Erratum: Global Polarization of Ξ and Ω Hyperons in Au+Au Collisions at sqrt[s_{NN}]=200 GeV [Phys. Rev. Lett. 126, 162301 (2021)]

This corrects the article DOI: 10.1103/PhysRevLett.126.162301.

Detection of the pathological exposure of pulp using an artificial intelligence tool: a multicentric study over periapical radiographs

BACKGROUND

Introducing artificial intelligence (AI) into the medical field proved beneficial in automating tasks and streamlining the practitioners' lives. Hence, this study was conducted to design and evaluate an AI tool called Make Sure Caries Detector and Classifier (MSc) for detecting pathological exposure of pulp on digital periapical radiographs and to compare its performance with dentists.

METHODS

This study was a diagnostic, multi-centric study, with 3461 digital periapical radiographs from three countries and seven centers. MSc was built using Yolov5-x model, and it was used for exposed and unexposed pulp detection. The dataset was split into a train, validate, and test dataset; the ratio was 8-1-1 to prevent overfitting. 345 images with 752 labels were randomly allocated to test MSc. The performance metrics used to test MSc performance included mean average precision (mAP), precision, F1 score, recall, and area under receiver operating characteristic curve (AUC). The metrics used to compare the performance with that of 10 certified dentists were: right diagnosis exposed (RDE), right diagnosis not exposed (RDNE), false diagnosis exposed (FDE), false diagnosis not exposed (FDNE), missed diagnosis (MD), and over diagnosis (OD).

RESULTS

MSc achieved a performance of more than 90% in all metrics examined: an average precision of 0.928, recall of 0.918, F1-score of 0.922, and AUC of 0.956 (P<.05). The results showed a higher mean of 1.94 for all right (correct) diagnosis parameters in MSc group, while a higher mean of 0.64 for all wrong diagnosis parameters in the dentists group (P<.05).

CONCLUSIONS

The designed MSc tool proved itself reliable in the detection and differentiating between exposed and unexposed pulp in the internally validated model. It also showed a better performance for the detection of exposed and unexposed pulp when compared to the 10 dentists' consensus.

Image findings of anti-neutrophil cytoplasmic antibody-associated vasculitis involving the skull base

AIM

To investigate computed tomography (CT) and magnetic resonance imaging (MRI) features of skull bases involving anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitides (AAV).

MATERIALS AND METHODS

A retrospective review was undertaken to identify an institutional historical cohort of 17 patients with confirmed AAV who underwent CT or MRI and had skull base involvement between 2002 and 2021. Two radiologists reviewed the extent and features of the lesions, bone changes, and other MRI findings.

RESULTS

A total of 17 patients (12 men; mean age ± standard deviation, 46.5 ± 17.1 years) were selected. AAV presented as infiltrative lesions with involvement at various sites. Most cases involved the paranasal sinuses (PNS; 88%, 15/17), nasopharynx (88%, 15/17), pterygopalatine fossa (82%, 14/17), and parapharyngeal space (82%, 14/17), frequently accompanied by mucosal irregularity of the PNS and nasopharynx (71%, 12/17). Central skull base and temporal bone involvement were seen in 53% (9/17) and 38% (6/16) of cases, respectively. On T1-weighted imaging (WI) and T2WI MRI, all lesions (15/15) showed predominant signal iso-intensity to grey matter.

CONCLUSIONS

Although radiological findings of AAV are non-specific and skull base involvement is less common, AAV may be considered if infiltrative lesions predominantly involving the PNS, nasopharynx, pterygopalatine fossa, and parapharyngeal space with combined bone changes of skull base are seen.

Measurements of the Elliptic and Triangular Azimuthal Anisotropies in Central ^{3}He+Au, d+Au and p+Au Collisions at sqrt[s_{NN}]=200 GeV

The elliptic (v_{2}) and triangular (v_{3}) azimuthal anisotropy coefficients in central ^{3}He+Au, d+Au, and p+Au collisions at sqrt[s_{NN}]=200  GeV are measured as a function of transverse momentum (p_{T}) at midrapidity (|η|<0.9), via the azimuthal angular correlation between two particles both at |η|<0.9. While the v_{2}(p_{T}) values depend on the colliding systems, the v_{3}(p_{T}) values are system independent within the uncertainties, suggesting an influence on eccentricity from subnucleonic fluctuations in these small-sized systems. These results also provide stringent constraints for the hydrodynamic modeling of these systems.

