We aimed to measure the diagnostic accuracy of the deep learning model (DLM) for ST-elevation myocardial infarction (STEMI) on a 12-lead electrocardiogram (ECG) according to culprit artery sorts. From January 2017 to December 2019, we recruited patients with STEMI who received more than one stent insertion for culprit artery occlusion. The DLM was trained with STEMI and normal sinus rhythm ECG for external validation. The primary outcome was the diagnostic accuracy of DLM for STEMI according to the three different culprit arteries. The outcomes were measured using the area under the receiver operating characteristic curve (AUROC), sensitivity (SEN), and specificity (SPE) using the Youden index. A total of 60,157 ECGs were obtained. These included 117 STEMI-ECGs and 60,040 normal sinus rhythm ECGs. When using DLM, the AUROC for overall STEMI was 0.998 (0.996-0.999) with SEN 97.4% (95.7-100) and SPE 99.2% (98.1-99.4). There were no significant differences in diagnostic accuracy within the three culprit arteries. The baseline wanders in false positive cases (83.7%, 345/412) significantly interfered with the accurate interpretation of ST elevation on an ECG. DLM showed high diagnostic accuracy for STEMI detection, regardless of the type of culprit artery. The baseline wanders of the ECGs could affect the misinterpretation of DLM.
Background and Objectives: This study assessed the prognostic value of underlying chronic kidney disease (CKD) and renal replacement therapy (RRT) on the clinical outcomes from out-of-hospital cardiac arrest (OHCA). Materials and Methods: This retrospective study was conducted utilizing the population-based OHCA data of South Korea between 2008 and 2018. Adult (>18 years) OHCA patients with a medical cause of cardiac arrest were included and classified into three categories based on the underlying CKD and RRT: (1) non-CKD group; (2) CKD without RRT group; and (3) CKD with RRT group. A total of 13,682 eligible patients were included (non-CKD, 9863; CKD without RRT, 1778; CKD with RRT, 2041). From the three comparison subgroups, data with propensity score matching were extracted. The influence of CKD and RRT on patient outcomes was assessed using propensity score matching and multivariate logistic regression analyses. The primary outcome was survival at hospital discharge and the secondary outcome was a good neurological outcome at hospital discharge. Results: The two CKD groups (CKD without RRT and CKD with RRT) showed no significant difference in survival at hospital discharge compared with the non-CKD group (CKD without RRT vs. non-CKD, p > 0.05; CKD with RRT vs. non-CKD, p > 0.05). The non-CKD group had a higher chance of having good neurological outcomes than the CKD groups (non-CKD vs. CKD without RRT, p < 0.05; non-CKD vs. CKD with RRT, p < 0.05) whereas there was no significant difference between the two CKD groups (CKD without RRT vs. CKD with RRT, p > 0.05). Conclusions: Compared with patients without CKD, the underlying cause of CKD—regardless of RRT—may be linked to poor neurological outcomes. Underlying CKD and RRT had no effect on the survival at hospital discharge.
(1) Background: Rapid and accurate negative discrimination enables efficient management of scarce isolated bed resources and adequate patient accommodation in the majority of areas experiencing an explosion of confirmed cases due to Omicron mutations. Until now, methods for artificial intelligence or deep learning to replace time-consuming RT-PCR have relied on CXR, chest CT, blood test results, or clinical information. (2) Methods: We proposed and compared five different types of deep learning algorithms (RNN, LSTM, Bi-LSTM, GRU, and transformer) for reducing the time required for RT-PCR diagnosis by learning the change in fluorescence value derived over time during the RT-PCR process. (3) Results: Among the five deep learning algorithms capable of training time series data, Bi-LSTM and GRU were shown to be able to decrease the time required for RT–PCR diagnosis by half or by 25% without significantly impairing the diagnostic performance of the COVID-19 RT–PCR test. (4) Conclusions: The diagnostic performance of the model developed in this study when 40 cycles of RT–PCR are used for diagnosis shows the possibility of nearly halving the time required for RT–PCR diagnosis.
Abstract The aim of this study was to determine which of 4 laryngoscopes, including A-LRYNGO, a newly developed channel-type video-laryngoscope with an embedded artificial intelligence-based glottis guidance system, is appropriate for tracheal intubation training in novice medical students wearing personal protective equipment (PPE). Thirty healthy senior medical school student volunteers were recruited. The participants underwent 2 tests with 4 laryngoscopes: Macintosh, McGrath, Pentax Airway-Scope and A-LRYNGO. The first test was conducted just after a lecture without any hands-on workshop. The second test was conducted after a one-on-one hands-on workshop. In each test, we measured the time required for tracheal intubation, intubation success rate, etc, and asked all participants to complete a short questionnaire. The time to completely insert the endotracheal tube with the Macintosh laryngoscope did not change significantly ( P = .177), but the remaining outcomes significantly improved after the hands-on workshop (all P < .05). Despite being novice practitioners with no intubation experience and wearing PPE, the, 2 channel-type video-laryngoscopes were associated with good intubation-related performance before the hands-on workshop (all P < .001). A-LRYNGO's artificial intelligence-based glottis guidance system showed 93.1% accuracy, but 20.7% of trials were guided by the vocal folds. To prepare to manage the airway of critically ill patients during the coronavirus disease 2019 pandemic, a channel-type video-laryngoscope is appropriate for tracheal intubation training for novice practitioners wearing PPE.
