If you have problem reading this email, you can cilck this LINK
or you can access our facebook or website from the links below.

To subscribe to our newsletter, please click
Facebook Facebook
Website Website
Email Email
Issue 016--Jan 2021

Hong Kong Society
for Emergency Medicine & Surgery
Message from the President...
Due to the ongoing threats of COVID-19, travelling and gathering restrictions, the Organizing Committee of ACEM 2021 has decided to change the conference into virtual more
From Editors...

In this issue we will explore the application of AI in Emergency Medicine. In members area, we will also bring to you an introduction to CUHK Medical more 
Council News...                 
Due to the COVID-19 pandemic, the SSEM this year has gone virtual and the half-day hiking activity and HK historical tour was more
From Members
Members Area
After years of expectation, the CUHK Medical Centre starts operating this month. As a newly established private hospital, CUHK adopts use of smart technology in its facilities and medical services. Dr Marc Yang Li-Chuan  will introduce to us this new hospital and his work in the emergency medical more 
Private EM Writes

The treatment of acute ischemic stroke has achieved much advancement in recent years and is no longer limited to aspirin prescription as in old days. The new chairman of Private Emergency Physician Chapter, Dr Lam Ka Keung will discuss the endovascular intervention in acute ischemic more

Nurse Corner
The pandemic of COVID 19 has posed severe challenge to frontline medical professionals as well as managerial staff. Ward manager of KWH A&E Ms Leung Shun Wah Ava will take us through the challenges for nurse managers during the COVID-19 pandemic and how she deals with the difficulties thus more 
Advance in EM
With the advance in technology, artificial intelligence has been increasingly applied in many aspects of our daily lives. It is not surprising that its use in medicine will become more prevalent with time. Our editor, Dr Wong Kwun Bun will introduce the application of artificial intelligence in emergency more
Message from the President      Dr Ben Kuang-An WAN
2020 is an unprecedented, difficult year. COVID-19 is still ravaging the world. At the time of writing, the 4th wave of local COVID-19 outbreak has yet to reach its peak. While many people celebrate for the development of COVID-19 vaccines, widespread skepticism about the efficacy and safety profile exist. Recently, a novel SARS-CoV-2 variant was first reported in UK and subsequently in several other countries as well. Preliminary reports by UK suggest that the
new variant is up to 70% more transmissible than the prevailing SARS-CoV-2 virus, and this will pose additional threats to public health.
Emotional wellbeing of healthcare professionals is being threatened by the stress from work and uncertainties of COVID-19. It is vital for us to maintain a healthy work-life balance and take good care of ourselves. Pay attention to those easily forgotten pillars of self-care – diet, sleep and exercise. These are not only important for physical wellbeing but also good from a mental health perspective. Despite the need of social distancing, we should still stay connected to others by virtual means and reinforce each other that no one is alone. Stay strong and healthy, and we will get through COVID-19 altogether.
Here I would also like to update you on the 11th Asian Conference on Emergency Medicine (ACEM 2021). Due to the ongoing threats of COVID-19, travelling and gathering restrictions, the Organizing Committee (OC) has decided to change the conference format. The ACEM 2021 will be the first virtual edition of ACEM. This was not an easy decision, but definitely an important move to safeguard the health of our participants. The OC is working hard to create an exciting opportunity for delegates around the world to share scientific contents without the need to meet face-to-face. A highly interactive digital platform will be developed to facilitate networking among delegates and sponsors.

Renowned local and international speakers will be invited to present the latest educational updates on a wide spectrum of EM topics and share their unique insights on the conference theme “Global Challenges - Asian Solution”. Please refer to the ACEM 2021 website ( for more details.
From Editors Editors in chief Dr Chor-man Lo, Dr Sam Siu-ming Yang

Dr Ho-yin Chan, Dr Wendy Cheng,
Dr Louis Chin-pang Cheung, Dr Kwun-bun Wong
Mr Chun Pong Leung

The year 2020 was difficult because the COVID-19 pandemic brought drastic changes to our ways of life, and created great challenges for every one of us. The Scientific Symposium on Emergency Medicine (SSEM) held in October 2020 was conducted through webinar. There was no audience at the venue but cameras and screens.

