ACS mortality prediction in Asian in-hospital patients with deep learning using machine learning feature selection
1
Citation
0
Reference
10
Related Paper
Citation Trend
Abstract:
Abstract Background Thrombolysis in Myocardial infarction (TIMI) is used in predicting the mortality rate of the acute coronary syndrome (ACS) patients. TIMI was developed based on the Western cohort with limited data on the Asian cohort. There are separate TIMI scores for STEMI and NSTEMI. Deep learning (DL) and machine learning (ML) algorithms such as support vector machine (SVM) in population-specific dataset resulted in a higher area under the curve (AUC) to TIMI. The limitation of DL is selected features by the algorithm is unknown compared to ML algorithms. Purpose To construct a single in-hospital mortality risk scoring system that combines SVM feature importance and the DL algorithm in ASIAN patients with ACS that is applicable for both STEMI and NSTEMI patients. To investigate DL performance constructed using predictors selected from SVM feature extraction and DL using complete features and compare with TIMI risk score for STEMI and NSTEMI patients. Methods We constructed four algorithms: i) DL and SVM algorithm with feature selected from SVM variable importance, ii) DL and SVM algorithm without feature selection. SVM feature importance with the backward elimination method is used to select and rank important variables. We used registry data from the National Cardiovascular Disease Database of 13190 patient's data. Fifty-four parameters including demographics, cardiovascular risk, medications and clinical variables were considered. AUC was used as the performance evaluation metric. All algorithms were validated using validation dataset and compared to the conventional TIMI for STEMI and NSTEMI. Results Validation results in Figure 1 are by STEMI and NTEMI patients. Both DL algorithms outperformed ML and TIMI score on validation data. Similar performance is observed for DL and SVM algorithms using all predictors (54 predictors) with DL and SVM algorithm using selected predictors (14 predictors). Predictors selected by the SVM feature selection are: age, heart rate, Killip class, fasting blood glucose, ST-elevation, CABG, cardiac catheterization, angina episode, HDLC, LDC, other lipid-lowering agents, statin, anti-arrhythmic agent, oralhypogly. CABG and pharmacotherapy drugs as selected predictors improve mortality prediction compared to TIMI score. In DL, 25.87% of STEMI patients and 19.71% of NSTEMI patients are estimated as high risk (risk probabilities of >50%). TIMI underestimated the risk of mortality of high-risk patients (≥5 risk scores) with 13.08% from STEMI patients and 4.65% from NSTEMI patients (Figure 2). Conclusions In the ASIAN multi-ethnicity population, patients with ACS can be better classified using one single algorithm compared to the conventional method like TIMI which requires two different scores. Combining ML feature selection with DL allows the identification of distinct factors related to in-hospital mortality of ACS patients in a unique ASIAN population for better mortality prediction. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Technology Development Fund 1 Figure 1. Performance resultsFigure 2. Analysis on the validation setKeywords:
TIMI
Feature (linguistics)
Abstract Background : Increased QT interval dispersion (QTd) has been found in patients with acute myocardial infarction (AMI). In previous studies this has been shown to decrease with thrombolysis. Hypothesis : The aim of this study was to compare the effects of reperfusion by primary percutaneous transluminal coronary angioplasty (PTCA) and by thrombolysis on QTd and correlate these results with the degree of reperfusion. Methods : We studied 60 patients with a first AMI. The study cohort included 40 consecutive patients who had received thrombolysis (streptokinase or rt‐PA); 20 additional consecutive patients with successful primary PTCA, all with preselected Thrombolysis in Myocardial Infarction (TIMI) grade 3 flow by predefined selection criteria (12 stents); and 20 controls. A 12‐lead ECG for QTd calculation was recorded before thrombolysis or PTCA and immediately after the procedure. All values were corrected according to Bazett's formula (QTcd). QTd and QTcd values before and after each procedure in three groups and the respective percent changes of ΔQTd and ΔQTcd were compared separately. Results : QTd and QTcd were significantly increased before thrombolysis/PTCA versus normals. An angiogram performed after thrombolysis showed adequate reperfusion (TIMI grade 2/3) in 20 patients, while in the other 20 only TIMI 0/1 reperfusion was achieved. Thrombolysis‐TIMI flow 2/3 and PTCA significantly reduced QTd (from 68 ± 10 to 35 ± 8 ms, p < 0.001, ΔQTd = 48 ± 11%, in the Thr‐TIMI flow 2/3 group, and from 79 ± 11 to 38 ± 9 ms, p < 0.001, ΔQTd = 52 ± 9%, in the PTCA group), while in the Thr‐TIMI flow 0/1 group no significant changes were recorded. A percent QTd decrease > 30 s had 96% sensitivity, 85% specificity, and 93% positive and 94% negative predictive value, respectively, for TIMI 2/3 flow. Conclusions : A significant decrease in QT dispersion may provide an additional electrocardiographic index for successful (TIMI 2/3) reperfusion.
