-
PLOS Digital Health May 2022Federated learning (FL) allows multiple institutions to collaboratively develop a machine learning algorithm without sharing their data. Organizations instead share...
OBJECTIVES
Federated learning (FL) allows multiple institutions to collaboratively develop a machine learning algorithm without sharing their data. Organizations instead share model parameters only, allowing them to benefit from a model built with a larger dataset while maintaining the privacy of their own data. We conducted a systematic review to evaluate the current state of FL in healthcare and discuss the limitations and promise of this technology.
METHODS
We conducted a literature search using PRISMA guidelines. At least two reviewers assessed each study for eligibility and extracted a predetermined set of data. The quality of each study was determined using the TRIPOD guideline and PROBAST tool.
RESULTS
13 studies were included in the full systematic review. Most were in the field of oncology (6 of 13; 46.1%), followed by radiology (5 of 13; 38.5%). The majority evaluated imaging results, performed a binary classification prediction task via offline learning (n = 12; 92.3%), and used a centralized topology, aggregation server workflow (n = 10; 76.9%). Most studies were compliant with the major reporting requirements of the TRIPOD guidelines. In all, 6 of 13 (46.2%) of studies were judged at high risk of bias using the PROBAST tool and only 5 studies used publicly available data.
CONCLUSION
Federated learning is a growing field in machine learning with many promising uses in healthcare. Few studies have been published to date. Our evaluation found that investigators can do more to address the risk of bias and increase transparency by adding steps for data homogeneity or sharing required metadata and code.
PubMed: 36812504
DOI: 10.1371/journal.pdig.0000033 -
Facts, Views & Vision in ObGyn Dec 2016In case of preterm birth in twins, it is not well established if the second twin benefits from a delayed-interval delivery. (Review)
Review
BACKGROUND
In case of preterm birth in twins, it is not well established if the second twin benefits from a delayed-interval delivery.
OBJECTIVE
The main objective of this systematic review is to evaluate survival benefit of the second twin from delayed interval delivery compared to the first twin. Secondly, we will evaluate the survival benefit of the procedure when performed equal to or after 24 weeks gestational age of the first born.
METHODS
Delayed interval delivery was defined as every attempt to perform a delayed interval delivery with at minimum placement of a high ligature of the umbilical cord and a delay of delivery of at least 24 hours. Based on the PRISMA method, a systematic review was performed. Controlled and observational studies reporting at least 3 cases of delayed interval delivery in dichorionic diamniotic twin pregnancy describing the outcome of the first and the second twin were included. Case reports and papers on triplet or higher order pregnancies were excluded. Primary data included gestational age and outcome of the first and second born. Metadata concern management strategies (tocolysis, antibiotics, cerclage), neonatal data (sex, birth weight and morbidity) and maternal complications. The methodological quality of included studies was assessed using the "IHE quality appraisal checklist for assessing the quality of case series". Meta-analysis was performed by computing relative risk (RR) with its 95% confidence interval (CI) using the random-effects model. Statistical heterogeneity was tested using the and Chi statistics. Since there is no control group for the secondary outcomes, these are presented by narrative synthesis.
RESULTS
Mortality data were extracted from 13 articles, reporting a total of 128 cases of delayed interval delivery. In the analysis, the second born had a significantly lower mortality risk compared to the first born (relative risk = 0.44, 95% confidence interval = 0.34 - 0.57, P<0.0001, I= 0%, P=0.70). For the analysis of mortality of the second born foetus versus the first born when the first delivery was at ≥24 weeks of gestational age, 12 articles were included. In the analysis 4 reports were excluded since there were no events (no mortality) in both groups (first and second born) making analysis impossible. For the 36 cases included, the second born had a significantly lower mortality risk compared to the first born if delivery of the first born occurred at ≥ 24 weeks gestational age (relative risk=0.37, 95% confidence interval= 0.17 - 0.82, P=0.014, I=0%, P=0.82).
