This paper presents an integrated circuit (IC) realization of a random forest (RF) machine learning classifier in a 65-nm CMOS. Algorithm, architecture, and circuits are co-optimized to achieve ...aggressive energy and delay benefits by taking advantage of the inherent error resiliency derived from the ensemble nature of an RF classifier. Deterministic subsampling (DSS) and regularized decision trees reduce interconnect complexity, and avoid irregular memory access patterns and computations, thereby reducing the energy-delay product (EDP). The prototype IC also employs low-swing analog in-memory computations embedded in a standard 6T SRAM to enable massively parallel tree node comparisons, thereby minimizing the memory fetches and reducing the EDP further. The 65-nm CMOS prototype IC achieves a 3.1× and 2.2× improved energy efficiency and throughput leading to 6.8× lower EDP compared to a conventional digital system at the same accuracies of 94% and 97.5% for two tasks: 1) eight-class traffic sign recognition and 2) face detection, respectively.
Conventional risk stratification models for mortality of acute myocardial infarction (AMI) have potential limitations. This study aimed to develop and validate deep-learning-based risk stratification ...for the mortality of patients with AMI (DAMI).
The data of 22,875 AMI patients from the Korean working group of the myocardial infarction (KorMI) registry were exclusively divided into 12,152 derivation data of 36 hospitals and 10,723 validation data of 23 hospitals. The predictor variables were the initial demographic and laboratory data. The endpoints were in-hospital mortality and 12-months mortality. We compared the DAMI performance with the global registry of acute coronary event (GRACE) score, acute coronary treatment and intervention outcomes network (ACTION) score, and the thrombolysis in myocardial infarction (TIMI) score using the validation data.
In-hospital mortality for the study subjects was 4.4% and 6-month mortality after survival upon discharge was 2.2%. The areas under the receiver operating characteristic curves (AUCs) of the DAMI were 0.905 95% confidence interval 0.902-0.909 and 0.870 0.865-0.876 for the ST elevation myocardial infarction (STEMI) and non ST elevation myocardial infarction (NSTEMI) patients, respectively; these results significantly outperformed those of the GRACE (0.851 0.846-0.856, 0.810 0.803-0.819), ACTION (0.852 0.847-0.857, 0.806 0.799-0.814 and TIMI score (0.781 0.775-0.787, 0.5930.585-0.603). DAMI predicted 30.9% of patients more accurately than the GRACE score. As secondary outcome, during the 6-month follow-up, the high risk group, defined by the DAMI, has a significantly higher mortality rate than the low risk group (17.1% vs. 0.5%, p < 0.001).
The DAMI predicted in-hospital mortality and 12-month mortality of AMI patients more accurately than the existing risk scores and other machine-learning methods.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
The computing power provided by the edge device is becoming increasingly improved in recent days, largely due to the technological development that realizes accelerators with a great amount of ...hardware parallelism. Thanks to the enhanced computing power, the application domain of the edge device is greatly expanding so that it is capable of performing new tasks previously inconceivable. In this paper, we investigate the potentials of the state-of-the-art edge devices in the context of scientific computing. We implement and perform stochastic biochemical simulation on a small cluster of modern, GPU-accelerated edge systems to examine the computational capability of the device. In addition, we perform and analyze the NAS Parallel Benchmarks (NPB) performance on the edge device. By comparing the performance of the edge device with other GPU (graphics processing unit) and CPU based systems across different platforms, we evaluate the applicability and usefulness of the modern hardware-accelerated edge devices in the scientific computing domain.
Out-of-hospital cardiac arrest (OHCA) is a major healthcare burden, and prognosis is critical in decision-making for treatment and the withdrawal of life-sustaining therapy. This study aimed to ...develop and validate a deep-learning-based out-of-hospital cardiac arrest prognostic system (DCAPS) for predicting neurologic recovery and survival to discharge.
The study subjects were patients from the Korea OHCA registry who experienced return of spontaneous circulation (ROSC) after OHCA. A total of 36,190 patients were exclusively divided into a set of 28,045 subjects for derivation data and 8,145 subjects for validation data. We used information available for the time of ROSC as predictor variables, and the endpoints were neurologic recovery (cerebral performance category 1 or 2) and survival to discharge. The DCAPS was developed using the derivation data and represented the favorability of prognosis with a score between 0 and 100.
The area under the receiver operating characteristic curve (AUROC) of DCAPS for predicting neurologic recovery for the validation data was 0.953 95% confidence interval 0.952–0.954; these results significantly outperformed those of logistic regression (0.947 0.943–0.948), random forest (0.943 0.942–0.945), support vector machine (0.930 0.929–0.932), and conventional methods of a previous study (0.817 0.815–0.820). The AUROC of the DCAPS for survival to discharge was 0.901 0.900–0.903, and this result significantly outperformed those of other models as well.