Neuropsychological, neuroimaging and autopsy findings of butane encephalopathy

BACKGROUND

Butane is an aliphatic hydrocarbon used in various commercial products. While numerous reports of sudden cardiac-related deaths from butane inhalation have been described, butane-associated acute encephalopathy has rarely been reported.

CASE PRESENTATION

A 38-year-old man presented with cognitive dysfunction after butane gas inhalation. Neuropsychological test results showed impairments in verbal and visual memory, and frontal executive function. Diffusion weighted MRI revealed symmetric high-signal changes in the bilateral hippocampus and globus pallidus. FDG-PET demonstrated decreased glucose metabolism in the bilateral precuneus and occipital areas and the left temporal region. At the 8-month follow-up, he showed still significant deficits in memory and frontal functions. Diffuse cortical atrophy with white matter hyperintensities and extensive glucose hypometabolism were detected on follow-up MRI and FDG-PET, respectively. Brain autopsy demonstrated necrosis and cavitary lesions in the globus pallidus.

CONCLUSIONS

Only a few cases of butane encephalopathy have been reported to date. Brain lesions associated with butane encephalopathy include lesions in the bilateral thalamus, insula, putamen, and cerebellum. To the best of our knowledge, this is the first report on bilateral hippocampal and globus pallidal involvement in acute butane encephalopathy. The pathophysiology of central nervous system complications induced by butane intoxication is not yet fully understood. However, the direct toxic effects of butane or anoxic injury secondary to cardiac arrest or respiratory depression have been suggested as possible mechanisms of edematous changes in the brain after butane intoxication.

Observation of Directed Flow of Hypernuclei _{Λ}^{3}H and _{Λ}^{4}H in sqrt[s_{NN}]=3 GeV Au+Au Collisions at RHIC

We report here the first observation of directed flow (v_{1}) of the hypernuclei _{Λ}^{3}H and _{Λ}^{4}H in mid-central Au+Au collisions at sqrt[s_{NN}]=3  GeV at RHIC. These data are taken as part of the beam energy scan program carried out by the STAR experiment. From 165×10^{6} events in 5%-40% centrality, about 8400 _{Λ}^{3}H and 5200 _{Λ}^{4}H candidates are reconstructed through two- and three-body decay channels. We observe that these hypernuclei exhibit significant directed flow. Comparing to that of light nuclei, it is found that the midrapidity v_{1} slopes of _{Λ}^{3}H and _{Λ}^{4}H follow baryon number scaling, implying that the coalescence is the dominant mechanism for these hypernuclei production in the 3 GeV Au+Au collisions.

Beam Energy Dependence of Triton Production and Yield Ratio (N_{t}×N_{p}/N_{d}^{2}) in Au+Au Collisions at RHIC

We report the triton (t) production in midrapidity (|y|<0.5) Au+Au collisions at sqrt[s_{NN}]=7.7-200  GeV measured by the STAR experiment from the first phase of the beam energy scan at the Relativistic Heavy Ion Collider. The nuclear compound yield ratio (N_{t}×N_{p}/N_{d}^{2}), which is predicted to be sensitive to the fluctuation of local neutron density, is observed to decrease monotonically with increasing charged-particle multiplicity (dN_{ch}/dη) and follows a scaling behavior. The dN_{ch}/dη dependence of the yield ratio is compared to calculations from coalescence and thermal models. Enhancements in the yield ratios relative to the coalescence baseline are observed in the 0%-10% most central collisions at 19.6 and 27 GeV, with a significance of 2.3σ and 3.4σ, respectively, giving a combined significance of 4.1σ. The enhancements are not observed in peripheral collisions or model calculations without critical fluctuation, and decreases with a smaller p_{T} acceptance. The physics implications of these results on the QCD phase structure and the production mechanism of light nuclei in heavy-ion collisions are discussed.

Characterization of human induced pluripotent stem cells line (PNUSCRi004-A) from a Parkinson's disease patient carrying L483P, A495P and V499V mutations

The hiPSC line was generated from peripheral blood mononuclear cells (PBMCs) collected from a female patient with young onset Parkinson's disease (PD), carrying on heterozygous c.1448 T > C (L483P), c1483 G > C (A495P) and c.1497 G > C (V499V) mutations in the GBA gene. The PBMCs was reprogrammed into an induced pluripotent stem cell (iPSC) line (GBA PD8 or PNUSCRi004-A hiPSCs) using non-integrative Sendai virus. The cell line, PNUSCRi004-A displayed a normal karyotype and expression of pluripotency markers capable of producing derivatives of three germ layers (Ectoderm, Endoderm and Mesoderm).

