SEARCH · PubMed Health

Results for “AI”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 235 records · Page 13Linked to original sources

PubMed2003

Comparison of Turkish Injury Scale (TIS) with the Abbreviated Injury Scale (AIS).

According to the Turkish Penal Code, Section 456, an assailant is punished in a correlation to the severity of the victim's injury. In this study, the injury scale used in Turkey in the basis code 456 is compared with Abbreviated Injury Scale (AIS). For this aim, a total of 984 cases out of the total amount reported at the Traumatology Section of the Turkish Council for Forensic Medicine were randomly selected and evaluated retrospectively. In all, 40.7% of injuries were caused by blunt trauma, whereas 59.3% were caused by a penetrating trauma. According to the Turkish Injury Scale (TIS), 40.3% of the cases were scored to be of a first degree of injury, 15.6% as second degree and 44.1% as third degree. When compared, the score points 3, 4 and 5 in the AIS were seen to be nearly equivalent to the TIS of third degree. From this point of view, in the modified AIS 91.1% of first degree of injury, 51.2% of second degree and 97.2% of third degree of injury are harmonious with TIS. Generally, 83.2% of the cases are harmonious with the AIS system. The purpose of this study is to determine what was the source of differences and to focus on particular traumatic lesions in order to determine a possible rearrangement of the Turkish Injury Scale.

Abbreviated Injury Scale↗

PubMed1984

Interaction of apolipoprotein AI from human serum high-density lipoprotein with egg yolk phosphatidylcholine.

Apolipoprotein AI from human serum high-density lipoprotein has been recombined with egg yolk lecithin from ternary complexes of detergent-lipid-protein to from homogeneous spherical particles with maximum binding of 220 mol of lipid/2 mol of AI. This complex differs from those formed when n-alkyl detergents or short chain saturated diacylphosphatidylcholines interact with AI in that the maximum hydrophobic volume accommodated by the protein is increased as the result of increased alpha-helix content. Additionally, it is shown that no interaction occurs between AI and didecanolyphosphatidylcholine or egg yolk lecithin above their thermotropic phase transitions and in the absence of single-tail amphiphiles.

Apolipoprotein A-I↗

PubMedDate not supplied

Novel atomic-level-based AI topological descriptors: application to QSPR/QSAR modeling.

Novel atomic level AI topological indexes based on the adjacency matrix and distance matrix of a graph is used to code the structural environment of each atomic type in a molecule. These AI indexes, along with Xu index, are successfully extended to compounds with heteroatoms in terms of novel vertex degree v(m), which is derived from the valence connectivity delta(v) of Kier-Hall to resolve the differentiation of heteroatoms in molecular graphs. The multiple linear regression (MLR) is used to develop the structure-property/activity models based on the modified Xu and AI indices. The efficiency of these indices is verified by high quality QSPR/QSAR models obtained for several representative physical properties and biological activities of several data sets of alcohols with a wide range of non-hydrogen atoms. The results indicate that the physical properties studied are dominated by molecular size, but other atomic types or groups have small influences dependent on the studied properties. Among all atomic types, -OH groups seem to be most important due to hydrogen-bonding interactions. On the contrary, -OH groups play a dominant role in biological activities studied, although molecular size is also an important factor. These results indicate that both Xu and AI indices are useful model parameters for QSPR/QSAR analysis of complex compounds.

Alcohols↗

PubMed2026

Reimagining research papers as interactive and reliable AI agents.