The DCAPS predicted neurologic recovery and survival to discharge of OHCA patients accurately and outperformed the conventional method and other machine-learning methods.
Schwann cells play a crucial role in successful nerve repair and regeneration by supporting both axonal growth and myelination. However, the sources of human Schwann cells are limited both for ...studies of Schwann cell development and biology and for the development of treatments for Schwann cell-associated diseases. Here, we provide a rapid and scalable method to produce self-renewing Schwann cell precursors (SCPs) from human pluripotent stem cells (hPSCs), using combined sequential treatment with inhibitors of the TGF-β and GSK-3 signaling pathways, and with neuregulin-1 for 18 days under chemically defined conditions. Within 1 week, hPSC-derived SCPs could be differentiated into immature Schwann cells that were functionally confirmed by their secretion of neurotrophic factors and their myelination capacity in vitro and in vivo. We propose that hPSC-derived SCPs are a promising, unlimited source of functional Schwann cells for treating demyelination disorders and injuries to the peripheral nervous system.
•hPSC-SCPs are highly expandable under chemically defined medium condition•hPSC-SCPs can rapidly and efficiently differentiate into functional Schwann cells•SCP-SCs myelinate axon and secrete various neurotrophic factors•SCP-SCs promote axonal regeneration in sciatic nerve-damaged mice
In this article, Cho and colleagues show that an efficient strategy for producing directly an unlimited supply of functional human Schwann cells (SCs) via successful derivation of expandable Schwann cell precursors (SCPs) from human pluripotent stem cells (hPSC-SCPs). Functional and molecular characteristic of SCs from hPSC-SCPs (SCP-SCs) were shown both in vitro and in vivo.
Stent thrombosis (ST) remains a catastrophic problem in patients undergoing percutaneous coronary intervention (PCI). However, a paucity of data exist regarding the incidence, implications, and ...predictors of ST in patients with acute myocardial infarction (AMI). We consecutively enrolled patients with AMI in the CardiOvascular Risk and idEntificAtion of potential high-risk population in AMI registry who underwent PCI from January 2004 to December 2009 and analyzed definite or probable ST according to Academic Research Consortium definitions. The median follow-up duration was 41.9 months. Definite or probable ST occurred in 136 patients (3.7%), including 44 with early ST (1.0%), 38 with late ST (0.9%), and 54 with very late ST (2.0%). The annual incidence of very late ST ranged from 0.5% to 0.6%. The all-cause mortality rate after ST was 29%, which was higher than that for patients without ST (17%; p <0.001). The independent predictors of ST were no-reflow phenomenon (hazard ratio HR 1.96, 95% confidence interval CI 1.28 to 3.03), decreased left ventricular ejection fraction (HR 1.70, 95% CI 1.21 to 2.40), anemia (HR 1.54, 95% CI 1.09 to 2.18), and a mean stent diameter <3.0 mm (HR 1.53, 95% CI 1.04 to 2.27). ST is not uncommon in patients with AMI and continues to occur beyond 1 year after PCI, irrespective of the stent type or clinical presentation. Patients with ST are associated with higher mortality than patients without ST. No reflow, decreased left ventricular ejection fraction, anemia, and a mean stent diameter <3.0 mm are independent predictors of ST.
Stent length has been considered an important predictor of adverse events after percutaneous coronary intervention, even with the first-generation drug-eluting stents (DESs). The introduction of ...newer-generation DES has further reduced the rates of adverse clinical events such as restenosis, myocardial infarction, and stent thrombosis. The aim of this study was to compare the impact of stent length on the long-term clinical outcomes between first- and newer-generation DESs. The effects of stent length (≥32 vs <32 mm) on the clinical outcomes were evaluated in 8,445 patients who underwent percutaneous coronary intervention using either a first-generation DES (sirolimus- and paclitaxel-eluting stents, n = 6,334) or a newer-generation DES (everolimus- and zotarolimus-eluting stents, n = 2,111) from January 2004 to December 2009. The 3-year adverse outcomes (composite of all-cause death, nonfatal myocardial infarction, target vessel revascularization, and stent thrombosis) were compared using the inverse probability of treatment-weighted method according to the stent length. After adjustment for differences in the baseline risk factors, a stent length of ≥32 mm was significantly associated with higher cumulative rates of target vessel revascularization and stent thrombosis in the patients treated with a first-generation DES (adjusted hazard ratio 1.875, 95% confidence interval 1.531 to 2.297, p <0.001; adjusted hazard ratio 2.964, 95% confidence interval 1.270 to 6.917, p = 0.012), but it was not associated with the clinical outcomes in patients treated with a newer-generation DES. In conclusion, stent length might not be associated with long-term clinical outcomes in newer-generation DES era, whereas stent length might be associated with long-term clinical outcomes in the first-generation DESs.