Measurement of Sequential ϒ Suppression in Au+Au Collisions at sqrt[s_{NN}]=200 GeV with the STAR Experiment

We report on measurements of sequential ϒ suppression in Au+Au collisions at sqrt[s_{NN}]=200  GeV with the STAR detector at the Relativistic Heavy Ion Collider (RHIC) through both the dielectron and dimuon decay channels. In the 0%-60% centrality class, the nuclear modification factors (R_{AA}), which quantify the level of yield suppression in heavy-ion collisions compared to p+p collisions, for ϒ(1S) and ϒ(2S) are 0.40±0.03(stat)±0.03(sys)±0.09(norm) and 0.26±0.08(stat)±0.02(sys)±0.06(norm), respectively, while the upper limit of the ϒ(3S) R_{AA} is 0.17 at a 95% confidence level. This provides experimental evidence that the ϒ(3S) is significantly more suppressed than the ϒ(1S) at RHIC. The level of suppression for ϒ(1S) is comparable to that observed at the much higher collision energy at the Large Hadron Collider. These results point to the creation of a medium at RHIC whose temperature is sufficiently high to strongly suppress excited ϒ states.

Beam Energy Dependence of Fifth- and Sixth-Order Net-Proton Number Fluctuations in Au+Au Collisions at RHIC

We report the beam energy and collision centrality dependence of fifth and sixth order cumulants (C_{5}, C_{6}) and factorial cumulants (κ_{5}, κ_{6}) of net-proton and proton number distributions, from center-of-mass energy (sqrt[s_{NN}]) 3 GeV to 200 GeV Au+Au collisions at RHIC. Cumulant ratios of net-proton (taken as proxy for net-baryon) distributions generally follow the hierarchy expected from QCD thermodynamics, except for the case of collisions at 3 GeV. The measured values of C_{6}/C_{2} for 0%-40% centrality collisions show progressively negative trend with decreasing energy, while it is positive for the lowest energy studied. These observed negative signs are consistent with QCD calculations (for baryon chemical potential, μ_{B}≤110  MeV) which contains the crossover transition range. In addition, for energies above 7.7 GeV, the measured proton κ_{n}, within uncertainties, does not support the two-component (Poisson+binomial) shape of proton number distributions that would be expected from a first-order phase transition. Taken in combination, the hyperorder proton number fluctuations suggest that the structure of QCD matter at high baryon density, μ_{B}∼750  MeV at sqrt[s_{NN}]=3  GeV is starkly different from those at vanishing μ_{B}∼24  MeV at sqrt[s_{NN}]=200  GeV and higher collision energies.

Optimization of Gelatin Extracting Condition from Korean Native Black Goat Skin and Quality Comparison with Commercial Gelatin

The practical use of Korean native black goat skin as a source of gelatin extraction is limited. The objective of this study was to optimize the extraction temperature and time of gelatin from Korean native black goat skin, and to compare the quality characteristics of goat skin gelatin and other commercial gelatin products. Response surface methodology was applied to optimize the extraction temperature and time of gelatin obtained from native Korean black goat skin. The effects of temperature (50°C-70°C) and time (2-4 h) on extraction yield and gel strength were investigated using a face-centered central composite design with 13 experiments. Gelatin extraction from Korean native black goat skin was prepared through the serial processes of alkali pre-treatment, bleaching, neutralization, hot-water extraction, and freeze-drying. Using the optimization plot of Minitab software, the optimized conditions for extracting temperature and time of goat skin gelatin were 59.49°C and 3.03 h, and the optimized values of extraction yield and gel strength were 12.52% and 263.37 g, respectively. Based on a quality comparison of goat skin gelatin with commercial gelatin, the pH value of gelatin extracted from Korean native black goat skin was 5.57. The color of gelatin extracted from Korean native black goat skin was darker than that of commercial gelatin (p<0.05). Higher emulsifying properties and gel strength of goat skin gelatin were observed when compared to those of commercial gelatin (p<0.05). Therefore, the results of this study indicate that Korean native black goat skin may be a valuable source for gelatin extraction.