Here we introduce Paper2Agent, an automated framework that converts research papers into artificial intelligence (AI) agents. Paper2Agent transforms research output from passive artefacts into active systems that accelerate use and discovery. Conventional research papers require readers to understand and adapt the paper's code, data and methods to their work, creating barriers to dissemination and reuse. Paper2Agent addresses this challenge by converting a paper into an AI agent that functions as a virtual corresponding author, exposing its manuscript, supplementary materials, datasets, code and workflows as active, agent-native knowledge rather than static text. It analyses the paper and codebase using multiple agents to construct a model context protocol (MCP) server, then generates and runs tests to refine and increase robustness of the MCP. These paper MCPs can be connected to a chat agent (such as Claude Code) to carry out complex scientific queries through natural language while invoking tools and workflows from the paper. We demonstrate Paper2Agent's effectiveness through case studies. Paper2Agent created an agent that leveraged AlphaGenome1 to interpret genomic variants and agents based on Scanpy2 and TISSUE (transcript imputation with spatial single-cell uncertainty estimation)3 to conduct single-cell and spatial transcriptomics analyses. We validate that these agents reproduce the results of the original papers and carry out novel user queries. Paper2Agent created multiple agents that collaborate to prioritize a causal gene for psoriasis. By turning static papers into interactive AI agents, Paper2Agent introduces a paradigm for knowledge dissemination and a collaborative ecosystem of AI co-scientists.

Journal Article↗

PubMed1980

Plasma high-density lipoprotein metabolism in subjects with primary hypertriglyceridaemia: altered metabolism of apoproteins AI and AII.

1. The metabolism of the major proteins of plasma high-density lipoprotein (HDL), apoproteins AI and AII, have been studied in 10 normotriglyceridaemic subjects and in 11 hypertriglyceridaemic subjects (plasma triglyceride 4.5--25 mmol/l) by kinetic analysis of the plasma specific radioactivity versus time curves of the apoproteins after intravenous injection of autologous 125I-labelled high-density lipoprotein. 2. The specific radioactivity versus time curves of both apoproteins (followed for 14 days) were bi-exponential in all subjects. 3. The plasma apoprotein AI and AII concentrations were significantly lower in the hypertriglyceridaemic subjects than in the normotriglyceridaemic subjects. Kinetic analysis showed that this was associated with a lower rate of synthesis of apoprotein AI (P < 0.01) and a higher fractional catabolic rate of apoprotein AII (P < 0.01) in the hypertriglyceridaemic group. 4. There were no significant differences between the two groups in the synthetic rate of apoprotein AII, the fractional catabolic rate of apoprotein AI or the intravascular/extravascular distributions of the apoproteins. 5. Thus hypertriglyceridaemia appears to be frequently associated with divergent abnormalities of the metabolism of the major high density lipoprotein apoproteins.

Adult↗

PubMed1998

Sperm characteristics and zona pellucida binding in relation to field fertility of frozen-thawed semen from dairy AI bulls.

The present study examined the relationship between bull sperm characteristics immediately post-thaw and some characteristics registered after swim-up, including the ability of spermatozoa to bind to homologous zona pellucidae (ZP) in vitro, and fertility after artificial insemination (AI) of 9426 females. Frozen-thawed semen from 22 AI bulls of the Swedish Red and White Breed, represented by 43 different frozen batches (1-4 batches/bull, 2 consecutive ejaculates/batch), was examined with the aim of determining concentration, motility patterns, morphology and membrane integrity. In addition, the frozen-thawed spermatozoa were subjected to a swim-up procedure and those separated in this way were tested with two assays of sperm-binding to the ZP of homologous oocytes in vitro (ZBA), using either a relative ZBA index against a control bull of proven high fertility or absolute binding (Absolute ZBA). The correlations of the various sperm traits and 56-day non-return rates (NRR) after field AI were retrospectively examined as single traits and as combinations of traits (combined measures), including regression analysis of significant traits. Among the sperm characteristics, positively significant (p < 0.01) correlations with NRR were found for linear motility post-thawing (r = 0.45-0.59) and the concentration of motile spermatozoa after swim-up (r = 0.43-0.63). Results obtained with the absolute ZBA approach were significantly (p < 0.05) correlated with NRR (r = 0.50), whereas the correlation between NRR and the ZBA index was not significant. The use of combined measures of sperm traits, including the ability to bind to ZP, showed a stronger predictive correlation with NRR (r = 0.68-0.75), compared with single traits. The results suggest that the combined analysis of sperm linear-motility patterns, swim-up separated sperm motility and absolute ZBA can provide a valuable assessment of the fertilizing capacity of AI bull semen.

