-
JMIR Medical Informatics May 2024With the increasing availability of data, computing resources, and easier-to-use software libraries, machine learning (ML) is increasingly used in disease detection and... (Review)
Review
BACKGROUND
With the increasing availability of data, computing resources, and easier-to-use software libraries, machine learning (ML) is increasingly used in disease detection and prediction, including for Parkinson disease (PD). Despite the large number of studies published every year, very few ML systems have been adopted for real-world use. In particular, a lack of external validity may result in poor performance of these systems in clinical practice. Additional methodological issues in ML design and reporting can also hinder clinical adoption, even for applications that would benefit from such data-driven systems.
OBJECTIVE
To sample the current ML practices in PD applications, we conducted a systematic review of studies published in 2020 and 2021 that used ML models to diagnose PD or track PD progression.
METHODS
We conducted a systematic literature review in accordance with PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines in PubMed between January 2020 and April 2021, using the following exact string: "Parkinson's" AND ("ML" OR "prediction" OR "classification" OR "detection" or "artificial intelligence" OR "AI"). The search resulted in 1085 publications. After a search query and review, we found 113 publications that used ML for the classification or regression-based prediction of PD or PD-related symptoms.
RESULTS
Only 65.5% (74/113) of studies used a holdout test set to avoid potentially inflated accuracies, and approximately half (25/46, 54%) of the studies without a holdout test set did not state this as a potential concern. Surprisingly, 38.9% (44/113) of studies did not report on how or if models were tuned, and an additional 27.4% (31/113) used ad hoc model tuning, which is generally frowned upon in ML model optimization. Only 15% (17/113) of studies performed direct comparisons of results with other models, severely limiting the interpretation of results.
CONCLUSIONS
This review highlights the notable limitations of current ML systems and techniques that may contribute to a gap between reported performance in research and the real-life applicability of ML models aiming to detect and predict diseases such as PD.
PubMed: 38771237
DOI: 10.2196/50117 -
La Clinica Terapeutica 2024Variations in cystic artery anatomy are not unusual in occurrence, hence considerably crucial during hepatobiliary surgical planning and execution. This systematic... (Meta-Analysis)
Meta-Analysis
BACKGROUND
Variations in cystic artery anatomy are not unusual in occurrence, hence considerably crucial during hepatobiliary surgical planning and execution. This systematic review and meta-analysis of the anatomical variations of cystic artery (CA) was undertaken to emphasize their significance in surgical practice.
METHODS
The PICO model was adopted, both MeSH term and free keywords were utilized for the search strategy. The risk of bias in each study was calculated by the anatomy quality assurance (AQUA) tool.
RESULT
The search strategy identified 8204 records, extracted 5529 studies, and evaluated 117 abstracts. Out of these 117 studies, 53 met the eligibility criteria. The CA was absent in 2% of instances (95% CI: 0.01-0.04), indicating that 98% of cases had the CA. In 10071 participants from 29 investigations, double cystic arteries were found in 13% (95% CI: 11-16%), with significant heterogeneity (I2 = 91%). In 46 studies with a total of 9928 participants, 89% of the individuals had CA originating from RHA (95% CI: 85%-92%) with significant heterogeneity (I2=94.3%) and a predictive range of 43%-99%.
CONCLUSION
The cystic artery is primarily derived from the right hepatic artery, followed by aberrant, proper, and left hepatic arteries. It is located anterior to common hepatic ducts and cystic ducts. The mean length and diameter of CA were 20.77 mm and 1.91 mm Short cystic arteries are common (20%) Congenital anomalies like absent and double cystic arteries have low prevalence but must be conside-red during surgery.
Topics: Humans; Hepatic Artery; Anatomic Variation
PubMed: 38767077
DOI: 10.7417/CT.2024.5061 -
Digital Health 2024Mobile health applications hold immense potential for enhancing health outcomes. Usability is one of the main factors for the adoption and use of mobile health... (Review)
Review
OBJECTIVE
Mobile health applications hold immense potential for enhancing health outcomes. Usability is one of the main factors for the adoption and use of mobile health applications. However, despite the growing importance of mHealth applications, clear standards for their evaluation remain elusive. The present study aimed to determine heuristics for the usability evaluation of health-related applications.
