-
Human Reproduction Open 2023What is the present performance of artificial intelligence (AI) decision support during embryo selection compared to the standard embryo selection by embryologists?
STUDY QUESTION
What is the present performance of artificial intelligence (AI) decision support during embryo selection compared to the standard embryo selection by embryologists?
SUMMARY ANSWER
AI consistently outperformed the clinical teams in all the studies focused on embryo morphology and clinical outcome prediction during embryo selection assessment.
WHAT IS KNOWN ALREADY
The ART success rate is ∼30%, with a worrying trend of increasing female age correlating with considerably worse results. As such, there have been ongoing efforts to address this low success rate through the development of new technologies. With the advent of AI, there is potential for machine learning to be applied in such a manner that areas limited by human subjectivity, such as embryo selection, can be enhanced through increased objectivity. Given the potential of AI to improve IVF success rates, it remains crucial to review the performance between AI and embryologists during embryo selection.
STUDY DESIGN SIZE DURATION
The search was done across PubMed, EMBASE, Ovid Medline, and IEEE Xplore from 1 June 2005 up to and including 7 January 2022. Included articles were also restricted to those written in English. Search terms utilized across all databases for the study were: ('Artificial intelligence' OR 'Machine Learning' OR 'Deep learning' OR 'Neural network') AND ('IVF' OR ' fertili*' OR 'assisted reproductive techn*' OR 'embryo'), where the character '*' refers the search engine to include any auto completion of the search term.
PARTICIPANTS/MATERIALS SETTING METHODS
A literature search was conducted for literature relating to AI applications to IVF. Primary outcomes of interest were accuracy, sensitivity, and specificity of the embryo morphology grade assessments and the likelihood of clinical outcomes, such as clinical pregnancy after IVF treatments. Risk of bias was assessed using the Modified Down and Black Checklist.
MAIN RESULTS AND THE ROLE OF CHANCE
Twenty articles were included in this review. There was no specific embryo assessment day across the studies-Day 1 until Day 5/6 of embryo development was investigated. The types of input for training AI algorithms were images and time-lapse (10/20), clinical information (6/20), and both images and clinical information (4/20). Each AI model demonstrated promise when compared to an embryologist's visual assessment. On average, the models predicted the likelihood of successful clinical pregnancy with greater accuracy than clinical embryologists, signifying greater reliability when compared to human prediction. The AI models performed at a median accuracy of 75.5% (range 59-94%) on predicting embryo morphology grade. The correct prediction (Ground Truth) was defined through the use of embryo images according to post embryologists' assessment following local respective guidelines. Using blind test datasets, the embryologists' accuracy prediction was 65.4% (range 47-75%) with the same ground truth provided by the original local respective assessment. Similarly, AI models had a median accuracy of 77.8% (range 68-90%) in predicting clinical pregnancy through the use of patient clinical treatment information compared to 64% (range 58-76%) when performed by embryologists. When both images/time-lapse and clinical information inputs were combined, the median accuracy by the AI models was higher at 81.5% (range 67-98%), while clinical embryologists had a median accuracy of 51% (range 43-59%).
LIMITATIONS REASONS FOR CAUTION
The findings of this review are based on studies that have not been prospectively evaluated in a clinical setting. Additionally, a fair comparison of all the studies were deemed unfeasible owing to the heterogeneity of the studies, development of the AI models, database employed and the study design and quality.
WIDER IMPLICATIONS OF THE FINDINGS
AI provides considerable promise to the IVF field and embryo selection. However, there needs to be a shift in developers' perception of the clinical outcome from successful implantation towards ongoing pregnancy or live birth. Additionally, existing models focus on locally generated databases and many lack external validation.
STUDY FUNDING/COMPETING INTERESTS
This study was funded by Monash Data Future Institute. All authors have no conflicts of interest to declare.
REGISTRATION NUMBER
CRD42021256333.
PubMed: 37588797
DOI: 10.1093/hropen/hoad031 -
BMJ Open Aug 2023To systematically investigate the associations between vision impairment and risk of motor vehicle crash (MVC) involvement, and evaluate vision-related interventions to... (Meta-Analysis)
Meta-Analysis
OBJECTIVES
To systematically investigate the associations between vision impairment and risk of motor vehicle crash (MVC) involvement, and evaluate vision-related interventions to reduce MVCs.
