• International Journal of Computer Science & Network Security
• /
• 제24권7호
• /
• pp.11-23
• /
• 2024

Triage is a practice of accurately prioritizing patients in emergency department (ED) based on their medical condition to provide them with proper treatment service. The variation in triage assessment among medical staff can cause mis-triage which affect the patients negatively. Developing ED triage system based on machine learning (ML) techniques can lead to accurate and efficient triage outcomes. This study aspires to develop a triage system using machine learning techniques to predict ED triage levels using patients' information. We conducted a retrospective study using Security Forces Hospital ED data, from 2021 through 2023 during Hajj period in Saudia Arabi. Using demographics, vital signs, and chief complaints as predictors, two machine learning models were investigated, naming gradient boosted decision tree (XGB) and deep neural network (DNN). The models were trained to predict ED triage levels and their predictive performance was evaluated using area under the receiver operating characteristic curve (AUC) and confusion matrix. A total of 11,584 ED visits were collected and used in this study. XGB and DNN models exhibit high abilities in the predicting performance with AUC-ROC scores 0.85 and 0.82, respectively. Compared to the traditional approach, our proposed system demonstrated better performance and can be implemented in real-world clinical settings. Utilizing ML applications can power the triage decision-making, clinical care, and resource utilization.

• 융합정보논문지
• /
• 제10권4호
• /
• pp.98-103
• /
• 2020

본 연구에서는 KTAS(Korean Triage and Acuity Scale) 결과가 분류를 시행한 주체의 직종에 따른 차이가 있는지를 알아보고자 한다. 2016년 1월 1일부터 2017년 12월 31일까지의 응급의료기관으로 내원한 환자 자료 중, 국가응급진료정보망으로 전송된 자료 총 10,960,359건을 분석하였다. 분류자 직종은 전문의, 전공의, 인턴, 일반의, 간호사, 응급구조사였다. 최초 중증도 분류와 최종 중증도 분류 결과의 일치율은 일반의가 98.9%로 가장 높았고, 인턴이 80.2%로 가장 낮았다. 과대 분류에서는 일반의가 0.6%로 가장 낮았고, 인턴은 16.0%로 가장 높았다. 또한 과소 분류는 전문의와 응급구조사가 0.4%로 가장 낮았고, 인턴이 3.8%로 가장 높았다. 중증도 분류 결과는 직종별 유의미한 차이가 있었다(p<0.001). 중증도 분류는 환자의 예후에 영향을 미치는 요인 중 하나로 직종별, 숙련도에 따라 그 결과가 달라져서는 안 된다. 때문에 정확한 중증도 분류를 위한 분류자의 역량 강화가 필요하다.

• 대한간호학회지
• /
• 제49권2호
• /
• pp.171-180
• /
• 2019

Purpose: The Korean Triage and Acuity Scale (KTAS) is a tool used to classify the severity and urgency of emergency department (ED) patients, focusing on their symptoms. In consideration of the importance of the KTAS, a web-based learning program has emerged as a new mode of education; it enables ED triage nurses to access it anytime and anywhere, and according to their own learning abilities. This study aimed to develop a web-based KTAS learning program and evaluate its effects on self-efficacy and triage performance ability in ED nurses. Methods: A quasi-experimental design with a non-equivalent control group pretest-posttest was used. The conceptual framework was Bandura's self-efficacy theory. There were 30 participants in the experimental group and 29 in the control group. The experimental group attended an orientation and 4 sessions of a web-based KTAS learning program. The learning program lasted 280 minutes over five weeks, consisting of 40 minutes of orientation and four 60-minute sessions. Results: The scores of self-efficacy, triage performance ability in KTAS level, and chief complaints significantly increased in the experimental group compared to the control group. In addition, the numbers of under-triage in KTAS significantly decreased in the experimental group in comparison to the control group. Conclusion: The results suggest that the learning program was effective in improving ED nurses' level of self-efficacy and triage performance ability (KTAS level and KTAS chief complaint). Accordingly, the web-based KTAS learning program can be applied as an education intervention to improve ED nurses' triage skill.

• 중환자간호학회지
• /
• 제5권2호
• /
• pp.61-71
• /
• 2012

Purpose: The aim of this study was to measure the inter-rater reliability of Emergency severity index (ESI) version 4 among triage nurse. Methods: This study was carried out from August 11, 2010 to September 7, 2010 in a regional emergency department. Data collection was done by ten triage nurses who trained ESI v.4. Two research nurses and ten triage nurses scored the ESI version 4 to the patients as references, independently. We calculated the weighted kappa between the triage nurses and research nurses to evaluate the consistency of the ESI v.4. Results: A total of 233 patients were enrolled in this study. Classification of ESI level was as follows - level 1 (0.4%), level 2 (21.0%), level 3 (67.8%), level 4 (9.4%), and level 5 (1.3%). Inter-rater reliability by weighted kappa was 0.79 (95% Confidence Interval= 0.74-0.83) and agreement rate was 87.1%. Under-triage rate by triage nurse was 6.0% and over-triage rate was 6.9%. Conclusion: For this study, inter-rater reliability was measured good level between triage nurses and research nurses in Korean single ED.