METHODS
We systematically searched multiple databases for relevant papers published between January 2008 and April 2021. Articles were reviewed, and data were extracted and categorized from those meeting inclusion criteria by two authors independently. Heuristics were identified based on statements, words, and concepts expressed in the studies. These heuristics were first mapped to Nielsen's heuristics based on their differences or similarities. The remaining heuristics that were very important for mobile applications were categorized into new heuristics.
RESULTS
Seventeen studies met the eligibility criteria. Seventy-nine heuristics were extracted from the papers. After combining the items with the same concepts and removing irrelevant items based on the exclusion criteria, 20 heuristics remained. Common heuristics such as "Visibility of system status" and "Flexibility and efficiency of use" were categorized into 10 previously established heuristics and new heuristics like "Navigation" and "User engagement" were recognized as new ones.
CONCLUSIONS
In our study, we have meticulously identified 20 heuristics that hold promise for evaluating and designing mHealth applications. These heuristics can be used by the researchers for the development of robust tools for heuristic evaluation. These tools, when adapted or tailored for health domain applications, have the potential to significantly enhance the quality of mHealth applications. Ultimately, this improvement in quality translates to enhanced patient safety.
PROTOCOL REGISTRATION
(10.17605/OSF.IO/PZJ7H).
PubMed: 38766365
DOI: 10.1177/20552076241253539 -
Psychiatry and Clinical... Dec 2023Major depressive disorder is the leading cause of mental health-related burden globally and up to one-third of major depressive disorder patients never achieve...
BACKGROUND
Major depressive disorder is the leading cause of mental health-related burden globally and up to one-third of major depressive disorder patients never achieve remission. Transcranial Direct Current Stimulation is a non-invasive intervention used to treat individuals diagnosed with major depressive disorder and bipolar disorder. Since the last transcranial direct current stimulation review specifically focusing on cognitive symptoms in major depressive disorder, twice as many papers have been published.
METHODS
A systematic review was conducted with 5 electronic databases from database inception until March 21, 2022. Randomized controlled trials with at least 1 arm evaluating transcranial direct current stimulation in adults (diagnosed with major depressive disorder or bipolar disorder using the Diagnostic and Statistical Manual of Mental Disorders or International Classification of Diseases criteria) aged 18 or older were included. Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines were adopted.
RESULTS
: A total of 972 participants were included across 14 studies (60.5% female; mean age of 47.0 years [SD = 16.8]). Nine studies focused on participants with major depressive disorder and all studies used the Diagnostic and Statistical Manual of Mental Disorders to diagnose the participants. Seven out of the 14 studies showed significant improvements in at least 1 cognitive outcome measure in the active transcranial direct current stimulation group compared to the sham group. Several cognitive measures were used across studies, and 12 of the 14 studies reported mild-to-moderate side effects from treatment.
CONCLUSION
: Current transcranial direct current stimulation literature has shown limited evidence for the treatment of cognitive impairments in major depressive disorder and bipolar disorder. Future research that applies machine learning algorithms may enable us to distinguish responders from non-responders, increasing clinical benefits of transcranial direct current stimulation.
PubMed: 38765850
DOI: 10.5152/pcp.2023.22583 -
Heliyon May 2024Deep Reinforcement Learning (DRL) has gained significant adoption in diverse fields and applications, mainly due to its proficiency in resolving complicated...
Deep Reinforcement Learning (DRL) has gained significant adoption in diverse fields and applications, mainly due to its proficiency in resolving complicated decision-making problems in spaces with high-dimensional states and actions. Deep Deterministic Policy Gradient (DDPG) is a well-known DRL algorithm that adopts an actor-critic approach, synthesizing the advantages of value-based and policy-based reinforcement learning methods. The aim of this study is to provide a thorough examination of the latest developments, patterns, obstacles, and potential opportunities related to DDPG. A systematic search was conducted using relevant academic databases (Scopus, Web of Science, and ScienceDirect) to identify 85 relevant studies published in the last five years (2018-2023). We provide a comprehensive overview of the key concepts and components of DDPG, including its formulation, implementation, and training. Then, we highlight the various applications and domains of DDPG, including Autonomous Driving, Unmanned Aerial Vehicles, Resource Allocation, Communications and the Internet of Things, Robotics, and Finance. Additionally, we provide an in-depth comparison of DDPG with other DRL algorithms and traditional RL methods, highlighting its strengths and weaknesses. We believe that this review will be an essential resource for researchers, offering them valuable insights into the methods and techniques utilized in the field of DRL and DDPG.