DESIGN
Medline (Ovid), EMBASE and Global Health electronic databases were systematically searched from inception to March 2022 for observational and interventional English-language studies. Screening, data extraction and appraisals using the Joanna Briggs Institute appraisal tools were completed by two reviewers independently. Where appropriate, measures of association were converted into risk ratios (RRs) or ORs for meta-analysis.
PARTICIPANTS
Drivers of four-wheeled vehicles of all ages with no cognitive declines.
PRIMARY AND SECONDARY OUTCOMES
MVC involvement (primary) and driving cessation (secondary).
RESULTS
101 studies (n=778 052) were included after full-text review. 57 studies only involved older drivers (≥65 years) and 85 were in high-income settings. Heterogeneity in the data meant that most meta-analyses were underpowered as only 25 studies, further split into different groups of eye diseases and measures of vision, could be meta-analysed. The limited evidence from the meta-analyses suggests that visual field defects (four studies; RR 1.51 (95% CI 1.23, 1.85); p<0.001; I=46.79%), and contrast sensitivity (two studies; RR 1.40 (95% CI 1.08, 1.80); p=0.01, I=0.11%) and visual acuity loss (five studies; RR 1.21 (95% CI 1.02, 1.43); p=0.03, I=28.49%) may increase crash risk. The results are more inconclusive for available evidence for associations of glaucoma (five studies, RR 1.27 (95% CI 0.67, 2.42); p=0.47; I=93.48%) and cataract (two studies RR 1.15 (95% CI 0.97, 1.36); p=0.11; I=3.96%) with crashes. Driving cessation may also be linked with glaucoma (two studies; RR 1.62 (95% CI 1.20, 2.19); p<0.001, I=22.45%), age-related macular degeneration (AMD) (three studies; RR 2.21 (95% CI 1.47, 3.31); p<0.001, I=75.11%) and reduced contrast sensitivity (three studies; RR 1.30 (95% CI 1.05, 1.61); p=0.02; I=63.19%). Cataract surgery halved MVC risk (three studies; RR 0.55 (95% CI 0.34, 0.92); p=0.02; I=97.10). Ranibizumab injections (four randomised controlled trials) prolonged driving in persons with AMD.
CONCLUSION
Impaired vision identified through a variety of measures is associated with both increased MVC involvement and cessation. Cataract surgery can reduce MVC risk. Despite literature being highly heterogeneous, this review shows that detection of vision problems and appropriate treatment are critical to road safety.
PROSPERO REGISTRATION NUMBER
CRD42020172153.
Topics: Humans; Visual Acuity; Ranibizumab; Accidents, Traffic; Macular Degeneration; Vision Disorders; Cataract
PubMed: 37567751
DOI: 10.1136/bmjopen-2022-065210 -
Neuroprotective Strategies for Nonarteritic Anterior Ischemic Optic Neuropathy: A Systematic Review.Korean Journal of Ophthalmology : KJO Aug 2023Nonarteritic anterior ischemic optic neuropathy (NAION) is the second most common form of optic neuropathy. Most patients show no improvement over time. Until now, there...
PURPOSE
Nonarteritic anterior ischemic optic neuropathy (NAION) is the second most common form of optic neuropathy. Most patients show no improvement over time. Until now, there is still no definitive therapy for NAION. The available literatures on the possible treatment of NAION are quite diverse and controversial. Neuroprotection strategies have been suggested as one of the potential treatments for NAION. This review aims to critically evaluate the literature on neuroprotective strategy for NAION.
METHODS
This report was written in accordance with PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) guidelines. We performed a systematic literature search in Pubmed, Science Direct, Proquest, and Cochrane databases. Only neuroprotective agents that directly work in protecting neurons were included. The outcome of interest in this review is retinal ganglion cell density and apoptosis for animal studies and retinal nerve fiber layer thickness for human studies.
RESULTS
The systematic search identified 591 studies of which 24 met the eligibility criteria, including 21 animal studies and three human studies. Only a few of the studies evaluated the same treatments, showing how diverse neuroprotector treatments are currently being evaluated as NAION treatment. From 21 animal studies, 14 studies showed significantly higher retinal ganglion cell density (1.49- to 2.81-fold) with neuroprotective treatment compared to control group. Two of three human studies in this review had also found a beneficial effect of preserving retinal nerve fiber layer thickness in NAION patients.