Animals↗

PubMed2000

T'ai chi ch'uan.

The Chinese practice of t'ai chi seems to be receiving increased interest in the West. This article gives a brief overview of t'ai chi, including its origins, development, principles and potential health benefits. The function of the essential elements of t'ai chi, namely the Form and chi kung are described and their potential benefits for patients and nurses are discussed. Exponents of t'ai chi believe that it has health benefits on physical, psychological and spiritual levels, thus promoting a feeling of well-being. In addition, regular practitioners are empowered to be in greater control of themselves, their health, and situations in which they find themselves.

Attention↗

PubMed2005

Paraoxonase-1 promoter polymorphism C--107T and serum apolipoprotein AI interact to modulate serum paraoxonase-1 status.

OBJECTIVES: The objective was to examine the hypothesis that modifications to paraoxonase-1 specific activity (SP, activity per unit mass peptide) could contribute to serum paraoxonase-1 status, a determinant of the clinical efficacy of the enzyme. METHODS: Enzyme activities and concentrations were determined in a large population (n=912) of patients and controls. SP were subsequently examined as a function of paraoxonase-1 gene polymorphisms, plasma lipids and lipoproteins, and physiological and pathophysiological parameters. RESULTS: Pathophysiological parameters (diabetes, metabolic syndrome, smoking, aging) did not promote variations in paraoxonase-1 SP, whilst coronary disease lowered SP (P<0.003). No serum lipid, apolipoprotein or lipoprotein component had an impact on specific activity, with the exception of apolipoprotein AI (P<0.005, both substrates). The paraoxonase-1 promoter C--107T and Q192R polymorphisms influenced SP and, together with apolipoprotein AI, were highly significant, independent determinants in regression models. There was an interaction between apolipoprotein AI and the C--107T polymorphism, which significantly modulated SP and serum paraoxonase-1 status. CONCLUSIONS: Enzyme inactivation giving rise to modulated activity per unit mass of peptide is not a major contributor to pathological effects of disease on serum paraoxonase-1 status. The C--107T polymorphism and serum apolipoprotein AI have major impacts individually on SP and also provide an example of gene-environment interaction to modulate such activities. These effects accentuate the differences between--107C and--107T allele carriers in terms of serum paraoxonase-1 status. The data underline the complexity of the factors that determine serum paraoxonase-1 status and suggest that the latter would benefit from therapeutic modulation of serum high density lipoproteins.

Apolipoprotein A-I↗

PubMed1991

Within-individual variation in serum cholesterol levels: association with DNA polymorphisms at the apolipoprotein B and AI-CIII-AIV loci in patients with peripheral arterial disease.

We have examined the association between variation at the apolipoprotein (apo) B gene and apo AI-CIII-AIV gene cluster and within-individual variation in serum cholesterol levels. Annual measurements were available over a period of 5-10 years in a group of 117 male patients with peripheral arterial disease. The overall within-individual coefficient of variation in cholesterol levels over time was 13.9%. For all patients, Restriction Fragment Length Polymorphism (RFLP) genotype at the apo B gene (XbaI and EcoRI) and apo AI-CIII-AIV gene cluster (XmnI, PstI and PvuII-CIII) had previously been determined. At the apo B locus, individuals heterozygous for either the XbaI or EcoRI RFLP showed significantly greater within-individual variability over time compared to individuals of other genotypes. At the apo AI-CIII-AIV gene cluster, individuals homozygous for the common allele of either the PstI or PvuIIA RFLPs showed the greatest within-individual variability over time but there was no difference in this estimate associated with XmnI genotype. Our observations suggest that variation at both the apo B and apo AI-CIII-AIV loci interacts with unidentified environmental factors to determine individual variability in serum cholesterol levels over time.

