-
Therapeutic Advances in Gastroenterology 2023Magnetically controlled capsule endoscopy (MCCE) is a non-invasive, painless, comfortable, and safe equipment to diagnose gastrointestinal diseases (GID), partially... (Review)
Review
BACKGROUND
Magnetically controlled capsule endoscopy (MCCE) is a non-invasive, painless, comfortable, and safe equipment to diagnose gastrointestinal diseases (GID), partially overcoming the shortcomings of conventional endoscopy and wireless capsule endoscopy (WCE). With advancements in technology, the main technical parameters of MCCE have continuously been improved, and MCCE has become more intelligent.
OBJECTIVES
The aim of this systematic review was to summarize the research progress of MCCE and artificial intelligence (AI) in the diagnosis and treatment of GID.
DATA SOURCES AND METHODS
We conducted a systematic search of PubMed and EMBASE for published studies on GID detection of MCCE, physical factors related to MCCE imaging quality, the application of AI in aiding MCCE, and its additional functions. We synergistically reviewed the included studies, extracted relevant data, and made comparisons.
RESULTS
MCCE was confirmed to have the same performance as conventional gastroscopy and WCE in detecting common GID, while it lacks research in detecting early gastric cancer (EGC). The body position and cleanliness of the gastrointestinal tract are the main factors affecting imaging quality. The applications of AI in screening intestinal diseases have been comprehensive, while in the detection of common gastric diseases such as ulcers, it has been developed. MCCE can perform some additional functions, such as observations of drug behavior in the stomach and drug damage to the gastric mucosa. Furthermore, it can be improved to perform a biopsy.
CONCLUSION
This comprehensive review showed that the MCCE technology has made great progress, but studies on GID detection and treatment by MCCE are in the primary stage. Further studies are required to confirm the performance of MCCE.
PubMed: 37900007
DOI: 10.1177/17562848231206991 -
The Lancet. Digital Health Dec 2023Machine learning and deep learning models have been increasingly used to predict long-term disease progression in patients with chronic obstructive pulmonary disease... (Meta-Analysis)
Meta-Analysis
Machine learning and deep learning predictive models for long-term prognosis in patients with chronic obstructive pulmonary disease: a systematic review and meta-analysis.
BACKGROUND
Machine learning and deep learning models have been increasingly used to predict long-term disease progression in patients with chronic obstructive pulmonary disease (COPD). We aimed to summarise the performance of such prognostic models for COPD, compare their relative performances, and identify key research gaps.
METHODS
We conducted a systematic review and meta-analysis to compare the performance of machine learning and deep learning prognostic models and identify pathways for future research. We searched PubMed, Embase, the Cochrane Library, ProQuest, Scopus, and Web of Science from database inception to April 6, 2023, for studies in English using machine learning or deep learning to predict patient outcomes at least 6 months after initial clinical presentation in those with COPD. We included studies comprising human adults aged 18-90 years and allowed for any input modalities. We reported area under the receiver operator characteristic curve (AUC) with 95% CI for predictions of mortality, exacerbation, and decline in forced expiratory volume in 1 s (FEV). We reported the degree of interstudy heterogeneity using Cochran's Q test (significant heterogeneity was defined as p≤0·10 or I>50%). Reporting quality was assessed using the TRIPOD checklist and a risk-of-bias assessment was done using the PROBAST checklist. This study was registered with PROSPERO (CRD42022323052).
FINDINGS
We identified 3620 studies in the initial search. 18 studies were eligible, and, of these, 12 used conventional machine learning and six used deep learning models. Seven models analysed exacerbation risk, with only six reporting AUC and 95% CI on internal validation datasets (pooled AUC 0·77 [95% CI 0·69-0·85]) and there was significant heterogeneity (I 97%, p<0·0001). 11 models analysed mortality risk, with only six reporting AUC and 95% CI on internal validation datasets (pooled AUC 0·77 [95% CI 0·74-0·80]) with significant degrees of heterogeneity (I 60%, p=0·027). Two studies assessed decline in lung function and were unable to be pooled. Machine learning and deep learning models did not show significant improvement over pre-existing disease severity scores in predicting exacerbations (p=0·24). Three studies directly compared machine learning models against pre-existing severity scores for predicting mortality and pooled performance did not differ (p=0·57). Of the five studies that performed external validation, performance was worse than or equal to regression models. Incorrect handling of missing data, not reporting model uncertainty, and use of datasets that were too small relative to the number of predictive features included provided the largest risks of bias.
