-
NPJ Digital Medicine Apr 2024The development of diagnostic tools for skin cancer based on artificial intelligence (AI) is increasing rapidly and will likely soon be widely implemented in clinical... (Review)
Review
The development of diagnostic tools for skin cancer based on artificial intelligence (AI) is increasing rapidly and will likely soon be widely implemented in clinical use. Even though the performance of these algorithms is promising in theory, there is limited evidence on the impact of AI assistance on human diagnostic decisions. Therefore, the aim of this systematic review and meta-analysis was to study the effect of AI assistance on the accuracy of skin cancer diagnosis. We searched PubMed, Embase, IEE Xplore, Scopus and conference proceedings for articles from 1/1/2017 to 11/8/2022. We included studies comparing the performance of clinicians diagnosing at least one skin cancer with and without deep learning-based AI assistance. Summary estimates of sensitivity and specificity of diagnostic accuracy with versus without AI assistance were computed using a bivariate random effects model. We identified 2983 studies, of which ten were eligible for meta-analysis. For clinicians without AI assistance, pooled sensitivity was 74.8% (95% CI 68.6-80.1) and specificity was 81.5% (95% CI 73.9-87.3). For AI-assisted clinicians, the overall sensitivity was 81.1% (95% CI 74.4-86.5) and specificity was 86.1% (95% CI 79.2-90.9). AI benefitted medical professionals of all experience levels in subgroup analyses, with the largest improvement among non-dermatologists. No publication bias was detected, and sensitivity analysis revealed that the findings were robust. AI in the hands of clinicians has the potential to improve diagnostic accuracy in skin cancer diagnosis. Given that most studies were conducted in experimental settings, we encourage future studies to further investigate these potential benefits in real-life settings.
PubMed: 38594408
DOI: 10.1038/s41746-024-01031-w -
Digital Health 2023Substance use disorders affect 36 million people globally, but only a small proportion of them receive the necessary treatment. E-health interventions have been... (Review)
Review
BACKGROUND
Substance use disorders affect 36 million people globally, but only a small proportion of them receive the necessary treatment. E-health interventions have been developed to address this issue by improving access to substance use treatment. However, concerns about participant engagement and adherence to these interventions remain. This review aimed to evaluate adherence to e-health interventions targeting substance use and identify hypothesized predictors of adherence.
METHODS
A systematic review of literature published between 2009 and 2020 was conducted, and data on adherence measures and hypothesized predictors were extracted. Meta-analysis and meta-regression were used to analyze the data. The two adherence measures were (a) the mean proportion of modules completed across the intervention groups and (b) the proportion of participants that completed all modules. Four meta-regression models assessed each covariate including guidance, blended treatment, intervention duration and recruitment strategy.
RESULTS
The overall pooled adherence rate was 0.60 (95%-CI: 0.52-0.67) for the mean proportion of modules completed across 30 intervention arms and 0.47 (95%-CI: 0.35-0.59) for the proportion of participants that completed all modules across 9 intervention arms. Guidance, blended treatment, and recruitment were significant predictors of adherence, while treatment duration was not.
CONCLUSION
The study suggests that more research is needed to identify predictors of adherence, in order to determine specific aspects that contribute to better exposure to intervention content. Reporting adherence and predictors in future studies can lead to improved meta-analyses and the development of more engaging interventions. Identifying predictors can aid in designing effective interventions for substance use disorders, with important implications for e-health interventions targeting substance use.
PubMed: 37780062
DOI: 10.1177/20552076231203876 -
Frontiers in Digital Health 2023Computer-mediated care is becoming increasingly popular, but little research has been done on it and its effects on emotion-related outcomes. This systematic literature... (Review)
Review
INTRODUCTION
Computer-mediated care is becoming increasingly popular, but little research has been done on it and its effects on emotion-related outcomes. This systematic literature review aims to create an overview that addresses the research question: "Is there a relationship between computer-mediated care and emotional expression, perception, and emotional and (long-term) emotion outcomes?"
METHOD
This systematic literature review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and used five eligibility criteria, namely, (1) participants: adults seeking support; (2) intervention: eHealth; (3) diagnostic criteria: transdiagnostic concept of difficulty identifying, expressing, and/or regulating emotions (e.g., alexithymia); (4) comparator: either face-to-face care or no comparator; and (5) study design: quantitative studies or qualitative studies. Quality was assessed using the QualSyst tool.