CONCLUSIONS
This review suggests the potential of neuroprotection as a viable option in the quest for an effective treatment strategy for NAION. Further studies, particularly clinical studies, are necessary to establish its efficacy in NAION patients.
Topics: Animals; Humans; Optic Neuropathy, Ischemic; Optic Disk; Neuroprotection; Visual Acuity; Tomography, Optical Coherence
PubMed: 37563973
DOI: 10.3341/kjo.2022.0166 -
Journal of Optometry 2023To analyse the scientific evidence about the efficacy of Syntonic phototherapy for producing changes in visual function. (Review)
Review
PURPOSE
To analyse the scientific evidence about the efficacy of Syntonic phototherapy for producing changes in visual function.
MATERIAL AND METHODS
A systematic review was performed to obtain studies on the effects of Syntonic phototherapy on vision. A search in health science databases (Medline, Scopus, Web of Science and PsycINFO) for studies published between 1980 and 2022 was conducted in accordance with the principles of Cochrane approach. The search identified 197 articles. Only clinical studies which used the Syntonic phototherapy as a vision therapy for any visual condition were included. Clinical cases and case series were excluded. Following the inclusion criteria, 8 clinical studies met inclusion, 5 of them being pseudo-experimental studies with an equivalent control group and 3 pre-post pseudo-experimental studies. GRADE tool was used to assess the certainty of the evidence of the studies. The GRADE evidence profile for the studies through the Soft table was made to analyse data.
RESULTS
The studies analysed seven outcomes: visual symptoms, functional visual fields, visual acuity, contrast sensitivity, deviation (phoria/tropia), stereopsis and reading abilities. Finding table about results (Soft Table) showed that for all outcomes reviewed, all studies yielded very low certainty of evidence. Results revealed a lack of scientific evidence of the efficacy of Syntonic optometric phototherapy to produce changes in the visual function.
CONCLUSION
This systematic review found no consistent evidence for the efficacy of Syntonic phototherapy to cause changes in visual function. There is no scientific evidence to support its clinical use for treating any type of visual anomalies.
Topics: Humans; Phototherapy; Visual Acuity; Vision Disorders; Contrast Sensitivity; Vision, Low
PubMed: 37230932
DOI: 10.1016/j.optom.2023.03.002 -
Ophthalmology. Retina Oct 2023To compare outcomes of scleral buckle (SB), pars plana vitrectomy (PPV), and combined PPV-SB to treat rhegmatogenous retinal detachments (RRDs) with inferior retinal... (Meta-Analysis)
Meta-Analysis Review
TOPIC
To compare outcomes of scleral buckle (SB), pars plana vitrectomy (PPV), and combined PPV-SB to treat rhegmatogenous retinal detachments (RRDs) with inferior retinal breaks (IRBs).
CLINICAL RELEVANCE
Rhegmatogenous retinal detachments with IRBs are not uncommon; their management is challenging with higher risk of failure. There is no consensus about their treatment, specifically whether SB, PPV, or PPV-SB should be performed.
METHODS
Systematic review and meta-analysis. Randomized controlled trials, case-control, and prospective/retrospective series (if n > 50) in English were eligible. Medline, Embase, and Cochrane databases were searched up to January 23, 2023. Standard systematic review methods were followed. The following outcomes at 3 (± 1) and 12 (± 3) months were evaluated: number of eyes with retinal reattachment after ≥ 1 surgeries, change in best-corrected visual acuity from preoperative to postoperative levels, and number of eyes with improvement of > 10 and > 15 ETDRS letters after surgery. Authors of eligible studies were asked for individual participant data (IPD) and IPD meta-analysis was undertaken. Risk of bias was assessed using National Institutes of Health study quality assessment tools. This study was registered prospectively in PROSPERO (CRD42019145626).