Efficacy of Silkworm Pupae Extract on Muscle Strength and Mass in Middle-Aged and Older Individuals: A Randomized, Double-Blind, Placebo-Controlled Trial

OBJECTIVES

We investigated the efficacy and safety of silkworm pupae extract (SWP) consumption for 12 weeks on muscle mass and strength in middle-aged and older individuals with relatively low skeletal muscle mass who do regular low-intensity exercise.

DESIGN

A randomized double-blinded placebo-controlled trial.

PARTICIPANTS

The study was conducted with 54 participants with relatively low skeletal muscle mass (SMM) (64.4 ± 6.1 years; body mass index, 23.8 ± 2.4 kg/m2).

INTERVENTION AND MEASUREMENTS

Participants were randomly assigned to one of two groups: 1000 mg of SWP/day plus regular exercise (SWP group, n=27) or placebo plus regular exercise (placebo group, n=27). All participants were required to engage in 30-60 minutes/day of walking for ≥3 days/week for 12 weeks. The primary outcome was knee extension/flexion strength (Nm), measured at the velocity of 60°/s. Secondary outcomes included body composition, biomarkers (creatine kinase and creatinine), handgrip strength, and quality of life questionnaire.

RESULTS

Both the intention-to-treat (ITT) and per-protocol (PP) analyses revealed no significant impact of SWP on knee strength compared to the placebo group over 12 weeks. On the other hand, the SWP group had significantly greater increases in right-handgrip strength by 1.94 kg (95% CI: 0.08-3.79; p = 0.041) and left-handgrip strength by 1.83 kg (0.25-3.41; p = 0.024) compared to the placebo group in the ITT population, after 12 weeks. Moreover, in the PP population, the SWP group revealed an even greater increase in right-handgrip strength by 2.07 kg (0.15-3. 98; p = 0.035) and left-handgrip strength by 2.21 kg (0.60-3.83; p = 0.008) for the 12-week period. However, this study resulted in a failure to detect significant differences in the body composition, biomarkers, quality of life questionnaire, physical activity, and caloric intake between the groups. None of the participants in the SWP group experienced any significant adverse events. In the placebo group, two participants experienced urticaria and allergic side effects, leading to their withdrawal from the study and two exhibited elevated levels of liver enzyme and increased diastolic blood pressure, respectively at 12 weeks.

CONCLUSION

SWP, in addition to low-intensity exercise, may enhance handgrip strengths in middle-aged and older adults with relatively lower SMM. Future studies need to use a large sample size over longer periods to validate our findings. This trial was registered at clinicaltrials.gov as NCT04994054.

Collision-System and Beam-Energy Dependence of Anisotropic Flow Fluctuations

Elliptic flow measurements from two-, four-, and six-particle correlations are used to investigate flow fluctuations in collisions of U+U at sqrt[s_{NN}]=193  GeV, Cu+Au at sqrt[s_{NN}]=200  GeV and Au+Au spanning the range sqrt[s_{NN}]=11.5-200  GeV. The measurements show a strong dependence of the flow fluctuations on collision centrality, a modest dependence on system size, and very little if any, dependence on particle species and beam energy. The results, when compared to similar LHC measurements, viscous hydrodynamic calculations, and trento model eccentricities, indicate that initial-state-driven fluctuations predominate the flow fluctuations generated in the collisions studied.

Generation of Parkinson's disease patient-derived human induced pluripotent stem cells line (PNUSCRi001-A) carrying a N227S mutation in GBA gene

The hiPSC line was generated from peripheral blood mononuclear cells (PBMCs) collected by a male patient with young onset Parkinson's disease, carrying on heterozygous c.680 A > G (N227S) mutation in the GBA gene. The PBMCs was reprogrammed into an induced pluripotent stem cell (iPSC) line (PNUSCRi001-A hiPSCs) using non-integrative sendai virus. The hiPSC line, PNUSCRi001-A displayed a normal karyotype and the Expression of pluripotency markers that is capable of producing derivatives of three germ layers (Ectoderm, Endoderm and Mesoderm).