CONCLUSIONS AND IMPLICATIONS
In carefully selected twin pregnancies the survival of the second born twin may improve with delayed interval delivery, also if the first was born at or after 24 weeks. Management protocols in the studies included vary, making it difficult to propose a uniform strategy for delayed interval delivery. Families must be informed about the possibility that a nonviable infant would survive to a periviable gestational age with a risk of severe sequels after birth as well as the possibility of maternal complications.
PubMed: 28210482
DOI: No ID Found -
International Journal of Qualitative... 2015The aim is to identify facilitators and barriers for physical activity (PA) experienced by morbidly obese adults in the Western world. Inactivity and a sedentary... (Review)
Review
The aim is to identify facilitators and barriers for physical activity (PA) experienced by morbidly obese adults in the Western world. Inactivity and a sedentary lifestyle have become a major challenge for health and well-being, particularly among persons with morbid obesity. Lifestyle changes may lead to long-term changes in activity level, if facilitators and barriers are approached in a holistic way by professionals. To develop lifestyle interventions, the perspective and experiences of this group of patients are essential for success. The methodology of the systematic review followed the seven-step procedure of the Joanna Briggs Institute and was published in a protocol. Six databases were searched using keywords and index terms. Manual searches were performed in reference lists and in cited citations up until March 2015. The selected studies underwent quality appraisal in the Joanna Briggs-Qualitative Assessment and Review Instrument. Data from primary studies were extracted and were subjected to a hermeneutic text interpretation and a data-driven coding in a five-step procedure focusing on meaning and constant targeted comparison through which they were categorized and subjected into a meta-synthesis. Eight papers were included for the systematic review, representing the experiences of PA among 212 participants. One main theme developed from the meta-data analysis: "Identity" with the three subthemes: "considering weight," "being able to," and "belonging with others." The theme and subthemes were merged into a meta-synthesis: "Homecoming: a change in identity." The experiences of either suffering or well-being during PA affected the identity of adults with morbid obesity either by challenging or motivating them. A change in identity may be needed to feel a sense of "homecoming" when active.
Topics: Attitude; Exercise; Health Behavior; Humans; Life Style; Motivation; Obesity, Morbid; Self Efficacy; Stress, Psychological
PubMed: 26400462
DOI: 10.3402/qhw.v10.28577 -
Systematic Reviews Oct 2016Overviews of methods are potentially useful means to increase clarity and enhance collective understanding of specific methods topics that may be characterized by... (Review)
Review
BACKGROUND
Overviews of methods are potentially useful means to increase clarity and enhance collective understanding of specific methods topics that may be characterized by ambiguity, inconsistency, or a lack of comprehensiveness. This type of review represents a distinct literature synthesis method, although to date, its methodology remains relatively undeveloped despite several aspects that demand unique review procedures. The purpose of this paper is to initiate discussion about what a rigorous systematic approach to reviews of methods, referred to here as systematic methods overviews, might look like by providing tentative suggestions for approaching specific challenges likely to be encountered. The guidance offered here was derived from experience conducting a systematic methods overview on the topic of sampling in qualitative research.
RESULTS
The guidance is organized into several principles that highlight specific objectives for this type of review given the common challenges that must be overcome to achieve them. Optional strategies for achieving each principle are also proposed, along with discussion of how they were successfully implemented in the overview on sampling. We describe seven paired principles and strategies that address the following aspects: delimiting the initial set of publications to consider, searching beyond standard bibliographic databases, searching without the availability of relevant metadata, selecting publications on purposeful conceptual grounds, defining concepts and other information to abstract iteratively, accounting for inconsistent terminology used to describe specific methods topics, and generating rigorous verifiable analytic interpretations. Since a broad aim in systematic methods overviews is to describe and interpret the relevant literature in qualitative terms, we suggest that iterative decision making at various stages of the review process, and a rigorous qualitative approach to analysis are necessary features of this review type.