Abstract This study aimed to investigate the prognostic difference between AUTOPULSE and LUCAS for out-of-hospital cardiac arrest (OHCA) adult patients. A retrospective observational study was performed nationwide. Adult OHCA patients after receiving in-hospital mechanical chest compression from 2012 to 2016 were included. The primary outcomes were sustained return of spontaneous circulation (ROSC) of more than 20 minutes and survival to discharge. Among 142,906 OHCA patients, 820 patients were finally included. In multivariate analysis, female (OR, 0.57; 95% CI, 0.33–0.99), witnessed arrest (OR, 2.10; 95% CI, 1.20–3.69), and arrest cause of non-cardiac origin (OR, 0.25; 95% CI, 0.10–0.62) were significantly associated with the increase in ROSC. LUCAS showed a lower survival than AUTOPULSE (OR, 0.23; 95% CI, 0.06–0.84), although it showed no significant association with ROSC. Percutaneous coronary intervention (OR, 6.30; 95% CI, 1.53–25.95) and target temperature management (TTM; OR, 7.30; 95% CI, 2.27–23.49) were the independent factors for survival. We categorized mechanical CPR recipients by witness to compare prognostic effectiveness of AUTOPULSE and LUCAS. In the witnessed subgroup, female (OR, 0.46; 95% CI, 0.24–0.89) was a prognostic factor for ROSC and shockable rhythm (OR, 5.04; 95% CI, 1.00–25.30), percutaneous coronary intervention (OR, 12.42; 95% CI, 2.04–75.53), and TTM (OR, 9.03; 95% CI, 1.86–43.78) for survival. In the unwitnessed subgroup, no prognostic factors were found for ROSC, and TTM (OR, 99.00; 95% CI, 8.9–1100.62) was found to be an independent factor for survival. LUCAS showed no significant increase in ROSC or survival in comparison with AUTOPULSE in both subgroups. The in-hospital use of LUCAS may have a deleterious effect for survival compared with AUTOPULSE.
This study aimed to evaluate the efficacy of i-gel blind intubation (IGI) as a rescue device for definitive airway management in ground intubation for pre-hospital trauma patients.A prospective randomized crossover study was conducted with 18 paramedics to examine intubation performance of two blind intubation techniques through a supraglottic airway devices (IGI and laryngeal mask airway Fastrach), compared with use of a Macintosh laryngoscope (MCL). Each intubation was conducted at two levels of patient positions (ground- and stretcher-level). Primary outcomes were the intubation time and the success rate for intubation.The intubation time (sec) of each intubation technique was not significantly different between the two positions. In both patient positions, the intubation time of IGI was shortest among the three intubation techniques (17.9±5.2 at the ground-level and 16.9±3.8 at the stretcher-level). In the analysis of cumulative success rate and intubation time, IGI was the fastest to reach 100% success among the three intubation techniques regardless of patient position (all P<0.017). The success of intubation was only affected by the intubation technique, and IGI achieved more success than MCL (odds ratio, 3.6; 95% confidence interval, 1.1 to 11.6; P=0.03).The patient position did not affect intubation performance. Additionally, the intubation time with blind intubation through supraglottic airway devices, especially with IGI, was significantly shorter than that with MCL.
'%3e%3cg id='b'%3e%3cpath id='a' stroke='%23000' stroke-width='2' d='M-6-25H6m-12 3H6m-12 3H6'/%3e%3cuse xlink:href='%23a' y='44'/%3e%3c/g%3e%3cpath stroke='%23fff' d='M0 17v10'/%3e%3ccircle r='12' fill='%23cd2e3a'/%3e%3cpath fill='%230047a0' d='M0-12A6 6 0 0 0 0 0a6 6 0 0 1 0 12 12 12 0 0 1 0-24z'/%3e%3c/g%3e%3cg transform='rotate(-123.69)'%3e%3cuse xlink:href='%23b'/%3e%3cpath stroke='%23fff' d='M0-23.5v3M0 17v3.5m0 3v3'/%3e%3c/g%3e%3c/svg%3e)
