Applying masked autoencoder-based self-supervised learning for high-capability vision transformers of electrocardiographies.

The generalization of deep neural network algorithms to a broader population is an important challenge in the medical field. We aimed to apply self-supervised learning using masked autoencoders (MAEs) to improve the performance of the 12-lead electrocardiography (ECG) analysis model using limited ECG data. We pretrained Vision Transformer (ViT) models by reconstructing the masked ECG data with MAE. We fine-tuned this MAE-based ECG pretrained model on ECG-echocardiography data from The University of Tokyo Hospital (UTokyo) for the detection of left ventricular systolic dysfunction (LVSD), and then evaluated it using multi-center external validation data from seven institutions, employing the area under the receiver operating characteristic curve (AUROC) for assessment. We included 38,245 ECG-echocardiography pairs from UTokyo and 229,439 pairs from all institutions. The performances of MAE-based ECG models pretrained using ECG data from UTokyo were significantly higher than that of other Deep Neural Network models across all external validation cohorts (AUROC, 0.913-0.962 for LVSD, p < 0.001). Moreover, we also found improvements for the MAE-based ECG analysis model depending on the model capacity and the amount of training data. Additionally, the MAE-based ECG analysis model maintained high performance even on the ECG benchmark dataset (PTB-XL). Our proposed method developed high performance MAE-based ECG analysis models using limited ECG data.

Genetic Risk Stratification of Primary Open-Angle Glaucoma in Japanese Individuals.

PURPOSE: To assess the impact of genetic risk estimation for primary open-angle glaucoma (POAG) in Japanese individuals. DESIGN: Cross-sectional analysis. PARTICIPANTS: Genetic risk scores (GRSs) were constructed based on a genome-wide association study (GWAS) of POAG in Japanese people. A total of 3625 Japanese individuals, including 1191 patients and 2434 controls (Japanese Tohoku), were used for the model selection. We also evaluated the discriminative accuracy of constructed GRSs in a dataset comprising 1034 patients and 1147 controls (the Japan Glaucoma Society Omics Group [JGS-OG] and the Genomic Research Committee of the Japanese Ophthalmological Society [GRC-JOS]) and 1900 participants from a population-based study (Hisayama Study). METHODS: We evaluated 2 types of GRSs: polygenic risk scores using the pruning and thresholding procedure and a GRS using variants associated with POAG in the GWAS of the International Glaucoma Genetics Consortium (IGGC). We selected the model with the highest areas under the receiver operating characteristic curve (AUC). In the population-based study, we evaluated the correlations between GRS and ocular measurements. MAIN OUTCOME MEASURE: Proportion of patients with POAG after stratification according to the GRS. RESULTS: We found that a GRS using 98 variants, which showed genome-wide significance in the IGGC, showed the best discriminative accuracy (AUC, 0.65). In the Japanese Tohoku, the proportion of patients with POAG in the top 10% individuals was significantly higher than that in the lowest 10% (odds ratio [OR], 6.15; 95% confidence interval [CI], 4.35-8.71). In the JGS-OG and GRC-JOS, we confirmed similar impact of POAG GRS (AUC, 0.64; OR [top vs. bottom decile], 5.81; 95% CI, 3.79-9.01). In the population-based study, POAG prevalence was significantly higher in the top 20% individuals of the GRS compared with the bottom 20% (9.2% vs. 5.0%). However, the discriminative accuracy was low (AUC, 0.56). The POAG GRS was correlated positively with intraocular pressure (r = 0.08: P = 4.0 × 10-4) and vertical cup-to-disc ratio (r = 0.11; P = 4.0 × 10-6). CONCLUSIONS: The GRS showed moderate discriminative accuracy for POAG in the Japanese population. However, risk stratification in the general population showed relatively weak discriminative performance. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Y-shaped graft replacement of an isolated innominate artery aneurysm via the transmanubrial approach.

We report the rare case of an obese woman with a 32-mm isolated innominate artery aneurysm. Due to the deep location of the aneurysm within the thoracic inlet, we selected a transmanubrial osteomuscular sparing approach to obtain a clear field of vision. During the aneurysmectomy and reconstruction with a Y-shaped graft, we performed a temporary bypass using a contralateral leg of the graft to the right common carotid artery. Through these techniques, we successfully achieved complete resection and reconstruction of the aneurysm without the need for cardiopulmonary bypass.