PubMed: 38765095
DOI: 10.1016/j.heliyon.2024.e30697 -
Cancer Treatment Reviews Jul 2024Local resection (LR) methods for rectal cancer are generally considered in the palliative setting or for patients deemed a high anaesthetic risk. This systematic review... (Meta-Analysis)
Meta-Analysis
BACKGROUND
Local resection (LR) methods for rectal cancer are generally considered in the palliative setting or for patients deemed a high anaesthetic risk. This systematic review and meta-analysis aimed to compare oncological outcomes of LR and radical resection (RR) for early rectal cancer in the context of staging and surveillance assessment.
METHODS
A literature search of MEDLINE, Embase and Emcare databases was performed for studies that reported data on clinical outcomes for both LR and RR for early rectal cancer from January 1995 to April 2023. Meta-analysis was performed using random-effect models and between-study heterogeneity was assessed. The quality of assessment was assessed using the Newcastle-Ottawa Scale for observational studies and the Cochrane Risk of Bias 2.0 tool for randomised controlled trials.
RESULTS
Twenty studies with 12,022 patients were included: 6,476 patients had LR and 5,546 patients underwent RR. RR led to an improvement in 5-year overall survival (OR 1.84; 95 % CI 1.54-2.20; p < 0.0001; I 20 %) and local recurrence (OR 3.06; 95 % CI 2.02-4.64; p < 0.0001; I 39 %) when compared to LR. However, when staging and surveillance methods were clearly adopted in LR cases, there was an improvement in R0 rates (96.7 % vs 85.6 %), 5-year disease-free survival (93.0 % vs 77.9 %) and overall survival (81.6 % vs 79.0 %) compared to when staging and surveillance was not reported/performed.
CONCLUSIONS
LR may be appropriate for selected patients without poor prognostic factors in early rectal cancer. This study also highlights that there is currently no single standardised staging or surveillance approach being adopted in the management of early rectal cancer. A more specified and standardised preoperative staging for patient selection as well as clinical and image-based surveillance protocols is needed.
Topics: Humans; Neoplasm Recurrence, Local; Neoplasm Staging; Rectal Neoplasms; Treatment Outcome
PubMed: 38761791
DOI: 10.1016/j.ctrv.2024.102753 -
Breast (Edinburgh, Scotland) May 2024There are a wide variety of intraoperative techniques available in breast surgery to achieve low rates for positive margins of excision. The objective of this systematic... (Review)
Review
PURPOSE
There are a wide variety of intraoperative techniques available in breast surgery to achieve low rates for positive margins of excision. The objective of this systematic review was to determine the pooled diagnostic accuracy of intraoperative breast margin assessment techniques that have been evaluated in clinical practice.
METHODS
This study was performed in accordance with PRISMA guidelines. A systematic search of the literature was conducted to identify studies assessing the diagnostic accuracy of intraoperative margin assessment techniques. Only clinical studies with raw diagnostic accuracy data as compared with final permanent section histopathology were included in the meta-analysis. A bivariate model for diagnostic meta-analysis was used to determine overall pooled sensitivity and specificity.
RESULTS
Sixty-one studies were eligible for inclusion in this systematic review and meta-analysis. Cytology demonstrated the best diagnostic accuracy, with pooled sensitivity of 0.92 (95 % CI 0.77-0.98) and a pooled specificity of 0.95 (95 % CI 0.90-0.97). The findings also indicate good diagnostic accuracy for optical spectroscopy, with a pooled sensitivity of 0.86 (95 % CI 0.76-0.93) and a pooled specificity of 0.92 (95 % CI 0.82-0.97).