RESULTS
The analysis includes 25 research papers. Self-paced interventions appear to have a positive effect on emotion regulation. Videoconferencing interventions improved emotion regulation from before to after treatment but worsened emotion regulation compared with face-to-face treatment.
DISCUSSION
The lack of variation in the modalities studied and the emotion measurements used make it difficult to draw responsible conclusions. Future research should examine how different modalities affect the real-time communication of emotions and how non-verbal cues influence this.
PubMed: 37720162
DOI: 10.3389/fdgth.2023.1216268 -
Digital Health 2023Augmented reality (AR) is a relatively new technology that merges virtual and physical environments, augmenting one's perception of reality. AR creates a... (Review)
Review
OBJECTIVE
Augmented reality (AR) is a relatively new technology that merges virtual and physical environments, augmenting one's perception of reality. AR creates a computer-generated environment that evokes a unique perception of reality, where real and virtual objects are registered with one another, which operates interactively and in real time. Recently, the medical application of AR technology has dramatically increased with other assisted technologies, from training to clinical practice. The ability to manipulate the real environment extensively has given AR interventions an advantage over traditional approaches. In this study, we aim to conduct a systematic review of the use of AR to have a better understanding of how the use of AR may affect patients with mental health-related conditions when combined with gamification.
METHOD
This systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines by searching Pubmed and Web of Science databases.
RESULTS AND CONCLUSION
We identified 48 relevant studies that fulfill the criteria. The studies were grouped into four categories: Neurodevelopmental disorders, anxiety and phobia, psychoeducation & well-being, and procedural & pain management. Our results revealed the effectiveness of AR in mental health-related conditions. However, the heterogeneity and small sample sizes demonstrate the need for further research with larger sample sizes and high-quality study designs.
PubMed: 37791140
DOI: 10.1177/20552076231203649 -
NPJ Digital Medicine Nov 2023Machine learning and deep learning are two subsets of artificial intelligence that involve teaching computers to learn and make decisions from any sort of data. Most... (Review)
Review
Machine learning and deep learning are two subsets of artificial intelligence that involve teaching computers to learn and make decisions from any sort of data. Most recent developments in artificial intelligence are coming from deep learning, which has proven revolutionary in almost all fields, from computer vision to health sciences. The effects of deep learning in medicine have changed the conventional ways of clinical application significantly. Although some sub-fields of medicine, such as pediatrics, have been relatively slow in receiving the critical benefits of deep learning, related research in pediatrics has started to accumulate to a significant level, too. Hence, in this paper, we review recently developed machine learning and deep learning-based solutions for neonatology applications. We systematically evaluate the roles of both classical machine learning and deep learning in neonatology applications, define the methodologies, including algorithmic developments, and describe the remaining challenges in the assessment of neonatal diseases by using PRISMA 2020 guidelines. To date, the primary areas of focus in neonatology regarding AI applications have included survival analysis, neuroimaging, analysis of vital parameters and biosignals, and retinopathy of prematurity diagnosis. We have categorically summarized 106 research articles from 1996 to 2022 and discussed their pros and cons, respectively. In this systematic review, we aimed to further enhance the comprehensiveness of the study. We also discuss possible directions for new AI models and the future of neonatology with the rising power of AI, suggesting roadmaps for the integration of AI into neonatal intensive care units.
PubMed: 38012349
DOI: 10.1038/s41746-023-00941-5 -
Digital Health 2023Cystic fibrosis causes mucus to build up in the lungs, digestive tract, and other areas. It is the most common chronic lung disease in children and young adults. It... (Review)
Review
BACKGROUND
Cystic fibrosis causes mucus to build up in the lungs, digestive tract, and other areas. It is the most common chronic lung disease in children and young adults. It requires daily medical care. Before the COVID-19 pandemic, telerehabilitation and telehealth were used, but it was after this that there was a boom in these types of assistance in order to continue caring for cystic fibrosis patients.
OBJECTIVE
The objective is to evaluate the effect of telemedicine programs in people with cystic fibrosis.