INTERPRETATION
There is limited evidence that conventional machine learning and deep learning prognostic models demonstrate superior performance to pre-existing disease severity scores. More rigorous adherence to reporting guidelines would reduce the risk of bias in future studies and aid study reproducibility.
FUNDING
None.
Topics: Adult; Humans; Reproducibility of Results; Deep Learning; Quality of Life; Pulmonary Disease, Chronic Obstructive; Prognosis
PubMed: 38000872
DOI: 10.1016/S2589-7500(23)00177-2 -
Journal of Stroke and Cerebrovascular... Nov 2023Stroke diagnosis is dependent on lengthy clinical and neuroimaging assessments, while rapid treatment initiation improves clinical outcome. Currently, more sensitive... (Meta-Analysis)
Meta-Analysis
BACKGROUND
Stroke diagnosis is dependent on lengthy clinical and neuroimaging assessments, while rapid treatment initiation improves clinical outcome. Currently, more sensitive biomarker assays of both non-coding RNA- and protein biomarkers have improved their detectability, which could accelerate stroke diagnosis. This systematic review and meta-analysis compares non-coding RNA- with protein biomarkers for their potential to diagnose and differentiate acute stroke (subtypes) in (pre-)hospital settings.
METHODS
We performed a systematic review and meta-analysis of studies evaluating diagnostic performance of non-coding RNA- and protein biomarkers to differentiate acute ischemic and hemorrhagic stroke, stroke mimics, and (healthy) controls. Quality appraisal of individual studies was assessed using the QUADAS-2 tool while the meta-analysis was performed with the sROC approach and by assessing pooled sensitivity and specificity, diagnostic odds ratios, positive- and negative likelihood ratios, and the Youden Index.
SUMMARY OF REVIEW
112 studies were included in the systematic review and 42 studies in the meta-analysis containing 11627 patients with ischemic strokes, 2110 patients with hemorrhagic strokes, 1393 patients with a stroke mimic, and 5548 healthy controls. Proteins (IL-6 and S100 calcium-binding protein B (S100B)) and microRNAs (miR-30a) have similar performance in ischemic stroke diagnosis. To differentiate between ischemic- or hemorrhagic strokes, glial fibrillary acidic protein (GFAP) levels and autoantibodies to the NR2 peptide (NR2aAb, a cleavage product of NMDA neuroreceptors) were best performing whereas no investigated protein or non-coding RNA biomarkers differentiated stroke from stroke mimics with high diagnostic potential.
CONCLUSIONS
Despite sampling time differences, circulating microRNAs (< 24 h) and proteins (< 4,5 h) perform equally well in ischemic stroke diagnosis. GFAP differentiates stroke subtypes, while a biomarker panel of GFAP and UCH-L1 improved the sensitivity and specificity of UCH-L1 alone to differentiate stroke.
Topics: Humans; Hemorrhagic Stroke; Stroke; Biomarkers; Ischemic Stroke; Glial Fibrillary Acidic Protein; MicroRNAs; RNA, Untranslated
PubMed: 37778160
DOI: 10.1016/j.jstrokecerebrovasdis.2023.107388 -
JMIR Medical Informatics Mar 2020Clinical narratives represent the main form of communication within health care, providing a personalized account of patient history and assessments, and offering rich... (Review)
Review
BACKGROUND
Clinical narratives represent the main form of communication within health care, providing a personalized account of patient history and assessments, and offering rich information for clinical decision making. Natural language processing (NLP) has repeatedly demonstrated its feasibility to unlock evidence buried in clinical narratives. Machine learning can facilitate rapid development of NLP tools by leveraging large amounts of text data.