RESULTS
A total of 542 studies were identified: 15 were eligible and included and 60% were retrospective. Individual participant data was obtained from 8 studies (1017 eyes). Given that only 26 patients had received SB alone, these data were not considered in the analysis. There was no evidence for differences between treatment groups (PPV versus PPV-SB) in the probability of having a flat retina at 3 or 12 months postoperatively after 1 (P = 0.067; odds ratio [OR], 0.47; P = 0.408; OR 2.55; respectively) or > 1 (OR, 0.54; P = 0.21; OR, 0.89; P = 0.926; respectively) surgery. Pars plana vitrectomy-SB showed less improvement in vision postoperatively at 3 months (estimate, 0.18; 95% confidence interval, 0.01-0.35; P = 0.044), but this difference was no longer observed at 12 months (estimate, -0.07; 95% confidence interval, -0.27, 0.13; P = 0.479).
CONCLUSION
Available evidence suggests a lack of benefit of adding SB to PPV to treat RRDs with IRBs. Evidence, however, comes mainly from retrospective series and, thus, despite the large number of eyes included, should be interpreted with caution. Further research is needed.
FINANCIAL DISCLOSURE(S)
The author(s) have no proprietary or commercial interest in any materials discussed in this article.
Topics: Humans; Retinal Detachment; Vitrectomy; Retrospective Studies; Prospective Studies; Visual Acuity; Retinal Perforations
PubMed: 37187441
DOI: 10.1016/j.oret.2023.05.006 -
Eye (London, England) Oct 2023Endothelial keratoplasty (EK) is a commonly performed transplant procedure used in the treatment of corneal endothelial dysfunction. The aim of this systematic review... (Meta-Analysis)
Meta-Analysis
BACKGROUND/OBJECTIVES
Endothelial keratoplasty (EK) is a commonly performed transplant procedure used in the treatment of corneal endothelial dysfunction. The aim of this systematic review and meta-analysis is to evaluate the differences in visual acuity outcomes, endothelial cell density (ECD) and complications between two forms of EK, ultrathin Descemet stripping automated endothelial keratoplasty (UT-DSAEK) and Descemet membrane endothelial keratoplasty (DMEK).
METHODS
A literature search of MEDLINE, Embase and Cochrane Library was conducted to identify studies reporting comparative results of UT-DSAEK versus DMEK. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement was used for search strategy. Of 141 titles, 7 studies met the inclusion criteria; best corrected visual acuity (BCVA) (LogMAR), ECD (cells/mm), and complications were compared, with all statistical analysis performed using Review Manager.
RESULTS
A total of 362 eyes were included for analysis. DMEK resulted in significantly better BCVA at 3 months (0.14 vs 0.22, p = 0.003), 6 months (0.08 vs 0.18, p = 0.005) and 1 year post-op (0.07 vs 0.14, p = 0.0005). UT-DSAEK resulted in significantly lower total complications (25.2% vs 57.3%, p = 0.0001) and rates of re-bubbling (11.0% vs 33.7%, p = 0.004). No differences were found in ECD between the two procedures (1541 vs 1605, p = 0.77).
CONCLUSIONS
DMEK results in superior visual acuity rates with quicker recovery. However, UT-DSAEK has a more favourable complication profile, particularly regarding lower rates of re-bubbling. Both are valuable options in the treatment of corneal endothelial disease and choice of procedure may depend on surgical expertise.
Topics: Humans; Descemet Membrane; Descemet Stripping Endothelial Keratoplasty; Corneal Diseases; Visual Acuity; Research Design; Retrospective Studies; Fuchs' Endothelial Dystrophy; Endothelium, Corneal
PubMed: 36934158
DOI: 10.1038/s41433-023-02467-2 -
Graefe's Archive For Clinical and... Aug 2023Retinal toxicity with long-term hydroxychloroquine (HCQ) treatment is a major concern. This systematic review aims to assess the application of optical coherence... (Meta-Analysis)
Meta-Analysis
BACKGROUND
Retinal toxicity with long-term hydroxychloroquine (HCQ) treatment is a major concern. This systematic review aims to assess the application of optical coherence tomography angiography (OCTA) to detect microvascular alterations in patients under HCQ.
METHODS
PubMed, Scopus, Web of Science, and Cochrane Library databases were systematically searched until January 14, 2023. Studies using OCTA as a primary diagnostic method to evaluate the macular microvasculature of HCQ users were included. Primary outcomes were macular vessel density (VD) and foveal avascular zone (FAZ) at the superficial (SCP) and deep (DCP) capillary plexus. Meta-analysis was performed using a random-effects model.