CONCLUSIONS
We believe that the principles and strategies provided here will be useful to anyone choosing to undertake a systematic methods overview. This paper represents an initial effort to promote high quality critical evaluations of the literature regarding problematic methods topics, which have the potential to promote clearer, shared understandings, and accelerate advances in research methods. Further work is warranted to develop more definitive guidance.
Topics: Databases, Bibliographic; Decision Making; Evidence-Based Medicine; Humans; Publications; Qualitative Research; Research Design
PubMed: 27729071
DOI: 10.1186/s13643-016-0343-0 -
Comprehensive Reviews in Food Science... Jul 2021Essential oils (EOs) are commercially important products, sources of compounds with antioxidant and antimicrobial activities considered indispensable for several fields,... (Meta-Analysis)
Meta-Analysis
Essential oils (EOs) are commercially important products, sources of compounds with antioxidant and antimicrobial activities considered indispensable for several fields, such as the food industry, cosmetics, perfumes, pharmaceuticals, sanitary and agricultural industries. In this context, this systematic review and meta-analysis, a novel approach will be presented using chemometric tools to verify and recognize patterns of antioxidant, antibacterial, and antifungal activities of EOs according to their geographic, botanical, chemical, and microbiological distribution. Scientific papers were selected following the Preferred Reporting Items for Systematic Review and Meta-Analyses statement flow diagram, and the data were evaluated by the self-organizing map and hierarchical cluster analysis. Overall, this novel approach allowed us to draw an overview of antioxidants and antimicrobials activities of EOs reported in 2019, through 585 articles evaluated, obtaining a dataset with more than 10,000 data, distributed in more than 80 countries, 290 plant genera, 150 chemical compounds, 30 genera of bacteria, and 10 genera of fungi. The networks for geographic, botanical, chemical, and microbiological distribution indicated that Brazil, Asia, the botanical genus Thymus, species Thymus vulgaris L. "thyme," the Lamiaceae family, limonene, and the oxygenated monoterpene class were the most representative in the dataset, while the species Escherichia coli and Candida albicans were the most used to assess the antimicrobial activity of EOs. This work can be seen as a guide for the processing of metadata using a novel approach with non-conventional statistical methods. However, this preliminary approach with EOs can be extended to other sources or areas of food science.
Topics: Candida albicans; Lamiaceae; Microbial Sensitivity Tests; Oils, Volatile; Thymus Plant
PubMed: 34125485
DOI: 10.1111/1541-4337.12773 -
Scandinavian Journal of Surgery : SJS :... 2022Patients presenting with synchronous colorectal liver metastases are increasingly being considered for a curative treatment, and the liver-first approach is gaining... (Meta-Analysis)
Meta-Analysis
BACKGROUND
Patients presenting with synchronous colorectal liver metastases are increasingly being considered for a curative treatment, and the liver-first approach is gaining popularity in this context. However, little is known about the completion rates of the liver-first approach and its effects on survival.
METHODS
A systematic review and meta-analysis of liver-first strategy for colorectal liver metastasis. The primary outcome was an assessment of the completion rates of the liver-first approach. Secondary outcomes included overall survival, causes of non-completion, and clinicopathologic data.
RESULTS
Seventeen articles were amenable for inclusion and the total study population was 1041. The median completion rate for the total population was 80% (range 20-100). The median overall survival for the completion and non-completion groups was 45 (range 12-69) months and 13 (range 10.5-25) months, respectively. Metadata showed a significant survival benefit for the completion group, with a univariate hazard ratio of 12.0 (95% confidence interval, range 5.7-24.4). The major cause of non-completion (76%) was liver disease progression before resection of the primary tumor. Pearson tests showed significant negative correlation between median number of lesions and median size of the largest metastasis and completion rate.