OBJECTIVE
The main aim of this study was to provide systematic evidence on the properties of text data used to train machine learning approaches to clinical NLP. We also investigated the types of NLP tasks that have been supported by machine learning and how they can be applied in clinical practice.
METHODS
Our methodology was based on the guidelines for performing systematic reviews. In August 2018, we used PubMed, a multifaceted interface, to perform a literature search against MEDLINE. We identified 110 relevant studies and extracted information about text data used to support machine learning, NLP tasks supported, and their clinical applications. The data properties considered included their size, provenance, collection methods, annotation, and any relevant statistics.
RESULTS
The majority of datasets used to train machine learning models included only hundreds or thousands of documents. Only 10 studies used tens of thousands of documents, with a handful of studies utilizing more. Relatively small datasets were utilized for training even when much larger datasets were available. The main reason for such poor data utilization is the annotation bottleneck faced by supervised machine learning algorithms. Active learning was explored to iteratively sample a subset of data for manual annotation as a strategy for minimizing the annotation effort while maximizing the predictive performance of the model. Supervised learning was successfully used where clinical codes integrated with free-text notes into electronic health records were utilized as class labels. Similarly, distant supervision was used to utilize an existing knowledge base to automatically annotate raw text. Where manual annotation was unavoidable, crowdsourcing was explored, but it remains unsuitable because of the sensitive nature of data considered. Besides the small volume, training data were typically sourced from a small number of institutions, thus offering no hard evidence about the transferability of machine learning models. The majority of studies focused on text classification. Most commonly, the classification results were used to support phenotyping, prognosis, care improvement, resource management, and surveillance.
CONCLUSIONS
We identified the data annotation bottleneck as one of the key obstacles to machine learning approaches in clinical NLP. Active learning and distant supervision were explored as a way of saving the annotation efforts. Future research in this field would benefit from alternatives such as data augmentation and transfer learning, or unsupervised learning, which do not require data annotation.
PubMed: 32229465
DOI: 10.2196/17984 -
Journal of the International Society of... 2016Carbohydrate supplements are widely used by athletes as an ergogenic aid before and during sports events. The present systematic review and meta-analysis aimed at... (Meta-Analysis)
Meta-Analysis Review
BACKGROUND
Carbohydrate supplements are widely used by athletes as an ergogenic aid before and during sports events. The present systematic review and meta-analysis aimed at synthesizing all available data from randomized controlled trials performed under real-life conditions.
METHODS
MEDLINE, EMBASE, and the Cochrane Central Register of Controlled Trials were searched systematically up to February 2015. Study groups were categorized according to test mode and type of performance measurement. Subgroup analyses were done with reference to exercise duration and range of carbohydrate concentration. Random effects and fixed effect meta-analyses were performed using the Software package by the Cochrane Collaboration Review Manager 5.3.
RESULTS
Twenty-four randomized controlled trials met the objectives and were included in the present systematic review, 16 of which provided data for meta-analyses. Carbohydrate supplementations were associated with a significantly shorter exercise time in groups performing submaximal exercise followed by a time trial [mean difference -0.9 min (95 % confidence interval -1.7, -0.2), p = 0.02] as compared to controls. Subgroup analysis showed that improvements were specific for studies administering a concentration of carbohydrates between 6 and 8 % [mean difference -1.0 min (95 % confidence interval -1.9, -0.0), p = 0.04]. Concerning groups with submaximal exercise followed by a time trial measuring power accomplished within a fixed time or distance, mean power output was significantly higher following carbohydrate load (mean difference 20.2 W (95 % confidence interval 9.0, 31.5), p = 0.0004]. Likewise, mean power output was significantly increased following carbohydrate intervention in groups with time trial measuring power within a fixed time or distance (mean difference 8.1 W (95 % confidence interval 0.5, 15.7) p = 0.04].