The potential of the transformer-based survival analysis model, SurvTrace, for predicting recurrent cardiovascular events and stratifying high-risk patients with ischemic heart disease.

INTRODUCTION: Ischemic heart disease is a leading cause of death worldwide, and its importance is increasing with the aging population. The aim of this study was to evaluate the accuracy of SurvTrace, a survival analysis model using the Transformer-a state-of-the-art deep learning method-for predicting recurrent cardiovascular events and stratifying high-risk patients. The model's performance was compared to that of a conventional scoring system utilizing real-world data from cardiovascular patients. METHODS: This study consecutively enrolled patients who underwent percutaneous coronary intervention (PCI) at the Department of Cardiovascular Medicine, University of Tokyo Hospital, between 2005 and 2019. Each patient's initial PCI at our hospital was designated as the index procedure, and a composite of major adverse cardiovascular events (MACE) was monitored for up to two years post-index event. Data regarding patient background, clinical presentation, medical history, medications, and perioperative complications were collected to predict MACE. The performance of two models-a conventional scoring system proposed by Wilson et al. and the Transformer-based model SurvTrace-was evaluated using Harrell's c-index, Kaplan-Meier curves, and log-rank tests. RESULTS: A total of 3938 cases were included in the study, with 394 used as the test dataset and the remaining 3544 used for model training. SurvTrace exhibited a mean c-index of 0.72 (95% confidence intervals (CI): 0.69-0.76), which indicated higher prognostic accuracy compared with the conventional scoring system's 0.64 (95% CI: 0.64-0.64). Moreover, SurvTrace demonstrated superior risk stratification ability, effectively distinguishing between the high-risk group and other risk categories in terms of event occurrence. In contrast, the conventional system only showed a significant difference between the low-risk and high-risk groups. CONCLUSION: This study based on real-world cardiovascular patient data underscores the potential of the Transformer-based survival analysis model, SurvTrace, for predicting recurrent cardiovascular events and stratifying high-risk patients.

Collapse of zone 0 landing TEVAR (Najuta) and the development of higher brain dysfunction.

Although thoracic endovascular repair (TEVAR)-specific complications often develop, stent-graft collapse is a rare, but fatal complication that requires attention. A 62-year-old male underwent TEVAR for a saccular distal arch aortic aneurysm. After the placement of the Gore TAG (W. L. Gore & Associates, Inc., Newark, DE, USA) from zones 2 to 4, a Najuta endograft (Kawasumi Laboratories, Inc., Tokyo, Japan) was deployed from zone 0. Neither intraoperative angiography nor postoperative contrast-enhanced computed tomography (CT) showed endoleaks or migration. On day 12 after surgery, the patient suddenly lost consciousness during rehabilitation, and CT revealed the collapse of the Najuta endograft. In emergency surgery, the Najuta endograft was removed and the TAG was sutured to the aorta. Although the patient survived, he developed irreversible higher brain dysfunction. The cause of the collapse was examined by the manufacturer and only a slight bird-beak configuration was noted. There were no other findings to indicate the cause of the collapse. The Najuta endograft is a semi-customized system that is created according to the three-dimensional morphology of each individual aortic arch and, thus, is expected to follow the flexion of this vessel. Nevertheless, the risk of collapse needs to be considered.

Sector-specific Association of Intraocular Pressure Dynamics in Dark-room Prone Testing and Visual Field Defect Progression in Glaucoma.

