer support in diabetes.

As in other domains, there has been unprecedented growth in diabetesrelated social media in the past decade. Although there is not yet enough evidence for the clinical benefits of patient-to-patient dialogue using emergent social media, patient empowerment through easier access to information has been proven to foster healthy lifestyles, and to delay or even prevent progression of secondary illnesses. In the design of diabetes-related social media, we need access to personal health data for modelling the core disease-related characteristics of the user. We discuss design aspects of mobile peer support, including acquisition of personal health data, and design artefacts for a healthcare recommender system. We also explore mentoring models as a tool for managing the transient relationships among peers with diabetes. Intermediate results suggest acquiring health data for modelling patients' health status is feasible for implementing a personalized and mobile peer-support system.