Electrically nonconducting UPW was electrolyzed without electrolyte through anion exchange membrane for evaluating applicability to EUV semiconductor cleaning. Produced anode water held positive ORP ...up to 900 mV, which is very oxidative. ORP, pH, and conductivity measurements showed delicately complementary each other for understanding anode water. Correlation of concurrent ORP decrease and conductivity increase in ultra-pure anode water domain was observed first time. The oxidative OH° was formed as the major species in anode water, causing positive ORP during ORP measurement. H
+
and OH
−
ions, and OH° radical coexisted in anode water at amphoteric nonequilibrium, while pH was less than 6. It was concluded that OH°, as a strong oxidant, transformed itself to OH
−
by ORP measurement. OH° radical would oxidize selectively and then remove nano-contaminants. Anode water is considered to fulfill the requirement of EUV semiconductor cleaning where no oxygen species should be required because of likely oxide layer formation during cleaning, and it will even remove the native oxide developed unintentionally before cleaning.
The available data are not sufficient to understand the clinical impact of statin intensity in elderly patients who undergo percutaneous coronary intervention (PCI) due to acute myocardial infarction ...(AMI).
Using the COREA-AMI registry, we sought to compare the clinical impact of high- versus low-to-moderate-intensity statin in younger (<75 years old) and elderly (≥75 years old) patients. Of 10,719 patients, we included 8,096 patients treated with drug-eluting stents. All patients were classified into high-intensity versus low-to-moderate-intensity statin group according to statin type and dose at discharge. The primary end point was target-vessel failure (TVF), a composite of cardiovascular death, target-vessel MI, or target-lesion revascularization (TLR) from 1 month to 12 months after index PCI.
In younger patients, high-intensity statin showed the better clinical outcomes than low-to-moderate-intensity statin (TVF: 79 5.4% vs. 329 6.8%, adjusted hazard ratio aHR 0.76; 95% confidence interval CI 0.59-0.99; P = 0.038). However, in elderly patients, the incidence rates of the adverse clinical outcomes were similar between two statin-intensity groups (TVF: 38 11.4% vs. 131 10.6%, aHR 1.1; 95% CI 0.76-1.59; P = 0.63).
In this AMI cohort underwent PCI, high-intensity statin showed the better 1-year clinical outcomes than low-to-moderate-intensity statin in younger patients. Meanwhile, the incidence rates of adverse clinical events between high- and low-to-moderate-intensity statin were not statistically different in elderly patients. Further randomized study with large elderly population is warranted.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
Transplant recipients are immunocompromised and vulnerable to developing tuberculosis. However, active tuberculosis incidence is rapidly declining in South Korea, but the trend of tuberculosis ...infection among transplant recipients has not been elucidated. This study aimed to evaluate the risk of active tuberculosis after transplantation, including risk factors for tuberculosis and standardized incidence ratios, compared with that in the general population.
This retrospective study was conducted based on the South Korean health insurance review and assessment database among those who underwent transplantation (62,484 recipients) between 2008 and 2020. Tuberculosis incidence was compared in recipients treated during higher- (2010-2012) and lower-disease burden (2016-2018) periods. Standardized incidence ratios were analyzed using the Korean Tuberculosis Surveillance System. The primary outcome was the number of new tuberculosis cases after transplantation.
Of 57,103 recipients analyzed, the overall cumulative incidence rate 1 year after transplantation was 0.8% (95% confidence interval CI: 0.7-0.8), significantly higher in the higher-burden period than in the lower-burden period (1.7% vs. 1.0% 3 years after transplantation, P < 0.001). Individuals who underwent allogeneic hematopoietic stem cell transplantation had the highest tuberculosis incidence, followed by those who underwent solid organ transplantation and autologous hematopoietic stem cell transplantation (P < 0.001). The overall standardized incidence ratio was 3.9 (95% CI 3.7-4.2) and was the highest in children aged 0-19 years, at 9.0 (95% CI 5.7-13.5). Male sex, older age, tuberculosis history, liver transplantation, and allogeneic hematopoietic stem cell transplantation were risk factors for tuberculosis.
Transplant recipients are vulnerable to developing tuberculosis, possibly influenced by their immunocompromised status, solid organ transplant type, age, and community prevalence of tuberculosis. Tuberculosis prevalence by country, transplant type, and age should be considered to establish an appropriate tuberculosis prevention strategy for high-risk groups.