CONCLUSION
Pooled data indicate that optical spectroscopy, cytology and frozen section have the greatest diagnostic accuracy of currently available intraoperative margin assessment techniques. However, long turnaround time for results and their resource intensive nature has prevented widespread adoption of these methods. The aim of emerging technologies is to compete with the diagnostic accuracy of these established techniques, while improving speed and usability.
PubMed: 38759577
DOI: 10.1016/j.breast.2024.103749 -
Clinical epidemiology and outcomes of emergency department-acute kidney injury: A systematic review.Heliyon May 2024Over half of all community-acquired acute kidney injury (CA-AKI) initially presented to emergency department (ED), but emergency department acute kidney injury (ED-AKI)...
BACKGROUND
Over half of all community-acquired acute kidney injury (CA-AKI) initially presented to emergency department (ED), but emergency department acute kidney injury (ED-AKI) is poorly characterised, poorly understood with no systematic review, often under-recognized and under-managed.
OBJECTIVE
To review the incidence, risk factors, and outcomes of ED-AKI, and risk factors of post-ED-AKI mortality globally.
METHODS
We included published prospective or retrospective observational studies, controlled trials, and systematic reviews reporting AKI in adult ED attendees within 24 h of ED admission. Iatrogenic causes of AKI from medical interventions were excluded. We used PubMed to identify articles from 1996 to August 14, 2021, and adopted the National Heart, Lung, and Blood Institute (NHLBI) Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies to assess risk of bias. We used a Forest plot to present pooled ED-AKI incidence rates and I statistics. Other parameters were summarized narratively.
RESULTS
Using 24 h from ED admission as the definition for ED-AKI we identified six articles from 2005 to 2018 in high-income settings and one article with a 48-h timeframe. The pooled incidence of ED-AKI was 20 per 1000 adult ED attendances. Risk factors for ED-AKI included increasing age, nursing home residence, previous hospital admission within 30 days, discharge diagnosis of diabetes, obstructive uropathy, sepsis, gastrointestinal medical conditions, high serum creatinine, bilirubin, C-reactive protein, white blood cell, alanine aminotransferase, low serum sodium or albumin on admission, poor premorbid renal function, antibiotic use, active malignancy, lung disease, hyperlipidaemia, and infection. Crude, all-cause 24-h mortality rate was 4.56 % and the one-year mortality rate was 35.04 %. Increasing age and comorbidities including cardiovascular disease and malignancy were associated with higher mortality rates.
CONCLUSION
The review reveals a paucity of relevant literature which calls for further research, increased vigilance, red flag identification, and standardized management protocols for ED-AKI.
PubMed: 38756601
DOI: 10.1016/j.heliyon.2024.e30580 -
PloS One 2024Non-invasive prenatal testing (NIPT) is a widely adopted maternal blood test that analyses foetal originating DNA to screen for foetal chromosomal conditions, including... (Meta-Analysis)
Meta-Analysis
BACKGROUND
Non-invasive prenatal testing (NIPT) is a widely adopted maternal blood test that analyses foetal originating DNA to screen for foetal chromosomal conditions, including Down's syndrome (DS). The introduction of this test, which may have implications for important decisions made during pregnancy, requires continual monitoring and evaluation. This systematic review aims to assess the extent of NIPT introduction into national screening programmes for DS worldwide, its uptake, and impact on pregnancy outcomes.
METHODS AND FINDINGS
The study protocol was published in PROSPERO (CRD42022306167). We systematically searched MEDLINE, CINAHL, Scopus, and Embase for population-based studies, government guidelines, and Public Health documents from 2010 onwards. Results summarised the national policies for NIPT implementation into screening programmes geographically, along with population uptake. Meta-analyses estimated the pooled proportions of women choosing invasive prenatal diagnosis (IPD) following a high chance biochemical screening result, before and after NIPT was introduced. Additionally, we meta-analysed outcomes (termination of pregnancy and live births) amongst high chance pregnancies identified by NIPT. Results demonstrated NIPT implementation in at least 27 countries. Uptake of second line NIPT varied, from 20.4% to 93.2% (n = 6). Following NIPT implementation, the proportion of women choosing IPD after high chance biochemical screening decreased from 75% (95% CI 53%, 88%, n = 5) to 43% (95%CI 31%, 56%, n = 5), an absolute risk reduction of 38%. A pooled estimate of 69% (95% CI 52%, 82%, n = 7) of high chance pregnancies after NIPT resulted in termination, whilst 8% (95% CI 3%, 21%, n = 7) had live births of babies with DS.