PURPOSE: To investigate sectoral differences in the relationship between intraocular pressure (IOP) dynamics during dark-room prone testing (DRPT) and visual field (VF) defect progression in primary open-angle glaucoma (POAG) patients. DESIGN: Retrospective, longitudinal study. PARTICIPANTS: This retrospective study included 116 eyes of 84 POAG patients who underwent DRPT and had at least 5 reliable VF tests conducted over a more than 2-year follow-up period. We excluded eyes with mean deviation worse than -20 dB or a history of intraocular surgery or laser treatment. METHODS: Average total deviation (TD) was calculated in the superior, central, and inferior sectors of the Humphrey 24-2 or 30-2 program. During DRPT, IOP was measured in the sitting position, and after 60 minutes in the prone position in a dark room, IOP was measured again. The relationship between IOP change during DRPT, IOP after DRPT, and TD slope in each quadrant was analyzed with a linear mixed-effects model, adjusting for other potential confounding factors. MAIN OUTCOME MEASURES: Total deviation slope in each quadrant, IOP change during DRPT, and IOP after DRPT. RESULTS: Intraocular pressure after DRPT and IOP change during DRPT were 18.16 ± 3.42 mmHg and 4.92 ± 3.12 mmHg, respectively. Superior TD slope was significantly associated with both IOP after DRPT (ß = -0.28, P = 0.003) and IOP change during DRPT (ß = -0.21, P = 0.029), while central (ß = -0.05, P = 0.595; ß = -0.05; P = 0.622) and inferior (ß = 0.05, P = 0.611; ß = 0.01, P = 0.938) TD slopes were not. CONCLUSION: Dark-room prone testing might be a useful test to predict the risk of superior VF defect progression in eyes with POAG. FINANCIAL DISCLOSURE(S): The author(s) have no proprietary or commercial interest in any materials discussed in this article.

Modified Commando procedure using a double valve composite through an aorto-annulo-septotomy.

Double valve replacement of aortic and mitral valves with intervalvular fibrous body reconstruction (Commando procedure) is a challenging operation. Particularly in redo surgery for prosthetic valve endocarditis, special techniques are needed for approaching and reconstructing the valve complex. We performed a modified Commando procedure using a double valve composite through an aorto-annulo-septotomy. This approach provided a good field of vision at the combined aortic and mitral annuli without incising the left atrial roof. The double valve composite with four-layer patch wings was useful for performing double valve replacement en bloc and aorto-annulo-septotomy closure serially. Using these techniques, we successfully performed the Commando procedure for complicated prosthetic valve endocarditis.

Multimodality Risk Assessment of Patients with Ischemic Heart Disease Using Deep Learning Models Applied to Electrocardiograms and Chest X-rays.

Comprehensive management approaches for patients with ischemic heart disease (IHD) are important aids for prognostication and treatment planning. While single-modality deep neural networks (DNNs) have shown promising performance for detecting cardiac abnormalities, the potential benefits of using DNNs for multimodality risk assessment in patients with IHD have not been reported. The purpose of this study was to investigate the effectiveness of multimodality risk assessment in patients with IHD using a DNN that utilizes 12-lead electrocardiograms (ECGs) and chest X-rays (CXRs), with the prediction of major adverse cardiovascular events (MACEs) being of particular concern.DNN models were applied to detection of left ventricular systolic dysfunction (LVSD) on ECGs and identification of cardiomegaly findings on CXRs. A total of 2107 patients who underwent elective percutaneous coronary intervention were categorized into 4 groups according to the models' outputs: Dual-modality high-risk (n = 105), ECG high-risk (n = 181), CXR high-risk (n = 392), and No-risk (n = 1,429).A total of 342 MACEs were observed. The incidence of a MACE was the highest in the Dual-modality high-risk group (P < 0.001). Multivariate Cox hazards analysis for predicting MACE revealed that the Dual-modality high-risk group had a significantly higher risk of MACE than the No-risk group (hazard ratio (HR): 2.370, P < 0.001), the ECG high-risk group (HR: 1.906, P = 0.010), and the CXR high-risk group (HR: 1.624, P = 0.018), after controlling for confounding factors.The results suggest the usefulness of multimodality risk assessment using DNN models applied to 12-lead ECG and CXR data from patients with IHD.

The association between intraocular pressure dynamics during dark-room prone testing and intraocular pressure over a relatively long-term follow-up period in primary open-glaucoma patients.