CONCLUSION
Due to the limitations of this systematic review, results can only be applied to a subset of athletes (trained male cyclists). For those, we could observe a potential ergogenic benefit of carbohydrate supplementation especially in a concentration range between 6 and 8 % when exercising longer than 90 min.
Topics: Athletes; Athletic Performance; Bicycling; Dietary Carbohydrates; Dietary Supplements; Exercise; Humans; Randomized Controlled Trials as Topic; Sports Nutritional Physiological Phenomena
PubMed: 27408608
DOI: 10.1186/s12970-016-0139-6 -
Nutrients Oct 2022Several studies have explored the effects of capsaicin and capsiate on endurance performance, with conflicting findings. This systematic review aimed to perform a... (Meta-Analysis)
Meta-Analysis Review
Several studies have explored the effects of capsaicin and capsiate on endurance performance, with conflicting findings. This systematic review aimed to perform a meta-analysis examining the effects of capsaicin and capsiate vs. placebo on endurance performance in humans. Seven databases were searched to find eligible studies. The effects of capsaicin and capsiate on aerobic endurance (e.g., time-trials or time-to-exhaustion tests), muscular endurance (e.g., repetitions performed to muscular failure), and rating of perceived exertion (RPE) were examined in a random-effects meta-analysis. Fourteen studies ( = 183) were included in the review. Most studies provided capsaicin or capsiate in the dose of 12 mg, 45 min before exercise. In the meta-analysis for aerobic endurance, there was no significant difference between the placebo and capsaicin/capsiate conditions (Cohen's : 0.04; 95% confidence interval: -0.16, 0.25; = 0.69). In subgroup meta-analyses, there were no significant differences between the placebo and capsaicin/capsiate conditions when analyzing only studies that used time-trials ( = 0.20) or time-to-exhaustion tests ( = 0.80). In the meta-analysis for muscular endurance, a significant ergogenic effect of capsaicin/capsiate was found (Cohen's : 0.27; 95% confidence interval: 0.10, 0.43; = 0.002). When analyzing set-specific effects, an ergogenic effect of capsaicin/capsiate was found in set 1, set 2, and set 3 (Cohen's : 0.21-29). Capsaicin/capsiate ingestion reduced RPE following muscular endurance ( = 0.03) but not aerobic endurance tests ( = 0.58). In summary, capsaicin/capsiate supplementation acutely enhances muscular endurance, while the effects on aerobic endurance are less clear.
Topics: Humans; Capsaicin; Performance-Enhancing Substances; Exercise; Physical Endurance
PubMed: 36364793
DOI: 10.3390/nu14214531 -
Journal of Biomaterials Applications Feb 2020Medical devices made of polydioxanone (a synthetic biodegradable polymer) have been available since the early 1980s. However, no review regarding their performance and...
BACKGROUND
Medical devices made of polydioxanone (a synthetic biodegradable polymer) have been available since the early 1980s. However, no review regarding their performance and safety has been published.
OBJECTIVE
This systematic review intends to review and assess commercially available polydioxanone implants and their safety and performance in patients.
METHODS
We searched for approved polydioxanone implants in several Food and Drug Administration databases. Then, we performed a literature search for publications and clinical trials where polydioxanone devices were implanted in patients. This search was performed on MEDLINE, Embase, Scopus and other databases. Safety and performance of polydioxanone implants in patients were assessed and compared with the implantation of non-polydioxanone devices, when possible, based on scoring systems developed by the authors that analyse surgical site infection rates, inflammatory reaction rates, foreign body response, postoperative pain and fever.