CONCLUSIONS
The liver-first approach offers a complete resection to most patients enrolled, with an overall survival benefit when completion can be assured. One-fifth fails to return to intended oncologic therapy and the major cause is interim metastatic progression, most often in the liver. Risk of non-completion is related to a higher number of lesions and large metastases. The majority of studies stem from primary rectal cancers, which may influence on the return to intended oncologic therapy as well. 170459.
Topics: Abdomen; Colorectal Neoplasms; Hepatectomy; Humans; Liver Neoplasms; Retrospective Studies; Treatment Outcome
PubMed: 34605325
DOI: 10.1177/14574969211030131 -
Translational Vision Science &... Feb 2024Retinal images contain rich biomarker information for neurodegenerative disease. Recently, deep learning models have been used for automated neurodegenerative disease...
PURPOSE
Retinal images contain rich biomarker information for neurodegenerative disease. Recently, deep learning models have been used for automated neurodegenerative disease diagnosis and risk prediction using retinal images with good results.
METHODS
In this review, we systematically report studies with datasets of retinal images from patients with neurodegenerative diseases, including Alzheimer's disease, Huntington's disease, Parkinson's disease, amyotrophic lateral sclerosis, and others. We also review and characterize the models in the current literature which have been used for classification, regression, or segmentation problems using retinal images in patients with neurodegenerative diseases.
RESULTS
Our review found several existing datasets and models with various imaging modalities primarily in patients with Alzheimer's disease, with most datasets on the order of tens to a few hundred images. We found limited data available for the other neurodegenerative diseases. Although cross-sectional imaging data for Alzheimer's disease is becoming more abundant, datasets with longitudinal imaging of any disease are lacking.
CONCLUSIONS
The use of bilateral and multimodal imaging together with metadata seems to improve model performance, thus multimodal bilateral image datasets with patient metadata are needed. We identified several deep learning tools that have been useful in this context including feature extraction algorithms specifically for retinal images, retinal image preprocessing techniques, transfer learning, feature fusion, and attention mapping. Importantly, we also consider the limitations common to these models in real-world clinical applications.
TRANSLATIONAL RELEVANCE
This systematic review evaluates the deep learning models and retinal features relevant in the evaluation of retinal images of patients with neurodegenerative disease.
Topics: Humans; Algorithms; Alzheimer Disease; Deep Learning; Machine Learning; Neurodegenerative Diseases; Datasets as Topic; Retina
PubMed: 38381447
DOI: 10.1167/tvst.13.2.16 -
PloS One 2016Research on perennial staple crops has increased in the past ten years due to their potential to improve ecosystem services in agricultural systems. However, multiple... (Comparative Study)
Comparative Study Review
Research on perennial staple crops has increased in the past ten years due to their potential to improve ecosystem services in agricultural systems. However, multiple past breeding efforts as well as research on traditional ratoon systems mean there is already a broad body of literature on perennial crops. In this review, we compare the development of research on perennial staple crops, including wheat, rice, rye, sorghum, and pigeon pea. We utilized the advanced search capabilities of Web of Science, Scopus, ScienceDirect, and Agricola to gather a library of 914 articles published from 1930 to the present. We analyzed the metadata in the entire library and in collections of literature on each crop to understand trends in research and publishing. In addition, we applied topic modeling to the article abstracts, a type of text analysis that identifies frequently co-occurring terms and latent topics. We found: 1.) Research on perennials is increasing overall, but individual crops have each seen periods of heightened interest and research activity; 2.) Specialist journals play an important role in supporting early research efforts. Research often begins within communities of specialists or breeders for the individual crop before transitioning to a more general scientific audience; 3.) Existing perennial agricultural systems and their domesticated crop material, such as ratoon rice systems, can provide a useful foundation for breeding efforts, accelerating the development of truly perennial crops and farming systems; 4.) Primary research is lacking for crops that are produced on a smaller scale globally, such as pigeon pea and sorghum, and on the ecosystem service benefits of perennial agricultural systems.