PURPOSE: To investigate the relationship between the dynamics of intraocular pressure (IOP) during dark-room prone testing (DRPT) and IOP over a relatively long-term follow-up period. METHODS: This retrospective study enrolled 84 eyes of 51 primary open-angle glaucoma patients who underwent DRPT for whom at least three IOP measurements made using Goldmann applanation tonometry were available over a maximum follow-up period of two years. We excluded eyes with a history of intraocular surgery or laser treatment and those with changes in topical anti-glaucoma medication during the follow-up period. In DRPT, IOP was measured in the sitting position, and after 60 min in the prone position in a dark room, IOP was measured again. In this study, IOP fluctuation refers to the standard deviation (SD) of IOP, and IOP max indicates the maximum value of IOP during the follow-up. The relationship between these parameters was analyzed with a linear mixed-effects model, adjusting for clinical parameters including age, gender, and axial length. RESULTS: IOP increased after DRPT with a mean of 6.13 ± 3.55 mmHg. IOP max was significantly associated with IOP after DRPT (ß = 0.38; p < 0.001). IOP fluctuation was significantly associated with IOP change in DRPT (ß = 0.29; p = 0.007). CONCLUSION: Our findings suggest that short-term and relatively long-term IOP dynamics are associated. Long-term IOP dynamics can be predicted by DRPT to some extent.

Deep Learning Models for Predicting Left Heart Abnormalities From Single-Lead Electrocardiogram for the Development of Wearable Devices.

BACKGROUND: Left heart abnormalities are risk factors for heart failure. However, echocardiography is not always available. Electrocardiograms (ECGs), which are now available from wearable devices, have the potential to detect these abnormalities. Nevertheless, whether a model can detect left heart abnormalities from single Lead I ECG data remains unclear.Methods and Results: We developed Lead I ECG models to detect low ejection fraction (EF), wall motion abnormality, left ventricular hypertrophy (LVH), left ventricular dilatation, and left atrial dilatation. We used a dataset comprising 229,439 paired sets of ECG and echocardiography data from 8 facilities, and validated the model using external verification with data from 2 facilities. The area under the receiver operating characteristic curves of our model was 0.913 for low EF, 0.832 for wall motion abnormality, 0.797 for LVH, 0.838 for left ventricular dilatation, and 0.802 for left atrial dilatation. In interpretation tests with 12 cardiologists, the accuracy of the model was 78.3% for low EF and 68.3% for LVH. Compared with cardiologists who read the 12-lead ECGs, the model's performance was superior for LVH and similar for low EF. CONCLUSIONS: From a multicenter study dataset, we developed models to predict left heart abnormalities using Lead I on the ECG. The Lead I ECG models show superior or equivalent performance to cardiologists using 12-lead ECGs.

Sex differences in the association between systemic oxidative stress status and optic nerve head blood flow in normal-tension glaucoma.

PURPOSE: To investigate the association of systemic oxidative stress markers and optic nerve head (ONH) blood flow in normal-tension glaucoma (NTG) patients, as well as sex differences in this association. METHODS: This was a cross-sectional study of 235 eyes with NTG of 134 patients (56 male, 78 female; mean age, 60.9±14.1 years). Laser speckle flowgraphy (LSFG) was used to measure ONH blood flow (mean blur rate in the tissue area of the ONH; MBR-T) and LSFG pulse-waveform parameters, including flow acceleration index in the tissue area of the ONH (FAI-T). Oxidative stress markers, diacron-reactive oxygen metabolites (d-ROMs), and biological antioxidant potential (BAP) were measured with a free radical elective evaluator. Spearman's rank correlation test and a multivariate linear mixed-effect model were used to investigate factors associated with ONH blood flow. RESULTS: MBR-T was significantly correlated with age (rs = -0.28, p < 0.001), mean arterial pressure (rs = -0.20, p = 0.002), intraocular pressure (rs = 0.24, p < 0.001), peripapillary retinal nerve fiber layer thickness (rs = 0.62, p < 0.001), and disc area (rs = -0.26, p < 0.001), but not with serum d-ROM level. Separate analyses of the subjects divided by sex showed that BAP was positively correlated to MBR-T (rs = 0.21, p = 0.036) and FAI-T (rs = 0.36, p < 0.001) only in male subjects. Similarly, BAP was significantly associated with MBR-T (ß = 0.25, p = 0.026) and FAI-T (ß = 0.37, p < 0.001) in male subjects in a multivariate linear mixed-effect model. CONCLUSION: A lower serum antioxidant level, as indicated by BAP, was associated with reduced ONH blood flow only in male NTG patients. Our findings suggest that there are sex differences in the involvement of oxidative stress in the pathogenesis of reduced ocular blood flow in NTG.