RESULTS
Of 211 screened abstracts, 13 were found eligible, enrolling 989 eyes from 778 patients. High-risk patients due to longer duration of treatment presented lower VD in the retinal microvasculature than those with low-risk in SCP (P = 0.02 in fovea; P = 0.004 in parafovea) and in DCP (P = 0.007 in fovea; P = 0.01 in parafovea). When compared with healthy controls, HCQ users had lower VD in both plexus-no quantitative synthesis was presented.
CONCLUSIONS
Microvascular changes were found in autoimmune patients under HCQ treatment without any documented retinopathy. However, the evidence produced so far does not allow to draw conclusion concerning the effect of drug as studies were not controlled for disease duration.
Topics: Humans; Hydroxychloroquine; Fluorescein Angiography; Retinal Vessels; Tomography, Optical Coherence; Visual Acuity; Macula Lutea
PubMed: 36884062
DOI: 10.1007/s00417-023-06023-2 -
Journal of Attention Disorders Sep 2023In this review, we examined if there is a deficit in facial recognition of emotion (FER) in children, adolescents, and adults with attention deficit hyperactivity...
OBJECTIVE
In this review, we examined if there is a deficit in facial recognition of emotion (FER) in children, adolescents, and adults with attention deficit hyperactivity disorder (ADHD).
BACKGROUND
Emotional regulation is impaired in ADHD. Although a facial emotion recognition deficit has been described in this condition, the underlying causal mechanisms remain unclear.
METHODS
The search was performed in six databases in September 2022. Studies assessing children, adolescents, or adults with isolated or comorbid ADHD that evaluated participants using a FER task were included.
RESULTS
Twelve studies out of 385 were selected, with participants ranging in age from 6 to 37.1 years. A deficit in FER specific to ADHD, or secondary to comorbid autism spectrum disorder, anxiety, and oppositional symptoms, was found.
CONCLUSIONS
There is a FER deficit in patients with ADHD. Adults showed improved recognition accuracy, reflecting partial compensation. ADHD symptoms and comorbidities appear to influence FER deficits.
Topics: Adolescent; Child; Adult; Humans; Young Adult; Attention Deficit Disorder with Hyperactivity; Facial Recognition; Autism Spectrum Disorder; Emotions; Recognition, Psychology; Facial Expression
PubMed: 36843351
DOI: 10.1177/10870547231154901 -
Neuropsychology Review Dec 2023Although attention and early associative learning in preverbal children is predominantly driven by rapid eye-movements in response to moving visual stimuli and... (Meta-Analysis)
Meta-Analysis Review
Although attention and early associative learning in preverbal children is predominantly driven by rapid eye-movements in response to moving visual stimuli and sounds/words (e.g., associating the word "bottle" with the object), the literature examining the role of visual attention and memory in ongoing vocabulary development across childhood is limited. Thus, this systematic review and meta-analysis examined the association between visual memory and vocabulary development, including moderators such as age and task selection, in neurotypical children aged 2-to-12 years, from the brain-based perspective of cognitive neuroscience. Visual memory tasks were classified according to the visual characteristics of the stimuli and the neural networks known to preferentially process such information, including consideration of the distinction between the ventral visual stream (processing more static visuo-perceptual details, such as form or colour) and the more dynamic dorsal visual stream (processing spatial temporal action-driven information). Final classifications included spatio-temporal span tasks, visuo-perceptual or spatial concurrent array tasks, and executive judgment tasks. Visuo-perceptual concurrent array tasks, reliant on ventral stream processing, were moderately associated with vocabulary, while tasks measuring spatio-temporal spans, associated with dorsal stream processing, and executive judgment tasks (central executive), showed only weak correlations with vocabulary. These findings have important implications for health professionals and researchers interested in language, as they advocate for the development of more targeted language learning interventions that include specific and relevant aspects of visual processing and memory, such as ventral stream visuo-perceptual details (i.e., shape or colour).
Topics: Child; Humans; Vocabulary; Memory; Visual Perception; Brain; Language
PubMed: 36136174
DOI: 10.1007/s11065-022-09561-4