Speakers spoke in front of cameras, discussion was conducted through on-line chat and group photo was taken by screen capture of the speakers and moderators. Although the element of face-to-face interaction was lacking, our members still showed enthusiastic support to this event through active participation. The experience of organizing this SSEM prepared us for the Asian Conference on Emergency Medicine, which is going to be held in December 2021.
The “Advance in Emergency Medicine” will introduce the topic on “artificial intelligence” (AI). With the rapid development of computer technology and big data, AI is gaining importance in every aspect of human life. Its use in medical field is becoming more common nowadays. In emergency medicine, it has seen its use in triage, image diagnosis and prognostication. While it is a powerful tool in helping patient management, worry arises as to whether it will replace nurses and even doctors. It is inevitable that part of our jobs will be replaced by the new technology. However, these will mostly be repetitive tasks that do not involve human interaction. AI will have difficulties in understanding feelings, interpreting emotion, and expressing empathy, at least in the foreseeable future. In medicine, communication skill is, and will be, an irreplaceable factor that constitutes good patient care.
Moving into year 2021, more good news will emerge with the development of effective vaccine against SARS-CoV-2. It is hoped that mass vaccination can be completed early so as to bring our lives back to normal, or at least near-normal. Let’s stay perseverant in our fight against the pandemic.

Council News

Dear Members of HKSEMS,
Welcome to the 4th issue of 2020. Let’s refresh the activities organized by the Council in these 3 months!!
SSEM 2020
SSEM 2020 organized by HKCEM, and supported by HKSEMS, HKCEN and HKENA was held virtually on 24th October 2020.
This year, with the theme of “New Frontier of Pediatric Emergency Medicine: Collaboration and Synergy”, the symposium featured inspirational plenary sessions and multi-disciplinary tracks of presentations.
Exciting moments can be reviewed at !
This is our first ever exchange forum with Taiwan Doctors, COVID-19 is never an obstacle to us to stay connected. Special group photos were taken with the speakers from Taiwan (on the screen).
This is a memorable year for the medical unit of Government Flying Service (GFS). They are celebrating their 20th establishment anniversary. A 20th Anniversary seminar was held, and 4 speakers shared the past and look into the future of the MU, GFS. Dr. Willis Kwok (Senior Auxiliary Member, the head of Auxiliary Unit) was leading us to walk through the past 20 years since the establishment of the MU.
Half-day hiking activity and HK historical tour 2020

This hiking activity was originally scheduled on 14th December 2020, but was postponed due to the 4th wave of COVID-19. We shall resume upon the stepping down of COVID-19.
Upcoming activities

Emergency Medicine Training Programme for Chinese Medicine Practitioners

HKSEMS is running the certificate course for CMP in EM since 2015, the course has been fully revamped in 2019 in order to help the CMP to prepare for the opening of the Chinese Medicine Hospital. It will be run in virtual due to COVID-19.