Clinical, electrophysiological, and genetic characteristics of cerebrotendinous xanthomatosis in South Korea

Cerebrotendinous xanthomatosis (CTX) is an autosomal recessive lipid storage disorder caused by 27-hydroxylase deficiency. We report the clinical characteristics of six Korean CTX patients. The median age of onset was 22.5 years, the median age at diagnosis was 42 years, and the diagnostic delay was 18.1 years. The most common clinical symptoms were tendon xanthoma and spastic paraplegia. Four of five patients exhibited latent central conduction dysfunction. All patients carried the same mutation in CYP27A1 (c.1214 G>A [p.R405Q]). CTX is a treatable neurodegenerative disorder; however, our results revealed that patients with CTX in Korea might receive the diagnosis after a prolonged delay.    .

Postoperative evaluation of modified abductor pollicis longus suspensionplasty using two anchors: Preliminary results

Among the various surgical treatments for basal joint arthritis, we modified abductor pollicis longus (APL) suspensionplasty by using two anchors. We hypothesized that this modification would prevent not only subsidence but also lateral migration of the first metacarpal. Thirteen thumbs that underwent APL suspensionplasty were investigated. Mean follow-up was 25 months (range, 12-69 months). Clinical and radiographic parameters were assessed preoperatively and postoperatively. Progression of subsidence (trapezial space) and lateral migration of the first metacarpal were investigated sequentially and with the thumb abducted or adducted against stress in a specifically designed mold. Clinical improvement was assessed by decrease in QuickDASH score (from 49.6 to 19.7). The immediate postoperative trapezial space decreased significantly by 39% (p = 0.003), and lateral migration was improved significantly by 14% (p = 0.007). At final follow-up, subsidence and lateral migration had not significantly progressed (p = 0.059 and 0.278, respectively). Under stress, the trapezial space ratio decreased significantly with the thumb in abduction (from 0.63 to 0.59, p = 0.011). APL suspensionplasty using two anchors in patients with basal joint arthritis maintained the position of the first metacarpal bone, and especially lateral subluxation. LEVEL OF EVIDENCE: IV.

Human induced pluripotent stem cells line (PNUSCRi002-A) from a patient with Parkinson's disease carrying a R159W mutation in the GBA gene

Mutation in the glucocerebrosidase encoding gene 1 (GBA) is one of the most frequent causes of Parkinson's disease (PD). Herein, we obtained peripheral blood mononuclear cells (PBMCs) from a patient with PD with a heterozygous c.475C > T (p.R159W) mutation in the GBA gene, and generated an induced pluripotent stem cell (iPSC) line (GBA PD9 or PNUSCRi002-A hiPSCs) using a non-integrative Sendai virus. The iPSC line expressed pluripotency markers (OCT4, NANOG, SSEA-4, TRA-1-60) and displayed differentiation properties in the three germ layers (ectoderm, endoderm, and mesoderm). Additionally, the patient had a normal karyotype.

Automated volumetric determination of high R2 * regions in substantia nigra: A feasibility study of quantifying substantia nigra atrophy in progressive supranuclear palsy

The establishment of an unbiased protocol for the automated volumetric measurement of iron-rich regions in the substantia nigra (SN) is clinically important for diagnosing neurodegenerative diseases exhibiting midbrain atrophy, such as progressive supranuclear palsy (PSP). This study aimed to automatically quantify the volume and surface properties of the iron-rich 3D regions in the SN using the quantitative MRI-R₂ ^* map. Three hundred and sixty-seven slices of R₂ ^* map and susceptibility-weighted imaging (SWI) at 3-T MRI from healthy control (HC) individuals and Parkinson's disease (PD) patients were used to train customized U-net++ convolutional neural network based on expert-segmented masks. Age- and sex-matched participants were selected from HC, PD, and PSP groups to automate the volumetric determination of iron-rich areas in the SN. Dice similarity coefficient values between expert-segmented and detected masks from the proposed network were 0.91 ± 0.07 for R₂ ^* maps and 0.89 ± 0.08 for SWI. Reductions in iron-rich SN volume from the R₂ ^* map (SWI) were observed in PSP with area under the receiver operating characteristic curve values of 0.96 (0.89) and 0.98 (0.92) compared with HC and PD, respectively. The mean curvature of the PSP showed SN deformation along the side closer to the red nucleus. We demonstrated the automated volumetric measurement of iron-rich regions in the SN using deep learning can quantify the SN atrophy in PSP compared with PD and HC.