Apolipoprotein A-I↗

PubMed1977

Forms AI and AII DNA-dependent RNA polymerases as components of two defined pools of polymerase activity in mammalian cells.

Two species of form A (or I) RNA polymerase have been identified in eucaryotic cells and there is evidence that this alpha-amanitin-insensitive activity exists as two discrete pools: a pool of 'free' activity, which is identified by its ability to transcribe poly d(A-T) in the presence of actinomycin in vitro, and a pool of enzyme in the form of a transcription complex ('engaged') which is unaffected by inhibitors of initiation of RNA synthesis. 1. The principles underlying and the practical application of the technique used to define the pool of 'free' RNA polymerase activity have been analysed in considerable detail. On the basis of actinomycin titrations of poly[d(A-T)]dependent activity in isolated organelles, it is concluded that a pool of 'free' RNA polymerase A activity exists in mammalian nuclei which, under certain circumstances, is lost from nuclei during their isolation. The evidence presented suggests that nucleoli, resolved from nuclei by the classical sonication technique, contain form A polymerase exclusively in the transcription complex form. 2. Different techniques used to solubilise RNA polymerase activity from nucleoli are shown to give rise to different proportions of the two form A RNA polymerase species (AI and AII, as defined by their differential elution from phosphocellulose): whereas low-ionic-strength extraction gives rise to form AII, high-salt, sonication extracts contain predominantly the form AI enzyme. It is shown that the sonication technique results in the conversion of form AII to form AI. By a careful appraisal of the products of these procedures and a novel polymerase solubilisation technique, it is concluded that RNA polymerase AII is the 'engaged' form of the enzyme found in the transcription complex. 3. Making use of the finding that the 'free' form of the enzyme is lost to the cytoplasmic fraction on nuclear isolation, this activity has been characterised without the requirement for solubilisation techniques which might result in the conversion of one form to another: the 'free' species is shown to be form AI RNA polymerase. 4. These conclusions that two discrete pools of form A RNA polymerase activity contain different species of the enzyme are briefly discussed in the light of other published information concerning their subunit structures and their potential role in the expression of the ribosomal RNA coding sequences.

Amanitins↗

PubMed1989

Replenishment of AI-doses with oestrogens in physiological amounts: effect on sow prolificacy in a field trial.

Basing on results about physiological functions of seminal oestrogens in the genital tract of sows, the effects of an oestrogen replenishment to AI-doses were investigated in a field trial. Each ejaculate was split into two halves, which were either diluted to normal AI-doses (controls, n = 353) or diluted and replenished with oestrogens in physiological amounts (n = 384). Insemination by qualified technicians led to an improvement of the pregnancy rate (82.8% vs. 77.1%; p less than 0.05) and the litter size (10.8% vs. 10.3%; p less than 0.05) in favour of the oestrogen replenishment. These results partly explain the known differences in prolificacy between natural mating and AI and thus provide a basis for improvement of pig AI.

Animals↗

PubMed1984

Familial apolipoprotein AI and apolipoprotein CIII deficiency. Subclass distribution, composition, and morphology of lipoproteins in a disorder associated with premature atherosclerosis.