ACEM 2021
Save the date !!
HKSEMS will be the organizer for the 11th ACEM.
This will be the ASEM’s first ever virtual conference!
Although the conference will go virtual due to COVID-19, it will not be an obstacle for organizing a great conference to share and stay connected!
Further information will be released in Facebook, IG and Twitter soon!
Please stay tune and join us!!
Members Area Dr. Marc Yang Li-Chuan
  Specialist in Emergency Medicine
Emergency Medicine Centre CUHKMC
Working in the CUHK Medical Centre
I have the privilege to take part in the commissioning of the new “The Chinese University of Hong Kong (CUHK) Medical Centre (CUHKMC)” since 2019.
CUHKMC is located within the campus of CUHK, right next to the train station. It is expected to have more than 500 inpatient beds and 28 operating theatres when fully functional. The Emergency Medicine Centre (EMC) in the CUHKMC is one of the 16 special medical centres within the hospital. Other centres include cardiac centre with cardiac catheterisation laboratory, endoscopy centre, assisted reproduction centre, etc. The intensive care unit of the hospital contains 8 beds with 2 of them having negative pressure isolation facilities.
My main involvement in the preparation for the opening of the hospital is in the EMC and the ICU.
A lot of information about the hospital can be found on the hospital website. Hence, I would like to share with the readers some of the interesting quirks of the hospital that are less publicised.  
  1. Hospital uniform. In addition to having an antimicrobial nano-coating, each piece of uniform garment is tagged with RFID. The size of each staff’s uniform was stored electronically in the staff card. Uniform will then be dispensed and collected automatically by machines in changing rooms and doctors’ lounges. My experience with using the uniform is good. They are comfortable, breathable, and any spilled blood or dirt just slides off the surface. They are well suited for the busy and hectic emergency department.
  2. Teaching facilities. The hospital is home to many teaching facilities. There is an auditorium large enough for a whole class of medical students. There are multiple rooms for conference, tutorial and seminar.
  3. Computer system. The Hospital Information System was commissioned from the ground up and encompasses all the major computer systems used in the hospital. This include consultation notes, inpatient progress notes, medications order entry, laboratory and radiology request, patient charges, etc. When fully functional, no paper record is necessary. The system is undergoing intensive testing for major bugs. In the EMC, the system integrates doctor scheduling, registration, triage, queue management, medication prescription and dispensing. Electronic display outside each room in the EMC automatically displays the information of the doctor and also guides patient into the correct room.
    A patient's phone application is also being tested. The app can remind the patient on appointments, medication compliance, checking of investigation results, and navigation and waiting time alert in the hospital.
  4. Medication dispensing. In addition to computerised prescription, the hospital is also equipped with automated medication handling system. Each ward and unit has an automatic dispensing cabinet that dispense non-ward-stocked medications after electronic prescription by the doctor. Non-ward stocked items will be dispensed by pharmacy in unit-dose packages and sent to ward via pneumatic tubes. For outpatients, we have Multiple Dose Dispensing where chronic medications would be packaged in ready to use pouches by administration time. This allow more accurate dispensing, better patient compliance and re-use of excess medications.
  5. One stop service. In the EMC, patients are registered, triaged, assessed clinically, payment, and medication collection, all are carried out within this complex, without going around.
 As for the EMC, it is planned to be fully operational in late 2021 and will provide 24 hours’ service. Prior to obtaining full Department of Health Accident and Emergency Services license, the unit will operate with shorter opening hours. The unit will be staffed by specialist emergency physicians and emergency nurses. The emergency medicine specialists in the centre are also responsible for inpatient resuscitation. The EMC is integrated tightly with the hospital’s patient deterioration detection system. Physically, the EMC consists of two resuscitation bays, and eight consultation rooms. Among the consultation rooms, two of them have negative pressure isolation capability; and three of them are designed to be child friendly. At the writing of this article, the EMC has operated as an OPD for one week, with minimal hiccups - most of which are related to the computer system. Most patients are very satisfied with our services.
Private EM Writes 
Dr Lam, Ka Keung
  Chairman of Private Fellows’ Chapter, HKCEM
Consultant, Hong Kong Baptist Hospital, Out-patient Centre