Cluster Analysis of DSC MRI, Dynamic Contrast-Enhanced MRI, and DWI Parameters Associated with Prognosis in Patients with Glioblastoma after Removal of the Contrast-Enhancing Component: A Preliminary Study

BACKGROUND AND PURPOSE

No report has been published on the use of DSC MR imaging, DCE MR imaging, and DWI parameters in combination to create a prognostic prediction model in glioblastoma patients. The aim of this study was to develop a machine learning-based model to find preoperative multiparametric MR imaging parameters associated with prognosis in patients with glioblastoma. Normalized CBV, volume transfer constant, and ADC of the nonenhancing T2 high-signal-intensity lesions were evaluated using K-means clustering.

MATERIALS AND METHODS

A total of 142 patients with glioblastoma who underwent preoperative MR imaging and total resection were included in this retrospective study. From the normalized CBV, volume transfer constant, and ADC maps, the parametric data were sorted using the K-means clustering method. Patients were divided into training and test sets (ratio, 1:1), and the optimal number of clusters was determined using receiver operating characteristic analysis. Kaplan-Meier survival analysis and log-rank tests were performed to identify potential parametric predictors. A multivariate Cox proportional hazard model was conducted to adjust for clinical predictors.

RESULTS

The nonenhancing T2 high-signal-intensity lesions were divided into 6 clusters. The cluster (class 4) with the relatively low normalized CBV and volume transfer constant value and the lowest ADC values was most associated with predicting glioblastoma prognosis. The optimal cutoff of the class 4 volume fraction of nonenhancing T2 high-signal-intensity lesions predicting 1-year progression-free survival was 9.70%, below which the cutoff was associated with longer progression-free survival. Two Kaplan-Meier curves based on the cutoff value showed a statistically significant difference (P = .037). When we adjusted for all clinical predictors, the cluster with the relatively low normalized CBV and volume transfer constant values and the lowest ADC value was an independent prognostic marker (hazard ratio, 3.04; P = .048). The multivariate Cox proportional hazard model showed a concordance index of 0.699 for progression-free survival.

CONCLUSIONS

Our model showed that nonenhancing T2 high-signal-intensity lesions with the relatively low normalized CBV, low volume transfer constant values, and the lowest ADC values could serve as useful prognostic imaging markers for predicting survival outcomes in patients with glioblastoma.

Short and long-term inequity in outpatient medical use by the type of medical institutions in Korea

Background

Many countries agree with the horizontal equity that medical resources should be allocated according to medical needs, regardless of income. Although the short-term equity index calculated through cross-sectional data doesn't reflect the dynamics of individual income and medical use, it can be supplemented by the long-term equity index using panel data. Koreans tend to choose expensive but highly specialized services without considering their medical needs because they are free to choose service providers. This study aims to empirically examine how the patterns of outpatient medical use that are not based on medical needs differ in terms of short- and long-term equity for each type of medical institution.

Methods

Using Korea Health Panel Survey(2014-2018), the equity of outpatient medical use(number of visits, medical expenses) of 10,244 people was measured by type of medical institution (tertiary general hospital, general hospital, hospital, clinic, and dentist). Wagstaff&van Doorslaer(2000)’s tool and Jones & Lopez-Nicolas(2004)’s tool were used to calculate the short and long-term horizontal equity index(HI), and mobility index(MI) to compare short and long-term inequity.

Results

In tertiary general hospitals and dentists, there were short and long-term pro-rich inequalities(HI &gt; 0, p &lt; 0.05). As a result of comparison, long-term inequality was greater in the number of visits (MI &lt; 0), while inequality was easing in the long-term in medical expenses(MI &gt; 0) in tertiary general hospitals. In dentists, long-term inequality was less than short-term inequality in both the number of visits and medical expenses (MI &gt; 0).

Conclusions

The short-term equity index is likely to underestimate or overestimate inequity in our society, so a long-term perspective is needed. Inequality patterns for each type of medical institution should be considered in healthcare reforms for fair distribution of medical resources.

Key messages

• Short-term equity index differs from the long-term equity index in outpatient medical use.

• The pattern of short and long-term equity indices may differ by type of medical institutions.

Prediction of atrial fibrillation in patients with embolic stroke with undetermined source using electrocardiogram deep learning algorithm and clinical risk factors

Background

Combining the artificial intelligence algorithm with the known clinical risk factors may provide enhanced accuracy for prediction of the hidden atrial fibrillation (AF) in patients with embolic stroke with undetermined source (ESUS).