Lipoprotein classes isolated from the plasma of two patients with apolipoprotein AI (apo AI) and apolipoprotein CIII (apo CIII) deficiency were characterized and compared with those of healthy, age- and sex-matched controls. The plasma triglyceride values for patients 1 and 2 were 31 and 51 mg/dl, respectively, and their cholesterol values were 130 and 122 mg/dl, respectively; the patients, however, had no measurable high density lipoprotein (HDL)-cholesterol. Analytic ultracentrifugation showed that patients' S degrees f 0-20 lipoproteins possess a single peak with S degrees f rates of 7.4 and 7.6 for patients 1 and 2, respectively, which is similar to that of the controls. The concentration of low density lipoprotein (LDL) (S degrees f 0-12) particles, although within normal range (331 and 343 mg/dl for patients 1 and 2, respectively), was 35% greater than that of controls. Intermediate density lipoproteins (IDL) and very low density lipoproteins (VLDL) (S degrees f 20-400) were extremely low in the patients. HDL in the patients had a calculated mass of 15.4 and 11.8 mg/dl for patients 1 and 2, respectively. No HDL could be detected by analytic ultracentrifugation, but polyacrylamide gradient gel electrophoresis (gge) revealed that patients possessed two major HDL subclasses: (HDL2b)gge at 11.0 nm and (HDL3b)gge at 7.8 nm. The major peak in the controls, (HDL3a)gge, was lacking in the patients. Gradient gel analysis of LDL indicated that patients' LDL possessed two peaks: a major one at 27 nm and a minor one at 26 nm. The electron microscopic structure of patients' lipoprotein fractions was indistinguishable from controls. Patients' HDL were spherical and contained a cholesteryl ester core, which suggests that lecithin/cholesterol acyltransferase was functional in the absence of apo AI. The effects of postprandial lipemia (100-g fat meal) were studied in patient 1. The major changes were the appearance of a 33-nm particle in the LDL density region of 1.036-1.041 g/ml and the presence of discoidal particles (12% of total particles) in the HDL region. The latter suggests that transformation of discs to spheres may be delayed in the patient. The simultaneous deficiency of apo AI and apo CIII suggests a dual defect in lipoprotein metabolism: one in triglyceride-rich lipoproteins and the other in HDL. The absence of apo CIII may result in accelerated catabolism of triglyceride-rich particles and an increased rate of LDL formation. Additionally, absence of apo CIII would favor rapid uptake of apo E-containing remnants by liver and peripheral cells. Excess cellular cholesterol would not be removed by the reverse cholesterol transport mechanism since HDL levels are exceedingly low and thus premature atherosclerosis occurs.

Apolipoproteins A↗

PubMed1995

Expression of human lecithin-cholesterol acyltransferase in transgenic mice. Effect of human apolipoprotein AI and human apolipoprotein all on plasma lipoprotein cholesterol metabolism.

Human (Hu) lecithin-cholesterol acyltransferase (LCAT) is a key enzyme in the plasma metabolism of cholesterol. To assess the effects of increased plasma levels of LCAT, four lines of transgenic mice were created expressing a Hu LCAT gene driven by either its natural or the mouse albumin enhancer promoter. Plasma LCAT activity increased from 1.2- to 1.6-fold higher than that found in control mouse plasma. Lipid profiles, upon comparing Hu LCAT transgenics to control animals, revealed a 20 t0 60% increase in total and cholesteryl esters that were mainly present in HDL. The in vivo substrate specificity of Hu LCAT was assessed by creating animals expressing Hu apo AI + Hu LCAT (HuAI/ LCAT), Hu apo AI + Hu apo AII + Hu LCAT (HuAI/ AII/LCAT), and Hu apo AII + Hu LCAT (HuAII/LCAT). Plasma cholesterol was increased up to 4.2-fold in HuAI/ LCAT transgenic mice and twofold in the HuAI/AII/LCAT transgenic mice, compared with HuAI and HuAI/AII transgenic mice. HDL cholesteryl ester levels were increased more than twofold in both the HuAI/LCAT and HuAI/AII/LCAT mice compared with the HuAI, HuAI/AII, and HuLCAT animals. The HDL particles were predominantly larger in the HuAI/LCAT and the HuAI/AII/LCAT mice compared with those in HuAI, HuAII/LCAT, and HuLCAT animals. The increase in LCAT activity in the HuAI/LCAT and HuAI/AII/LCAT mice was associated with 62 and 27% reductions respectively, in the proportion of Hu apo AI in the pre beta-HDL fraction, when compared with HuAI and HuAI/AII transgenic mice. These data demonstrate that moderate increases in LCAT activity are associated with significant changes in lipoprotein cholesterol levels and that Hu LCAT has a significant preference for HDL containing Hu apo AI.