Real-time intelligent classification of COVID-19 and thrombosis via massive image-based analysis of platelet aggregates.

Microvascular thrombosis is a typical symptom of COVID-19 and shows similarities to thrombosis. Using a microfluidic imaging flow cytometer, we measured the blood of 181 COVID-19 samples and 101 non-COVID-19 thrombosis samples, resulting in a total of 6.3 million bright-field images. We trained a convolutional neural network to distinguish single platelets, platelet aggregates, and white blood cells and performed classical image analysis for each subpopulation individually. Based on derived single-cell features for each population, we trained machine learning models for classification between COVID-19 and non-COVID-19 thrombosis, resulting in a patient testing accuracy of 75%. This result indicates that platelet formation differs between COVID-19 and non-COVID-19 thrombosis. All analysis steps were optimized for efficiency and implemented in an easy-to-use plugin for the image viewer napari, allowing the entire analysis to be performed within seconds on mid-range computers, which could be used for real-time diagnosis.

A Case of Endovascular Aortic Repair for Abdominal Aortic Aneurysm with a Saccular Aneurysm in the Severely Angulated Proximal Landing Zone.

We report a case of endovascular aortic repair (EVAR) for the abdominal aortic aneurysm complicated by a saccular aneurysm due to a penetrating atherosclerotic ulcer in the severely angulated proximal landing zone. To secure the zone, coil embolization of the saccular aneurysm was performed before stent grafting to treat the abdominal aortic aneurysm. To precisely follow the severely angled proximal neck, we used the Excluder stent-graft system inserted by the body floss technique method instead of the stiff wire method to avoid accordion folding the proximal landing zone. These techniques may expand the indications of EVAR.

Focus movement distance per pulse dependence of electrical conductivity and diameter of diamond internal modification induced by picosecond laser.

Internal and local modifications via ultrashort pulsed laser illumination to diamond are promising for manufacturing diamond electronic devices. The relationship between the diameter/electrical conductivity of modified regions and the laser fluence distribution was investigated. Picosecond laser illumination without scanning the laser focus fabricated short modified regions in diamond. As a result, the calculated laser fluence distribution matches the distribution of the modified regions. Wire-shaped modified regions were fabricated via laser illumination with scanning of the laser focus, and the corresponding diameter and electrical conductivity were investigated by controlling the laser focus movement distance per pulse (Vf). The modified regions fabricated with varying Vf were divided into three categories depending on the trend of the relationship between the diameter and electrical conductivity. The diameters of the modified regions were constant at the maximum values when Vf was sufficiently small, decreased with increasing Vf, and reached a minimum when Vf was sufficiently large. The modified regions became more electrically conductive with increasing Vf, even when the deposited energy per unit length decreased. Moreover, the electrical conductivity decreased significantly when the diameter became constant at the minimum value. Finally, the relationship between the diameter/electrical conductivity of the modified regions and the laser fluence distribution was elucidated.

Age prediction from coronary angiography using a deep neural network: Age as a potential label to extract prognosis-related imaging features.

Coronary angiography (CAG) is still considered the reference standard for coronary artery assessment, especially in the treatment of acute coronary syndrome (ACS). Although aging causes changes in coronary arteries, the age-related imaging features on CAG and their prognostic relevance have not been fully characterized. We hypothesized that a deep neural network (DNN) model could be trained to estimate vascular age only using CAG and that this age prediction from CAG could show significant associations with clinical outcomes of ACS. A DNN was trained to estimate vascular age using ten separate frames from each of 5,923 CAG videos from 572 patients. It was then tested on 1,437 CAG videos from 144 patients. Subsequently, 298 ACS patients who underwent percutaneous coronary intervention (PCI) were analysed to assess whether predicted age by DNN was associated with clinical outcomes. Age predicted as a continuous variable showed mean absolute error of 4 years with R squared of 0.72 (r = 0.856). Among the ACS patients stratified by predicted age from CAG images before PCI, major adverse cardiovascular events (MACE) were more frequently observed in the older vascular age group than in the younger vascular age group (p = 0.017). Furthermore, after controlling for actual age, gender, peak creatine kinase, and history of heart failure, the older vascular age group independently suffered from more MACE (hazard ratio 2.14, 95% CI 1.07 to 4.29, p = 0.032). The vascular age estimated based on CAG imaging by DNN showed high predictive value. The age predicted from CAG images by DNN could have significant associations with clinical outcomes in patients with ACS.