RESULTS
Food and Drug Administration databases search revealed that 48 implants have been approved since 1981, with 1294 adverse reactions or product malfunction in the last decade and 16 recalls. A total of 49 clinical trials and 104 scientific publications were found. Polydioxanone sutures and meshes/plates had low rates of surgical site infection, inflammatory reaction, foreign body response and postoperative fever. Polydioxanone clips/staples reported high rates of surgical site infection, postoperative fever and pain, with sub-optimal clinical performance and poor safety rates. The remaining implants identified showed high levels of safety and performance. Safety scores of polydioxanone implants and non-polydioxanone alternatives are similar. Polydioxanone monofilament sutures perform better than non-polydioxanone alternatives but performance did not differ with remaining polydioxanone implant types.
CONCLUSIONS
Although polydioxanone clips/staples should be implanted with caution and monitored carefully, in general, safety and performance scores of other polydioxanone implants did not differ from non-polydioxanone alternatives. This review will be a useful reference for researchers and industries developing new polydioxanone medical devices.
Topics: Absorbable Implants; Biocompatible Materials; Device Approval; Humans; Inflammation; Medical Device Recalls; Polydioxanone
PubMed: 31771403
DOI: 10.1177/0885328219888841 -
Sports Health 2022With the current Centers for Disease Control and Prevention recommendations for mask use to minimize transmission of coronavirus 2019 (COVID-19) coupled with concern for...
CONTEXT
With the current Centers for Disease Control and Prevention recommendations for mask use to minimize transmission of coronavirus 2019 (COVID-19) coupled with concern for future pandemics that would require mask wearing, providing data-driven guidance with respect to athletic performance is essential.
OBJECTIVE
The purpose of this study was to perform a systematic review of existing literature on the use of face masks while exercising to assess the physiologic effects of face masks worn during athletic activities.
DATA SOURCES
A systematic review was conducted of studies on face mask use during exercise according to Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines. Potential studies were identified through searches of MEDLINE, Embase, Cochrane CENTRAL and CINAHL databases.
STUDY SELECTION
Screening was completed independently by 2 coauthors who sought to identify studies that described the effects of oronasal mask use, if any, on sports/exercise/physical activity, for any age, gender, or level of sport. Articles describing mask effects without exercise, articles published before 1980, and non-English language studies were excluded.
STUDY DESIGN
Systematic review.
LEVEL OF EVIDENCE
Level 3.
DATA EXTRACTION
Data extraction focused on physiologic parameters measured during physical activity performed while wearing a face mask.
RESULTS
Twenty-two articles met all inclusion criteria. Study analysis revealed that the use of masks in healthy volunteers during exercise had no significant effect on physiologic parameters measured including heart rate (HR), respiratory rate (RR), oxygen saturation, and perceived exertion. Of the studies that investigated N95 masks in the healthy adult population, 2 reported modest changes in RR and maximum power output indicative of decreased athletic performance when subjects were exercising at maximum effort. Similar findings were seen in studies of subpopulations including children and pregnant women.
CONCLUSION
Available data suggest that healthy individuals can perform moderate-to-vigorous exercise while wearing a face mask without experiencing changes in HR, RR, and oxygen saturation that would compromise individual safety or athletic performance. In the specific situation in which an N95 mask is worn, maximum power generated may be impaired.
WHAT IS KNOWN ABOUT THE SUBJECT
To date, there has been no systematic review of the existing literature to provide a clear consensus on whether face mask use significantly impacts athletic performance. Mask use has been demonstrated safe in the workplace; however, the use of face masks during exercise has not been examined on a large scale, particularly with respect to physiologic parameters.
WHAT THIS STUDY ADDS TO EXISTING KNOWLEDGE
This analysis highlights that available data suggest that healthy individuals can perform heavy exercise in face masks with minimal physiologic changes. This is the first systematic review of studies analyzing exercise use wearing masks. With the evidence presented here commonly cited concerns about both safety and performance decrements with mask use during physical activities may be allayed.