Animals↗

PubMed1996

Polymorphisms at the 5'-end of the apolipoprotein AI gene and severity of coronary artery disease.

Elevated HDL-cholesterol (C) and apo AI are associated with decreased coronary artery disease (CAD) risk. We determined distributions of two MspI polymorphisms of the apo AI gene, associated in other studies with increased HDL-C, among 644 patients aged < or = 65 years in relation to circulating lipids and CAD severity assessed angiographically. The rare allele distributions at both sites were in Hardy-Weinberg equilibrium in these patients but the base changes were not associated with HDL-C and apo AI levels. However, patients homozygous for the -75 bp substitution were more likely to have one or more significantly diseased vessels (> 50% luminal obstruction)(OR: 4.75, 95%CI: 1.10- 20.46) as also were patients with the rare +83 bp alleles (OR: 2.56, 95%CI: 1.13-5.81). While there was an additive effect of the two polymorphisms to have severe CAD (OR: 6.33, 95%CI: 1.33-30.02), the polymorphism at +83 bp remained significant in predicting CAD severity after adjusting for other variables in a logistic regression analysis (OR: 2.95, 95%CI: 1.26-6.90), which was also strongly associated with the positive family CAD history (P = 0.009). We conclude that patients with these base changes in this Australian coronary population do not have increased HDL-C and apo AI levels but do have more severe CAD.

Aged↗

PubMed2001

Self-reported benefits of t'ai chi practice by older women.

Self-rated health is a powerful and consistent predictor of self-care capability and health outcomes including mobility, morbidity, and mortality. Exercise is important for health and functioning of older adults. Although daily physical activity is advocated for reducing many health risks and maintaining mobility, older women are generally not heeding the message. Exercise interventions for older women should be age appropriate. T'ai chi, an ancient Chinese martial art, involves an integration of the mind and body in slow, circular movements and changes in the center of gravity. Although there is a growing body of literature on the health benefits of t'ai chi exercises, few studies focus on the self-assessment of health benefits of t'ai chi for older women. This within-participants, single-factor study of women aged 72 to 96 years resulted in statistically significant improvement in self-assessed health as well as numerous self-reported benefits after 3 months of t'ai chi exercise participation.

Activities of Daily Living↗

PubMed1999

Molecular mechanisms of androgen-independent growth of human prostate cancer LNCaP-AI cells.

The goal of this study is to investigate the molecular mechanisms of androgen-independent growth in prostate cancer. We have established an androgen-independent prostatic carcinoma LNCaP-AI (defined as a LNCaP cell line that is capable of growing in charcoal-stripped serum) from the androgen-dependent LNCaP-FGC cells. In contrast to the androgen-independent PC-3 human prostate cancer cells, LNCaP-AI cells still express a similar level of androgen receptor as their parental cells and are sensitive to androgen stimulation. Compared with the parental LNCaP-FGC cells, LNCaP-AI cells are more resistant to apoptosis induced by 12-O-tetradecanoylphorbol-13-acetate and express a much higher level of antiapoptotic gene bcl-2 and cyclin-dependent kinase inhibitor p21, which may confer an enhanced antiapoptosis phenotype. On the other hand, expression of cyclin-dependent kinase inhibitor p16 is significantly reduced in the LNCaP-AI cells, implying the release of an inhibitory effect of p16 on cell cycle progression. Taken together, our results suggest that multiple factors contribute to the development of androgen-independent growth of prostatic carcinoma cells, including enhancement of cell antiapoptosis function, release of cell cycle inhibition, and stimulation of cell proliferation by alternative signaling pathways.

Androgens↗

PubMed1983

Serum apolipoprotein AI synthesis in rat hepatocytes and its secretion as proform.