Endovascular Stent Grafting for Recurrent Strokes Due to Fragile Innominate Artery Plaque: A Case Report.

Here we describe a case of recurrent ischemic strokes due to fragile innominate artery plaque successfully treated using endovascular stent grafting. An 80-year-old man presented with a history of recurrent strokes that were refractory to medical treatment. Computed tomography and magnetic resonance images of the thorax revealed a gross intramural plaque in the innominate artery. He was successfully treated using endovascular stent grafting. An AFX stent graft device was used to prevent further embolic strokes. The AFX stent graft has a unique endoskeleton design with a thin-walled expanded polytetrafluoroethylene fabric-known as active sealing structure-attached to the implant. Postoperatively, the patient has experienced no recurrent strokes in over 2 years of follow-up. The stent grafting procedure could be an optimal treatment option for treating fragile innominate artery plaques.

Photostable near-infrared-absorbing diradical-platinum(ii) complex solubilized by albumin toward a cancer photothermal therapy agent.

A hydrophobic diradical-platinum(ii) complex was solubilized in aqueous solutions by using bovine serum albumin and exhibited photothermal conversion under near-infrared (NIR) light irradiation. The complex was introduced into cancer cells and induced cell death upon absorption of NIR. These results imply that the complex can function as a photothermal therapeutic agent.

Limitations of median nerve somatosensory evoked potential monitoring during carotid endarterectomy.

OBJECTIVE: Hypoperfusion during carotid artery cross-clamping (CC) for carotid endarterectomy (CEA) may result in the major complication of perioperative stroke. Median nerve somatosensory evoked potential (MNSSEP) monitoring, which is an established method for the prediction of cerebral ischemia, has low sensitivity in detecting such hypoperfusion. In this study the authors sought to explore the limitations of MNSSEP monitoring compared to tibial nerve somatosensory evoked potential (TNSSEP) monitoring for the detection of CC-related hypoperfusion. METHODS: The authors retrospectively analyzed data from patients who underwent unilateral CEA with routine shunt use. All patients underwent preoperative magnetic resonance angiography and were monitored for intraoperative cerebral ischemia by using MNSSEP, TNSSEP, and carotid stump pressure during CC. First, the frequency of MNSSEP and TNSSEP changes during CC were analyzed. Subsequently, variables related to stump pressure were determined by using linear analysis and those related to each of the somatosensory evoked potential (SSEP) changes were determined by using logistic regression analysis. RESULTS: A total of 94 patients (mean age 74 years) were included in the study. TNSSEP identified a greater number of SSEP changes during CC than MNSSEP (20.2% vs 11.7%; p < 0.05). Linear regression analysis demonstrated that hypoplasia of the contralateral proximal segment of the anterior cerebral artery (A1 hypoplasia) (p < 0.01) and hypoplasia of the ipsilateral precommunicating segment of the posterior cerebral artery (P1 hypoplasia) (p = 0.02) independently and negatively correlated with stump pressure. Both contralateral A1 hypoplasia (OR 26.25, 95% CI 4.52-152.51) and ipsilateral P1 hypoplasia (OR 8.75, 95% CI 1.83-41.94) were independently related to the TNSSEP changes. However, only ipsilateral P1 hypoplasia (OR 8.76, 95% CI 1.61-47.67) was independently related to MNSSEP changes. CONCLUSIONS: TNSSEP monitoring appears to be superior to MNSSEP in detecting CC-related hypoperfusion. Correlation with stump pressure and SSEP changes indicates that TNSSEP, and not MNSSEP monitoring, is a reliable indicator of cerebral ischemia in the territory of the anterior cerebral artery.