Topics: Bibliometrics; Cajanus; Crops, Agricultural; Databases, Bibliographic; Models, Theoretical; Oryza; Secale; Sorghum; Triticum
PubMed: 27213283
DOI: 10.1371/journal.pone.0155788 -
Reproduction (Cambridge, England) Jun 2023Adverse trends in reproductive function are a concern in humans, companion, livestock, and wildlife species. This study indicates that equine populations are at risk of...
IN BRIEF
Adverse trends in reproductive function are a concern in humans, companion, livestock, and wildlife species. This study indicates that equine populations are at risk of a comparable decline in sperm progressive motility.
ABSTRACT
There is increasing evidence reporting geographically sensitive adverse trends in human semen quality, with parallel trends observed in the dog sentinel. Despite significant economic and welfare complications associated with poor testicular function, trends in current equine populations are undetermined. Given the predictive value of sperm progressive motility (PMOT) in male factor infertility and fertilisation potential, research determining trends in this parameter is warranted. This research analysed trends in stallion sperm PMOT through systematic review and meta-regression. Using a comprehensive search strategy, Scopus, Embase (Ovid), Medline (Ovid), and VetMed (CAB direct) were scoped for eligible data. Using best practices, 230 meta-data points from 229 articles published from 1991 to 2021 were collated for meta-regression analysis. Sperm PMOT declined significantly between 1984 and 2019 (simple linear regression: b -0.340, P = 0.017; meta-regression: b -0.610, P ≤ 0.001). Overall and yearly PMOT declines were predicted at 33.51 and 0.96%, respectively (1984: 63.69 ± 5.07%; 2019: 42.35 ± 3.69%). Trends remained consistent irrespective of sensitivity analyses. Yearly and overall declines were stronger in western (yearly: 0.75%, overall: 26.29%) compared to non-western (yearly: 0.46%, overall: 10.65%) populations. Adverse trends contribute vital data to the debate surrounding declining semen quality, supporting the use of equines as novel comparative models for human reproduction. Results could have significant economic, health, and welfare consequences for equine breeding sectors. A comparable decline in human, dog, and horse sperm quality is indicative of a common environmental aetiology, indicating the need for a holistic One Health approach in determining causes and developing preventative strategies.
Topics: Male; Horses; Animals; Humans; Dogs; Semen Analysis; Semen; Sperm Motility; Spermatozoa; Infertility, Male; Sperm Count
PubMed: 37000597
DOI: 10.1530/REP-22-0490 -
PeerJ 2023The emerging field of environmental DNA (eDNA) research lacks universal guidelines for ensuring data produced are FAIR-findable, accessible, interoperable, and...
The emerging field of environmental DNA (eDNA) research lacks universal guidelines for ensuring data produced are FAIR-findable, accessible, interoperable, and reusable-despite growing awareness of the importance of such practices. In order to better understand these data usability challenges, we systematically reviewed 60 peer reviewed articles conducting a specific subset of eDNA research: metabarcoding studies in marine environments. For each article, we characterized approximately 90 features across several categories: general article attributes and topics, methodological choices, types of metadata included, and availability and storage of sequence data. Analyzing these characteristics, we identified several barriers to data accessibility, including a lack of common context and vocabulary across the articles, missing metadata, supplementary information limitations, and a concentration of both sample collection and analysis in the United States. While some of these barriers require significant effort to address, we also found many instances where small choices made by authors and journals could have an outsized influence on the discoverability and reusability of data. Promisingly, articles also showed consistency and creativity in data storage choices as well as a strong trend toward open access publishing. Our analysis underscores the need to think critically about data accessibility and usability as marine eDNA metabarcoding studies, and eDNA projects more broadly, continue to proliferate.
Topics: DNA, Environmental; Biodiversity; DNA Barcoding, Taxonomic
PubMed: 36992947
DOI: 10.7717/peerj.14993