TIMI
Cite
Citations (23)
TIMI
Reperfusion Therapy
Cite
Citations (25)
TIMI
Cite
Citations (8)
Objective Evaluation of the benefit and safety of combined thrombolysis and precutaneous coronary intervention in acute myocardial infarction by retrospective analysis. Methods Precutaneous coronary intervention were performed in 45 patients with acute myocardial infarction immediately after thrombolysis. The clinical and angiographic data were compared with that of 31 patients with thrombolysis alone and 74 patients with primary precutaneous coronary intervention in the same period. Results Angiographic data showed that patients with PCI plus thrombolysis had more frequency of TIMI 3 flow than patients with thrombolysis alone (88.9% vs. 74.2%, P=0.087). Patients with PCI plus thrombolysis and with primary PCI had similar frequency of TIMI 3 flow (88.9% vs. 91.9% P=0.404). Clinical data showed that patients with PCI plus thrombolysis had less major adverse cardiovascular events in hospital than patients with thrombolysis alone (4.4% vs. 12.9%, P=0.181). Patients with PCI plus thrombolysis and with primary PCI had similar major adverse cardiovascular events (4.4% vs. 1.4%, P=0.319). Patients with three strategies of treatment had similar mortality (4.4% vs. 6.5% vs. 4.1%). Patients with three strategies of treatment had similar major bleeding events (4.4% vs. 3.2% vs. 1.4%) in hospital. Conclusion Combined thrombolysis and precutaneous coronary intervention maybe surpass thrombolysis alone, at least similar to primary PCI. This strategy of treatment is safe.
TIMI
Cite
Citations (0)
Although pre-interventional thrombolysis has recently been shown to restore early patency and preserve left ventricular function in patients with acute myocardial infarction, the significance of Thrombolysis in Myocardial Infarction (TIMI) grade flow early after thrombolysis remains unclear. Patients were classified into 3 groups according to TIMI grade flow 45 min after thrombolysis; 38 patients with TIMI grade 0 or 1 flow (group T0) and 46 with TIMI grade 2 flow (group T2) additionally received immediate percutaneous coronary intervention (PCI) and 50 patients with TIMI grade 3 flow (group T3) were treated conservatively after thrombolysis. Although the door-to-balloon times did not differ in groups T0 and T2, group T2 had lower peak creatine kinase, a higher rate of complete (≥70%) ST resolution and better regional wall motion at discharge as compared with group T0, similar to group T3 (group T2, group T3 vs group T0; 2,857±1,756, 2,314±1,948 vs 3,779 ±2,214 mU/ml; 57, 72 vs 34%; -1.5±1.6, -1.2±1.6 vs -2.2±1.6; all p<0.01, respectively). These results suggest TIMI grade 2 flow at 45 min after thrombolysis followed by immediate PCI, as well as TIMI grade 3 flow, is associated with greater myocardial salvage than TIMI grade 0 or 1 flow. (Circ J 2003; 67: 238 - 242)
TIMI
Cite
Citations (5)
TIMI
Stroke
Cite
Citations (22)
Background Thrombolysis with streptokinase (STK) is the most widely used reperfusion strategy for ST elevation myocardial infarction (STEMI) in India. Achieving full reperfusion as evidenced by thrombolysis in myocardial infarction (TIMI) flow grade 3 in coronary angiography (CAG) is associated with better outcomes. Recent studies show that hematological indices like neutrophil-lymphocyte ratio (NLR) and mean platelet volume (MPV) estimated before thrombolysis could predict TIMI 3 flow. We studied clinical, electrocardiographic and hematological parameters associated with TIMI 3 flow after thrombolysis with STK. Methods We prospectively studied 201 adult patients with STEMI presenting within 12 hours of onset of chest pain. Before thrombolysis, blood sample was collected for estimating NLR and MPV. Timing of CAG after thrombolysis was decided by consultant cardiologists. Patients were followed up for one month after discharge. Results Of 201 patients, 162 (81%) had relief of chest pain and 131 (65%) had ST segment recovery of ≥50% at 90 minutes after thrombolysis. CAG was performed within median (IQR) of four (3-5) days after thrombolysis. TIMI 3 flow was observed in 112 (56%) patients. NLR and MPV had no significant association with TIMI 3 flow. In multivariable analysis, ST-segment recovery of ≥50% at 90 minutes was associated with TIMI 3 flow (adjusted OR 3.47, 95% CI: 1.84-6.53, P = <0.001). Of 198 patients followed up for one month after discharge, 13 (6.5%) died. Conclusions In patients with STEMI, ST-segment recovery of ≥50% at 90 minutes after thrombolysis with STK predicted TIMI 3 flow independently. NLR and MPV values were not predictive of TIMI 3 flow.
TIMI
Cite
Citations (1)
TIMI
Reperfusion Therapy
Cite
Citations (36)
TIMI
Cite
Citations (49)
Objective To investigate the curative effect and complications of intra-arterial thrombolysis (IAT) with recombinant tissue plasminogen activator(rt-PA) for patients with acute ischemic stroke, and analyse the factors related the outcomes.Methods 12 patients were treated by IAT with rt-PA in our hospital from Oct.2002 to Oct.2003. Primary neuroradiological assessment was performed with CT in all patients. Mechanical disruption of clot remnants were attempted after rt-PA infusion. Angiographic recanalization was classified according to thrombolysis in myocardial infarction (TIMI) grades. Clinical evaluation was undertaken 20 d after thrombolysis with classification of modified rankin scale (MRS) scores, good for 0 to 3 and poor for 4 to 6. Results Before thrombolysis the scores for TIMI 1 in 1 case was 8.33% and TIMI 0 in 11 cases was 91.67%. The rates of complete or partial recanalization just after IAT were 75%(9/12), less or no recanalization were 25%(3/12). Good outcome in 66.7%(8/12), poor outcome in 33.33%(4/12). Cerebral hemorrhage occurred in 1 case.Conclusions Intra-arterial thrombolysis ( IAT) with rt-PA is feasible and safe in treatment of acute ischemic stroke.
TIMI
Stroke
Cite
Citations (0)