METHODS
For the search, the PubMed, Scopus, Web of Science, PEDro, Cochrane, and CINAHL databases were used. Randomized controlled trials, pilot studies, and clinical trials have been included. The exclusion criteria have considered that the population did not have another active disease or that telemedicine was not used as the main intervention. This study follows the PRISMA statement and has been registered in the PROSPERO database (CRD42021257647).
RESULTS
A total of 11 articles have been included in the systematic review. No improvements have been found in quality of life, forced expiratory volume, and forced vital capacity. Good results have been found in increasing physical activity and early detection of exacerbations. Adherence and satisfaction are very positive and promising.
CONCLUSIONS
Despite not obtaining significant improvements in some of the variables, it should be noted that the adherence and satisfaction of both patients and workers reinforce the use of this type of care. Future studies are recommended in which to continue investigating this topic.
PubMed: 37654722
DOI: 10.1177/20552076231197023 -
Cureus Jan 2024Dupuytren's disease (DD) is a fibroproliferative disorder that manifests as an abnormal growth of myofibroblasts, causing nodule formation and contractures and affecting... (Review)
Review
Comparing Complications and Patient Satisfaction Following Injectable Collagenase Versus Limited Fasciectomy for Dupuytren's Disease: A Systematic Review and Meta-Analysis.
Dupuytren's disease (DD) is a fibroproliferative disorder that manifests as an abnormal growth of myofibroblasts, causing nodule formation and contractures and affecting digit function. If left untreated, these contractures can lead to a loss of mobility and potentially impact hand function. This systematic review critically compares and evaluates the existing literature on the complications and patient satisfaction following injectable collagenase (CCH) versus limited fasciectomy (LF) for DD. We performed a comprehensive search of the PubMed, Medical Literature Analysis and Retrieval System Online (MEDLINE), The Cochrane Library, and Excerpta Medica database (EMBASE) databases from 2006 to August 2023. This research targeted all clinical studies involving adults who underwent injectable collagenase and/or limited fasciectomy in the management of DD. Out of the 437 identified studies, only 53 were considered eligible for our analysis, and merely 14 met our inclusion criteria. These selected studies encompassed a total of 967 patients with 1,344 treated joints, with an average follow-up duration of 19.22 (ranging from one to 84.06) months. Within this cohort, 498 joints from 385 patients underwent LF, while 846 joints from 491 patients received CCH injections. Notably, among the 491 patients treated with CCH, 1,060 complications were reported, averaging 2.15 complications per patient, with the most common being contusion/bruising/hematoma/ecchymosis (22.54%), and edema/swelling (18.96%). In contrast, among the 385 patients treated with LF, only 97 complications were reported, translating to 0.25 complications per patient, with the most frequent being paraesthesia or numbness (23.7%), scar sequelae like skin laceration, tear, fissure, or hypertrophic scar (23.7%), and neuropraxia or nerve injury (22.6%). Our meta-analysis indicates that paraesthesia or numbness is more frequently observed in LF than CCH injections, although without statistical significance, with a risk ratio (RR) of 0.39 (95% confidence interval (CI) 0.13-1.18, p-value 0.1). However, scar sequelae (hypertrophic scar, skin laceration, tear, or fissure) show a contrasting pattern, being more commonly associated with CCH injections than LF, with an RR of 1.98 (95% CI 0.26-14.85, p-value 0.51), which, upon eliminating the source of heterogeneity, becomes statistically significant, with an RR of 4.98 (95% CI 1.40-17.72, p-value 0.01). Our data revealed a higher frequency of complications with CCH compared to LF, although more severe adverse effects were observed in the LF group, such as neuropraxia or nerve injury. Scar sequelae were more common with CCH injections. Despite both treatments showing increased patient satisfaction at the final follow-up, CCH injection resulted in earlier improvements in satisfaction.
PubMed: 38420076
DOI: 10.7759/cureus.53147 -
Digital Health 2023During the Coronavirus Disease 2019 (COVID-19) pandemic, digital health technologies (DHTs) became increasingly important, especially for older adults. The objective of... (Review)
Review
OBJECTIVE
During the Coronavirus Disease 2019 (COVID-19) pandemic, digital health technologies (DHTs) became increasingly important, especially for older adults. The objective of this systematic review was to synthesize evidence on the rapid implementation and use of DHTs among older adults during the COVID-19 pandemic.