• Journal of Trauma and Injury
• /
• 제33권1호
• /
• pp.13-17
• /
• 2020

Purpose: Accurate and appropriate prehospital field triage is essential for a trauma system. The Korean trauma system (established in 2014) uses the trauma field triage algorithm of the United States Centers for Disease Control and Prevention (CDC). This study evaluated the suitability of the CDC field triage criteria for major trauma cases (injury severity score >15) in Korea. Methods: This retrospective cohort study evaluated trauma patients who presented at the authors' regional trauma center from January 1 to May 7, 2017. The undertriage and overtriage rates of each CDC field triage step were calculated. Receiver operating characteristic curves were constructed, and the area under the curve (AUC) was evaluated for each step. Results: Among the 1,009 enrolled patients, 168 (16.7%) had major trauma. The undertriage/overtriage rates of each step (steps I, II, III, and IV) of CDC field triage were 9.2%/47.4%, 6.3%/50.8%, 4.5%/59.4%, and 5.3%/78.9%, respectively. The AUC values of each CDC triage step were 0.722, 0.783, 0.791, and 0.615, respectively. The AUC values of the separate components of each step (physiologic criteria, anatomic criteria, mechanism-of-injury criteria, and special considerations) were 0.722, 0.648, 0.647, and 0.456, respectively. Conclusions: The CDC field triage system is acceptable, but not ideal, for Korean trauma care. If we follow the protocol, it would be preferable to omit step IV. The Korean Triage and Acuity Scale may be a good indicator for in-hospital triage. However, a new triage protocol that is simple to estimate on-scene while having good performance should be developed.

• Journal of Medicine and Life Science
• /
• 제17권1호
• /
• pp.7-15
• /
• 2020

The field triage guidelines have been widely implemented in the Korean trauma system. This study aimed to evaluate and validate whether it is reliable to use the field triage guidelines for predicting severe traumatic brain injury (TBI) and traumatic spinal injury (TSI) patients. This study retrospectively analyzed in-hospital cohort registries of all TBI and TSI patients, who visited the emergency department (ED) of the Jeju National University Hospital from 1 January 2013 to 31 December 2015. The primary outcome was defined as TBI and TSI patients with an injury severity score (ISS)>15. Secondary outcomes were defined as cases in which one or more of the following conditions: in-hospital death, ISS>15, admission to the intensive care unit, emergency surgery. We enrolled 14,889 TBI and TSI patients who visited ED, of which 7,966 (53.5%) were triage positive. The overall sensitivity, specificity and area under the curve (AUC) of the full cumulative field triage guidelines step's model (Step 1+3+4 criteria) for primary outcome were 82.8%, 47.0%, and 0.646, respectively. In the results for secondary outcomes, the specificity did not show a significant difference, but the sensitivity decreased to 66.5% and AUC to 0.568. The results of this study suggest that further optimization of the field triage guidelines is needed to identify high-risk TBI and TSI patients.

• Journal of Trauma and Injury
• /
• 제27권4호
• /
• pp.126-132
• /
• 2014

Purpose: The field trauma triage for injured patients is essential for trauma care system. In this study, agreement of patient evaluation between by prehospital EMS personnel and by hospital staffs and the appropriateness of prehospital triage were evaluated. Methods: This observational study was conducted from September to October 2012 for 5 weeks. During this period, EMT evaluated patient's severity according to guideline for field triage and recorded. Same guideline was applied in 26 hospitals for patients with EMS use. Kappa statistics were used to measure agreement for each item of guideline. Finally, over-triage and under-triage rate of EMT were calculated. Results: During study period, total 3,106 patients were transferred to 26 hospital emergency departments with EMS use. Kappa statistics for "vital signs" items were 0.45 for mentality lower than V and 0.44 for systolic blood pressure lower than 90 mmHg as a moderate agreement. In "anatomy of injury" items Kappa statistics were very low. In "mechanism of injury" items Kappa statistics were 0.28 for high-rise fall down and 0.27 for high energy traffic accident but in other items Kappa statistics were very low. 362 patients (12.0%) were over-triaged and 281 patients (9.3%) were under-triaged. Conclusion: Field triage can be applied but need to evaluate and modify in order to become accurate and sensitive for decision of transportation.