Topics: Adult; Athletes; COVID-19; Child; Exercise; Female; Humans; Pandemics; Pregnancy; SARS-CoV-2
PubMed: 35855525
DOI: 10.1177/19417381221111395 -
Sports (Basel, Switzerland) May 2020The purpose of this article was to review the data on the relationship between multi-joint isometric strength test (IsoTest) force-time characteristics (peak force, rate... (Review)
Review
The purpose of this article was to review the data on the relationship between multi-joint isometric strength test (IsoTest) force-time characteristics (peak force, rate of force development and impulse) and dynamic performance that is available in the current literature. Four electronic databases were searched using search terms related to IsoTest. Studies were considered eligible if they were original research studies that investigated the relationships between multi-joint IsoTest and performance of dynamic movements; published in peer-reviewed journals; had participants who were athletes or active individuals who participate in recreational sports or resistance training, with no restriction on sex; and had full text available. A total of 47 studies were selected. These studies showed significant small to large correlations between isometric bench press (IBP) force-time variables and upper body dynamic performances ( = 0.221 to 0.608, < 0.05) and significant small to very large correlation between isometric squat (ISqT) ( = 0.085 to 0.746, < 0.05) and isometric mid-thigh pull (IMTP) ( = 0.120 to 0.941, < 0.05) force-time variables with lower body dynamic performances. IsoTest force-time characteristics were shown to have small to very large correlations with dynamic performances of the upper and lower limbs as well as performance of sporting movements ( = 0.118 to 0.700, < 0.05). These data suggest that IsoTest force-time characteristics provide insights into the force production capability of athletes which give insight into dynamic performance capabilities.
PubMed: 32429176
DOI: 10.3390/sports8050063 -
Physical and Engineering Sciences in... Mar 2022To conduct a systematic survey of published techniques for automated diagnosis and prognosis of COVID-19 diseases using medical imaging, assessing the validity of... (Review)
Review
OBJECTIVES
To conduct a systematic survey of published techniques for automated diagnosis and prognosis of COVID-19 diseases using medical imaging, assessing the validity of reported performance and investigating the proposed clinical use-case. To conduct a scoping review into the authors publishing such work.
METHODS
The Scopus database was queried and studies were screened for article type, and minimum source normalized impact per paper and citations, before manual relevance assessment and a bias assessment derived from a subset of the Checklist for Artificial Intelligence in Medical Imaging (CLAIM). The number of failures of the full CLAIM was adopted as a surrogate for risk-of-bias. Methodological and performance measurements were collected from each technique. Each study was assessed by one author. Comparisons were evaluated for significance with a two-sided independent t-test.
FINDINGS
Of 1002 studies identified, 390 remained after screening and 81 after relevance and bias exclusion. The ratio of exclusion for bias was 71%, indicative of a high level of bias in the field. The mean number of CLAIM failures per study was 8.3 ± 3.9 [1,17] (mean ± standard deviation [min,max]). 58% of methods performed diagnosis versus 31% prognosis. Of the diagnostic methods, 38% differentiated COVID-19 from healthy controls. For diagnostic techniques, area under the receiver operating curve (AUC) = 0.924 ± 0.074 [0.810,0.991] and accuracy = 91.7% ± 6.4 [79.0,99.0]. For prognostic techniques, AUC = 0.836 ± 0.126 [0.605,0.980] and accuracy = 78.4% ± 9.4 [62.5,98.0]. CLAIM failures did not correlate with performance, providing confidence that the highest results were not driven by biased papers. Deep learning techniques reported higher AUC (p < 0.05) and accuracy (p < 0.05), but no difference in CLAIM failures was identified.
INTERPRETATION
A majority of papers focus on the less clinically impactful diagnosis task, contrasted with prognosis, with a significant portion performing a clinically unnecessary task of differentiating COVID-19 from healthy. Authors should consider the clinical scenario in which their work would be deployed when developing techniques. Nevertheless, studies report superb performance in a potentially impactful application. Future work is warranted in translating techniques into clinical tools.
Topics: Artificial Intelligence; COVID-19; COVID-19 Testing; Humans; Publishing; Radiography; SARS-CoV-2
PubMed: 34919204
DOI: 10.1007/s13246-021-01093-0