CONCLUSIONS
NIPT has rapidly gained global acceptance, but population uptake is influenced by healthcare structures, historical screening practices, and cultural factors. Our findings indicate a reduction in IPD tests following NIPT implementation, but limited pre-NIPT data hinder comprehensive impact assessment. Transparent, comparable data reporting is vital for monitoring NIPT's potential consequences.
Topics: Humans; Down Syndrome; Female; Pregnancy; Noninvasive Prenatal Testing; Prenatal Diagnosis; Pregnancy Outcome
PubMed: 38753891
DOI: 10.1371/journal.pone.0298643 -
PloS One 2024Improved sanitation refers to those that effectively avoid human contact with excreta in a hygienic manner. Having improved latrines is a key factor in adopting safe... (Meta-Analysis)
Meta-Analysis Review
INTRODUCTION
Improved sanitation refers to those that effectively avoid human contact with excreta in a hygienic manner. Having improved latrines is a key factor in adopting safe ways of disposing of child feces. However, previous studies in Africa that examined how owning improved latrine facilities associated with household child feces disposal practices has shown inconsistent results, and no systematic review of these findings has been done. Therefore, this study aims to synthesize the evidence on the significance of households having improved latrine facilities for safe child feces disposal practices among households with under five-year-old children in Africa.
METHODS
The searched databases include: PubMed/Medline, Ovid/Embase, ScienceDirect, AJOL and the Cochrane Library. In the search process, Google Scholar and references of other studies were considered. This review included studies that were published in English without any time restrictions. The outcome of this study was an estimate of the association between the ownership of an improved latrine and the disposal practices of children's feces. Two reviewers used the Excel data extraction tool to extract the relevant data from the studies that were included in the review. Using Stata version 16, a meta-analysis was performed with a random effects statistical model. The inverse index of variance (I2) was used to assess heterogeneity. Forest plots were used to show the pooled estimate with a 95% confidence interval. Publication bias was assessed using Egger's test and a funnel plot.
RESULTS
Out of the 616 studies that were retrieved, 15 were included in the systematic review analysis and 10 were included in the meta-analysis. All studies that were included are cross-sectional studies done in Ethiopia, Nigeria, Gambia, Malawi, Eswatini, Ghana, Zambia, and a study used data from sub-Saharan Africa. Improved latrine facilities significantly enhanced the practice of safe child feces disposal, as shown by the overall effect size (OR = 2.74; 95% CI = 1.24-1.35, I2 = 99.95%). In the subgroup analysis by sample size, the presence of improved latrines significantly enhanced safe child feces disposal in studies with sample sizes less than 1000 (OR = 3.24; 95% CI = 2.86-3.62, I2 = 61.38%), while there was no significant difference in studies with sample sizes greater than 1000 (OR = 2.67; 95% CI = 0.69-4.64, I2 = 99.97%). However, studies that involved children under 5 years old indicated that improved latrine facilities significantly enhanced the practice of safe child feces disposal (OR = 4.02; 95% CI = 2.03-6.09; I2 = 99.96%).
CONCLUSIONS
In this research study, we examined the ownership of improved latrine facilities among households with five-year-old children to enhance the disposal of child feces in a safer manner in Africa. The high heterogeneity among the studies and the cross-sectional design of the included studies limit the causal inference and generalizability of the findings. Therefore, meta-analyses of longitudinal and experimental studies are needed to confirm the causal relationship between improved latrine facilities and safe child feces disposal practices in Africa.
Topics: Toilet Facilities; Humans; Feces; Africa; Child, Preschool; Refuse Disposal; Sanitation; Child; Ownership
PubMed: 38753650
DOI: 10.1371/journal.pone.0303754