Rat hepatocytes in monolayer or suspension culture synthesize serum lipoprotein AI. It is secreted into the serum-free culture medium. Synthesis and secretion processes were studied in the presence of radiolabelled amino acids. The synthesis product of the hepatocytes and the secretion product from the medium were isolated by immunoprecipitation with a mono-specific rabbit antiserum against rat apolipoprotein AI. The intracellular and secreted products were homogeneous and identical in polyacrylamide gel electrophoresis but had reduced electrophoretic mobility as compared to native apolipoprotein AI. They were submitted to automated Edman degradation. They were present in their proform, the N-terminus of which is extended by a hexapeptide. In the presence of rat serum the proform is proteolytically transformed into the mature form of apolipoprotein AI.

Amino Acid Sequence↗

PubMed2026

Performance of AI-Based Screening Tools for Obstructive Sleep Apnea Across Apnea-Hypopnea Index Thresholds: Systematic Review and Meta-Analysis.

BACKGROUND: Obstructive sleep apnea (OSA) is highly prevalent but remains substantially underdiagnosed. Polysomnography (PSG) is the reference standard, but its cost and limited availability constrain large-scale case identification. AI-based screening tools may support risk stratification and referral prioritization, but their diagnostic accuracy across apnea-hypopnea index (AHI) thresholds remains uncertain. OBJECTIVE: This review aimed to systematically evaluate the diagnostic accuracy of AI-based OSA screening tools at AHI thresholds of ≥5, ≥15, and ≥30 events/hour, with emphasis on models using non-PSG-derived inputs. METHODS: PubMed, Embase, Scopus, and Web of Science were searched for studies published from January 1, 2016, to May 3, 2026. Eligible studies included adults evaluated for suspected OSA or recruited from population-based cohorts, assessed AI-based models intended or interpretable for OSA screening, risk prediction, or screening-oriented severity classification, used PSG as the reference standard, and reported sufficient data to construct or reconstruct 2×2 contingency tables. Diagnostic accuracy was synthesized separately by AHI threshold and input source using bivariate random-effects models, with 95% CIs and prediction intervals (PIs). Risk of bias and certainty of evidence were assessed using QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies 2) and GRADE (Grading of Recommendations Assessment, Development, and Evaluation), respectively. RESULTS: A total of 60 studies were included, of which 47 contributed data to the meta-analysis. At AHI thresholds of ≥5, ≥15, and ≥30 events/hour, pooled sensitivities were 0.94 (95% CI 0.92-0.96; 95% PI 0.71-0.99), 0.87 (95% CI 0.84-0.89; 95% PI 0.66-0.96), and 0.83 (95% CI 0.79-0.87; 95% PI 0.61-0.94), respectively; the corresponding specificities were 0.77 (95% CI 0.69-0.84; 95% PI 0.30-0.96), 0.81 (95% CI 0.75-0.85; 95% PI 0.39-0.96), and 0.91 (95% CI 0.87-0.94; 95% PI 0.55-0.99), respectively. The corresponding areas under the summary receiver operating characteristic curves were 0.943, 0.907, and 0.920. For non-PSG-derived tools, sensitivities were 0.92, 0.85, and 0.81, and specificities were 0.70, 0.74, and 0.85 at the 3 thresholds, respectively. For PSG-derived models, sensitivities were 0.96, 0.90, and 0.85, and specificities were 0.82, 0.88, and 0.96, respectively. Exploratory subgroup analyses suggested performance variation across selected study and model characteristics, including region, algorithmic framework, data source, and validation method. CONCLUSIONS: AI-based tools showed generally favorable screening performance for OSA across clinically relevant AHI thresholds, although wide PIs suggest variable performance across future comparable populations and settings. By synthesizing diagnostic accuracy across 3 AHI thresholds and distinguishing non-PSG-derived from PSG-derived models, this review extends previous broad or modality-specific reviews and offers a clinically interpretable, pathway-specific basis for linking model performance to intended use. The findings may clarify potential roles for non-PSG-derived tools in front-end screening and referral prioritization and for PSG-derived models in reduced-channel assessment and sleep-laboratory workflow support. Given substantial heterogeneity, limited external validation, and low or very low certainty of evidence, prospective validation is needed before routine implementation.

Humans↗