Global Journal of Nursing & Forensic Studies
Open Access

American Stroke Association; World Health Organisation; Surgical Emergency Care

Introduction

Evidence for insufficient understanding and management of pain is mounting, perhaps even more so in the prehospital setting. The challenges that emergency care workers face on a regular basis in addressing prehospital pain are yet unknown [1]. In the Western Cape of South Africa, this study aims to learn more about the assessment and management of acute prehospital pain. Medical imaging workflows are supported by hospital information systems (HIS), electronic health record (EHR) systems, radiological information systems (RIS), picture archiving and communication systems (PACS), and, in rare cases, teleradiology systems [2]. These numerous information systems disseminate a great deal of crucial information on stroke patients, thus their databases are significant sources of information for the long-term development of the care procedures [3].

This study's main area of interest is the modelling of highly unbalanced tasks utilising improved artificial learning techniques. A model of balanced datasets serves as the basis for this strategy, which is then followed by a stage in which ML classifiers are applied and their performance metrics are compared. In this paper, the multiclass classification problem was examined using ensemble learning techniques. This approach was chosen partly due to how straightforward and effective it is at addressing issues with class disparity. The reality of ACS is that, unlike its other cousins (i.e., NSTEMI and UA), which require additional decision support, STEMI frequently has a simple EKG-based diagnosis [4].

According to the World Health Organization (WHO), a stroke is a regional or generalized neurological impairment of vascular origin that can be fatal and is caused by an obstruction in the blood supply to the brain. An estimated 15 million people get strokes annually, and six million of them pass away. This data validates the prevalent concern about enhancing the treatment of acute stroke victims. Because of all the detrimental physical, psychological, and social implications, stroke continues to be one of the leading causes of long-term impairment [5].

The benefits of reperfusion therapy, however, certainly depend on the length of the treatment period; the likelihood of a good outcome for patients receiving intravenous thrombolysis compared to placebo decreases as the length of the treatment period increases [6]. Thus, it is essential to reduce the interval between the onset and the beginning of treatment. Reperfusion therapy should be administered in this case within 4.5 hours of the onset of symptoms and within an hour of hospital admission, according to the American Heart Association/ American Stroke Association (AHA/ASA) recommendations for the early care of acute ischemic stroke [7].

Methods

Between January 2017 and August 2020, we collected considerable clinical electronic health data on patient interactions with clinical concerns for ACS from a large metropolitan emergency department (ED). In order to improve the data quality, we applied an analytical framework with a variety of well-developed methodologies to address missing values, dimensionality reduction, and data imbalance [8].

To categorise patients based on whether or not their symptoms were brought on by ACS, we used ensemble learning algorithms. The accuracy, sensitivity, precision, F1-score, and the area under the receiver operating characteristic (AUROC) were used to evaluate the model's performance [9].

Result

Ninety-two people, averaging 72.41 years old, were examined. The most common diagnosis (76, 83.52%) was colorectal cancer that had progressed to the right (37, 41.51%), left (35, 39.33%), and rectum (17, 19.10%). Ischemia (2, 2.17%), blockage without perforation (61, 67.03%), and perforation (25, 27.17%) were the clinical symptoms. Overall, mortality was 6.52%. The three main causes of death were pulmonary embolism (one, 16.67%), respiratory failure (three, 50%), and septic shock (three, 50%). Morbidity rates were 41.30% in surgical cases and 26.08% in medical cases. Among the procedures carried out for all causes were resection with primary anastomosis (62, 71.26%), Hartmann's procedure (11, 12.64%), and loop colostomy (12, 13.79%).

The typical procedure took 159.86 minutes. The risk of death during emergency colorectal surgery increased by 7.6, 16.5 and 0.08 times, respectively, depending on whether the patient had pre-existing heart disease, clinical perforation, or depended on a ventilator [10].

Conclusion

Our findings imply that our proposed strategy may enable us to identify critical variables and offer extremely accurate and sensitive diagnoses of NSTEMI, UA, and non-ACS. These discoveries will probably change how ACS patients are now identified and will lessen the chance of a potentially catastrophic STEMI developing. Patients with stable NSTEMI that was misdiagnosed are at extremely high risk of suffering a major adverse cardiovascular event. Early diagnosis of stable NSTEMI and UA offers a major therapeutic benefit for the alternatives involved in addressing these conditions (such as an early invasive approach like angiography and cardiac revascularization with percutaneous coronary intervention). Our work uses information from a single centre to construct a diagnostic decision support tool has a significant therapeutic benefit. Our study creates a diagnostic decision support tool using data from a single Centre.

Acknowledgement

None

1. Liu CH, Huang YC (2011) Comparison of STEMI and NSTEMI patients in the emergency department. J Acute Med 1:1-4.

Indexed at, Google Scholar, Crossref

2. Gao CZ, Ma QQ, Wu J, Liu R, Wang F, et al. (2018) Comparison of the effects of ticagrelor and clopidogrel on inflammatory factors, vascular endothelium functions and short-term prognosis in patients with acute ST-segment elevation myocardial infarction undergoing emergency percutaneous coronary intervention: a pilot study. Cell Physiol Biochem 48:385-396.

Indexed at, Google Scholar, Crossref

3. Granger CB, Goldberg RJ, Dabbous O, Pieper KS, Eagle KA, et al. (2003) Predictors of hospital mortality in the global registry of acute coronary events. Arch Intern Med 163:2345-2353.

Indexed at, Google Scholar, Crossref

4. McNamara RL, Kennedy KF, Cohen DJ, Diercks DB, Moscucci M, et al. (2016) Predicting in-hospital mortality in patients with acute myocardial infarction. J Am Coll Cardiol 68:626-635.

Indexed at, Google Scholar, Crossref

5. Freixa X, Bellera N, Ortiz-Perez JT, Jimenez M, Pare C, et al. (2012) Ischaemic postconditioning revisited: lack of effects on infarct size following primary percutaneous coronary intervention. Eur Heart J 33:103-112.

Indexed at, Google Scholar, Crossref

6. Huang W, Ying TW, Chin WLC, Baskaran L, Marcus OEH, et al. (2022) Application of ensemble machine learning algorithms on lifestyle factors and wearables for cardiovascular risk prediction. Sci Rep 12:1033.

Indexed at, Google Scholar, Crossref

7. Madeline MJ, Prashant M, Sally KS, Brandon CK, Mohsen S (2022) Optimizing Pediatric Patient Safety in the Emergency Care Setting. Pediatrics 150:e2022059673.

Indexed at, Google Scholar, Crossref

8. Steinberg EM, Valenzuela-Araujo D, Zickafoose JS, Kieffer E, DeCamp LR (2016) The “battle” of managing language barriers in health care. Clin Pediatr (Phila) 55:1318-1327.

Indexed at, Google Scholar, Crossref

9. Gallagher RA, Porter S, Monuteaux MC, Stack AM (2013) Unscheduled return visits to the emergency department: the impact of language. Pediatr Emerg Care. 29:579-583.

Indexed at, Google Scholar, Crossref

10. Zamor R, Byczkowski T, Zhang Y, Vaughn L, Mahabee-Gittens EM (2020) Language barriers and the management of bronchiolitis in a pediatric emergency department. Acad Pediatr 20:356-363.

Indexed at, Google Scholar, Crossref