• Pediatric Gastroenterology, Hepatology & Nutrition
• /
• 제21권4호
• /
• pp.278-288
• /
• 2018

Purpose: Dehydration is a paediatric medical emergency but there is no single standard parameter to evaluate it at the emergency department. Our aim was to evaluate the reliability and validity of capillary refilling time as a triage parameter to assess dehydration in children. Methods: This was a prospective pilot cohort study of children who presented to two paediatric emergency departments in Italy, with symptoms of dehydration. Reliability was assessed by comparing the triage nurse's measurements with those obtained by the physician. Validity was demonstrated by using 6 parameters suggestive of dehydration. Comparison between refilling time (RT) and a validated Clinical Dehydration Score (CDS) was also considered. The scale's discriminative ability was evaluated for the outcome of starting intravenous rehydration therapy by using a receiver operating characteristic (ROC) curve. Results: Participants were 242 children. All nurses found easy to elicit the RT after being trained. Interobserver reliability was fair, with a Cohen's kappa of 0.56 (95% confidence interval [CI], 0.41 to 0.70). There was a significant correlation between RT and weight loss percentage (r-squared=-0.27; 95% CI, -0.47 to -0.04). The scale's discriminative ability yielded an area under the ROC curve (AUC) of 0.65 (95% CI, 0.57 to 0.73). We found a similarity between RT AUC and CDS-scale AUC matching the two ROC curves. Conclusion: The study showed that RT represents a fast and handy tool to recognize dehydrated children who need a prompt rehydration and may be introduced in the triage line-up.

• Child Health Nursing Research
• /
• 제5권2호
• /
• pp.136-150
• /
• 1999

The subjects, under 18 old, 2,694 children who visited ER during 1998, were surveyed by the Triage and other statitical data. The results were as follows 1. The male to female ratio was 1.7 : 1, and the toddler(1-4 old) was the majority. 2. Triage ; critical 1.3%, acute 14.6%, urgent 29.5%, nonemergent 54.6%. 3. The reasons of visiting ER ; 1) The children had diseases(46.5%), injury (36.9%), TA(5.6%) and toxication(1.0%). 2) In diseases, male to female ratio was 1.5 : 1 and in injury, male to female was 2 : 1. 3) Among the children having in TA, 12-18 old groups was the majority(34.9%). 4. The time of visiting : the 20:00 - 22:00 was the majority(16.9%). 5. By monthly and seasonal distribution ; Jan. (9.7%), Mar. and May(9% respectively, Dec. and July(6.7%) respectively. The children who visted ER in spring and autumn showed higher portion than those of summer and winter. 6. Results ; admission(27.4%), discharge(68.4%), operations(2.8%), and DOA and DAA(0.4%), The mortality of the infancy and toddler groups was 83.3%. The infancy group showed the highest rate of admission. 7. The time of staying in ER ; 1-2 hrs was the major group(23.3%) and the average was 4.6 hours. 8. By clinical departments ; Ped. was 34.4%, PS was 20.8%, Dental Surgery was 10.3% and Dermatology was 0.9%. 9. The types of visting ; the group who visited by themselves was 80%, transfer from the primary and secondary clinic was 17% and OPD was 3.0%. 10. The traffic means . by the own cars and taxi were 87.6%, by hospital ambulance was 6.1% and by 119 ambulance was 4.3%.

• Journal of Trauma and Injury
• /
• 제21권2호
• /
• pp.71-77
• /
• 2008

Purpose: The Glasgow Coma Scale (GCS), though it is widely used for triage, has been criticized as being unnecessarily complex. Recently, a 3-point Simplified Motor Score (SMS, defined as obeys commands=2; localizes pain=1; withdrawals to pain or worse=0) was developed from the motor component of the GCS and was found to have a similar test performance for triage after traumatic brain injury when compared with the GCS as the criterion standard. The purpose of this study was to validate the SMS. Methods: We analyzed the patients who visited Kyungpook National University Hospital emergency center after traumatic brain injury from 2006 January to 2006 June. The test performance of the GCS, its motor component, and SMS relative to three clinically relevant traumatic brain injury outcomes (abnormal brain CT scans, Abbreviated Injury Scale $(AIS){\geq}4$, and mortality) were evaluated with areas under the receiver operating characteristic curves (AUCs). Results: Of 504 patients included in the analysis, 25.6% had an abnormal brain CT scans, 13.1% had $AIS{\geq}4$, and 5.0% died. The AUCs for the GCS, its motor component, and SMS with respect to the abnormal CT scans were 0.776, 0.715, and 0.716, and respectively, those for $AIS{\geq}4$ and mortality, were 0.969, 0.973, and 0.968, and 0.931, 0.909, and 0.909, respectively. Conclusion: The 3-point SMS demonstrated similar test performance when compared with the 15-point GCS score and its motor component for triage after traumatic brain injury in our populations.