METHODS
A structured, electronic search was conducted on 9 November 2021, and updated on 5 January 2023, among five databases to select DHT interventional studies conducted among older adults during the pandemic. The bias of studies was assessed using Version 2 of the Cochrane Risk-of-Bias Tool for randomized trials (RoB 2) and Risk of Bias in Non-randomized Studies of Interventions (ROBINS-I).
RESULTS
Among 20 articles included in the review, 14 (70%) focused on older adults with chronic diseases or symptoms, such as dementia or cognitive impairment, type 2 diabetes, and obesity. DHTs included traditional telehealth interventions via telephone, video, and social media, as well as emerging technologies such as Humanoid Robot and Laser acupuncture teletherapy. Using RoB 2 and ROBINS-I, four studies (20%) were evaluated as high or serious overall risk of bias. DHTs have shown to be effective, feasible, acceptable, and satisfactory for older adults during the COVID-19 pandemic compared to usual care. In addition, some studies also highlighted challenges with technology, hearing difficulties, and communication barriers within the vulnerable population.
CONCLUSIONS
During the COVID-19 pandemic, DHTs had the potential to improve various health outcomes and showed benefits for older adults' access to health care services.
PubMed: 37529545
DOI: 10.1177/20552076231191050 -
NPJ Digital Medicine Apr 2024The integration of robotics in surgery has increased over the past decade, and advances in the autonomous capabilities of surgical robots have paralleled that of... (Review)
Review
The integration of robotics in surgery has increased over the past decade, and advances in the autonomous capabilities of surgical robots have paralleled that of assistive and industrial robots. However, classification and regulatory frameworks have not kept pace with the increasing autonomy of surgical robots. There is a need to modernize our classification to understand technological trends and prepare to regulate and streamline surgical practice around these robotic systems. We present a systematic review of all surgical robots cleared by the United States Food and Drug Administration (FDA) from 2015 to 2023, utilizing a classification system that we call Levels of Autonomy in Surgical Robotics (LASR) to categorize each robot's decision-making and action-taking abilities from Level 1 (Robot Assistance) to Level 5 (Full Autonomy). We searched the 510(k), De Novo, and AccessGUDID databases in December 2023 and included all medical devices fitting our definition of a surgical robot. 37,981 records were screened to identify 49 surgical robots. Most surgical robots were at Level 1 (86%) and some reached Level 3 (Conditional Autonomy) (6%). 2 surgical robots were recognized by the FDA to have machine learning-enabled capabilities, while more were reported to have these capabilities in their marketing materials. Most surgical robots were introduced via the 510(k) pathway, but a growing number via the De Novo pathway. This review highlights trends toward greater autonomy in surgical robotics. Implementing regulatory frameworks that acknowledge varying levels of autonomy in surgical robots may help ensure their safe and effective integration into surgical practice.
PubMed: 38671232
DOI: 10.1038/s41746-024-01102-y -
The Lancet. Digital Health Feb 2024Machine learning (ML)-based risk prediction models hold the potential to support the health-care setting in several ways; however, use of such models is scarce. We aimed... (Review)
Review
Machine learning (ML)-based risk prediction models hold the potential to support the health-care setting in several ways; however, use of such models is scarce. We aimed to review health-care professional (HCP) and patient perceptions of ML risk prediction models in published literature, to inform future risk prediction model development. Following database and citation searches, we identified 41 articles suitable for inclusion. Article quality varied with qualitative studies performing strongest. Overall, perceptions of ML risk prediction models were positive. HCPs and patients considered that models have the potential to add benefit in the health-care setting. However, reservations remain; for example, concerns regarding data quality for model development and fears of unintended consequences following ML model use. We identified that public views regarding these models might be more negative than HCPs and that concerns (eg, extra demands on workload) were not always borne out in practice. Conclusions are tempered by the low number of patient and public studies, the absence of participant ethnic diversity, and variation in article quality. We identified gaps in knowledge (particularly views from under-represented groups) and optimum methods for model explanation and alerts, which require future research.
Topics: Humans; Health Personnel; Qualitative Research; Machine Learning; Attitude of Health Personnel; Risk Assessment; Patient Preference
PubMed: 38278615
DOI: 10.1016/S2589-7500(23)00241-8