Purpose

We aimed to develop enhanced prediction models for AF with deep learning algorithm (DLA) and clinical predictors in patients with ESUS. The DLA was created to identify the patients with paroxysmal AF based on their electrocardiograms (ECG) during sinus rhythm.

Methods

We analyzed the 221 patients who underwent insertable cardiac monitor (ICM) for AF detection following ESUS. The DLA was previously developed with sinus rhythm ECGs of 10,605 paroxysmal AF patients and 50,522 non-AF patients. The convolutional neural network was used for the DLA. The primary endpoint was defined as any AF episode lasting over 5 min by ICM. The atrial ectopic burden (AEB) was calculated as the percentage of the number of conducted QRS by atrial ectopy on Holter monitoring.

Results

AF (≥5 min) was detected by ICM in 32 patients (14.5%) during follow-up period of 15.1±8.6 months. AF patients had higher AEB (0.199% vs 0.023%, p&lt;0.001), larger left atrial diameters (LAD, 41.2 mm vs 35.7 mm, p&lt;0.001), and larger left atrial volume index (LAVI, 46.4 ml/m2 vs 32.3 ml/m2, p&lt;0.001) than those without AF. The means of calculated probabilities of AF by DLA were higher in patients with AF than those without AF (63.8% vs 40.2%, p&lt;0.001). In the receiver operating characteristic curve analysis, the areas under the curve (AUC) were the highest in DLA (0.824) followed by AEB (0.784), LAVI (0.780), and LAD (0.768). The multivariable model with AEB, LAVI, and DLA demonstrated excellent prediction accuracy for paroxysmal AF (AUC: 0.902, Figure 1)

Conclusions

In patients with ESUS, the DLA outperformed other clinical risk factors for prediction of AF. Combining DLA with AEB, LAD and LAVI could is a potential useful tool to predict AF in ESUS patients.

Funding Acknowledgement

Type of funding sources: None.

Deep learning-based electrocardiogram analysis detecting paroxysmal atrial fibrillation during sinus rhythm in patients with cryptogenic stroke: validation study using implantable cardiac monitoring

Background

Atrial fibrillation (AF) is the most cause of cardioembolic source causing cryptogenic stroke. In these, anticoagulation therapy could reduce recurrence of stroke. However, paroxysmal AF would not be detected even by 24 hours Holter monitoring. Deep learning-based electrocardiogram (ECG) analysis models were recently developed to detect AF during sinus rhythm.

Purpose

We aimed to develop a deep learning algorithm (DLA) to detect AF during sinus rhythm and validate the model in patients with cryptogenic stroke who underwent implantable cardiac monitoring (ICM) to diagnose paroxysmal AF.

Methods

This cohort study involved three hospitals (A, B, and C). We developed a DLA to detect AF using sinus rhythm 10 s 12-lead ECG. We included adult patients aged ≥18 years from hospital A and B. We used development data from AF adult patients who had at least one atrial fibrillation rhythm in the study period (Jan 2016 to Dec 2021) and non-AF patients who had no reference to AF in the ECG and electronic medical record. DLA was based on convolutional neural network (CNN) using 10 s 12-lead. For external validation, the ECGs from 217 patients (hospital C) with cryptogenic stroke who underwent ICM were analyzed by using the DLA for validating the accuracy in the real-world clinical situations.

Results

We included 10,605 AF adult patients and 50,522 non-AF patients as development data. During the internal validation, the area under the curve (AUC) of the final DLA based on CNN was 0.793 (95% Confidence interval 0.778–0.807). In external validation data from cryptogenic stroke patients, the mean ICM duration was 15.1 months, and AF &gt;5 mins was detected in 32 patients (14.5%). The diagnostic accuracy of DLA was 0.793 to detect AF during sinus rhythm, and AUC was 0.824. The sensitivity, specificity, positive predictive value, and negative predictive value of the model were 0.844, 0.784, 0.403, and 0.967, respectively, which outperformed other conventional predictive methods based on clinical factors, such as CHARGE-AF, C2hest, and HATCH.

Conclusions

In this study, DLA accurately detected paroxysmal AF using 12-leads normal sinus rhythm ECG in patients with cryptogenic stroke and outperformed the conventional models. The DLA could be used as a screening tool to identify the cause of stroke in the future.

Funding Acknowledgement

Type of funding sources: None.