New Era of Care: Endovascular Intervention in Acute Ischemic Stroke
Treatment of acute ischaemic stroke has been advanced in recent decade. In my junior days more than twenty years ago, treatment was limited to conservative approaches including aspirin, better medical condition control and rehabilitation along with supportive measures. The concept of “Time is Brain” to save the viable
penumbral brain tissue had been taught, but nothing materialized in practice. 
The implication of thrombolytic agent together with intensive care changed treatment into a more active role. However, it was hindered by harsh criteria of ultra-short presentation time of 3 to 4.5 hours from symptom onset1.
Five landmark randomized controlled trials (RCT) in 2010-2014 (MR CLEAN2, EXTEND-IA3, SWIFT PRIME4, REVASCAT5 and ESCAPE6) provided best evidence supporting endovascular mechanical thrombectomy in the treatment of acute ischaemic stroke by acute large vessel occlusion (LVO) in anterior cerebral circulation within 6 hours of symptom onset. The American Heart Association/American Stroke Association (AHA/ ASA) 2015 guidelines classified this intervention with stent retriever as Class I recommendation with level A evidence1. Furthermore, although there was limited data for this intervention in blood vessels other than middle cerebral artery, the AHA/ ASA 2015 guidelines still suggest mechanical thrombectomy may be a reasonable treatment for acute occlusion of the anterior cerebral arteries, vertebral arteries, basilar artery, or posterior cerebral arteries (Class IIb, Level C evidence)1
The presence of collateral arteries serves an important role in acute stroke. Diagnostic imaging of Computer Tomography Perfusion scan or Magnetic resonance imaging with diffuse weight imaging (MRI with DWI) provide neurovascular surgeon/ vascular interventionist more information to decide whether or not to proceed for thrombectomy on those who present late. The ESCAPE trial already showed that thrombectomy was beneficial up to within 12 hours of symptoms onset.
Two more RCTs published in 2018: DEFUSE 3 and DAWN further extended symptom onset time up to 16 and 24 hours respectively7,8.
The latest AHA/ ASA guidelines in 2019 stated “in selected patients with acute ischaemic stroke (AIS) within 6 to 16 hours of last known normal who have LVO in the anterior circulation and meet other DAWN or DEFUSE 3 eligibility criteria, mechanical thrombectomy is recommended” (Class I, level A evidence)1. And “In selected patients with AIS within 16 to 24 hours of last known normal who have LVO in the anterior circulation and meet other DAWN eligibility criteria, mechanical thrombectomy is reasonable.” (Class IIa Level B evidence)1.
This is a break-through against the time-barrier. Those with wake up stroke or delayed presentation may benefit from the intervention, provided they fit into criteria, although they are minority.
However, the basics do not change: Time is Brain. The occluded artery must be recanalized as soon as possible. The key is early diagnosis to start the chain of survival. 
Two cases of AIS are illustrated here, one was acute basilar artery occlusion and the other was proximal middle cerebral artery occlusion attending a few years ago. Both had endovascular thrombectomy procedures performed in private hospital. 
Case 1: A young man in his 20s presented with severe headache, repeated vomiting and hemianopia. Non-contrast Computer Tomography of Brain (NCCT brain) showed acute left occipital infarct and basilar artery thrombosis. CT angiogram was performed immediately after NCCT brain, which revealed acute basilar artery occlusion:
Occluding clot was retrieved by thrombectomy. Post-procedure Magnetic Resonance Angiography (MRA) revealed recanalization of basilar artery. Patient recovered well and was discharged with residual hemianopia.
Case 2: A 60 years old male presented with acute right hemiplegia (muscle power 0/5) and MRA of brain showed left proximal middle cerebral artery occlusion:
Cerebral arteriogram showed before (Left) and after (middle) clot retrieved (right) by thrombectomy. Patient was discharged with residual facial asymmetry without limb weakness.
  1. Powers WJ, Rabinsteinet AA, Ackerson T, Adeoye OM, Bambakidis NC et al. Guidelines for the early management of patients with acute ischemic stroke: 2019 update to the 2018 guidelines for the early management of acute ischemic stroke: A guideline for healthcare professionals from the American Heart Association/American Stroke Association.  Stroke. 2019;50:e344–e418.
  2. Berkhemer OA, Fransen PSS, Beumer D, van den Berg LA, Lingsma HF, et al. A randomized trial of intra-arterial treatment for acute, ischemic stroke. N Engl J Med 2015;372:11-20.
  3. Campbell BCV, Mitchell PJ, Kleinig TJ, Dewey HM, Churilov DL et al. Endovascular therapy for ischemic stroke with perfusion-imaging selection. N Engl J Med 2015;372:1009-18.
  4. Saver JL, Goyal M, Bonafe A, Diener HC, Levy EI et al. Stent-retriever thrombectomy after intravenous t-PA vs. t-PA alone in stroke. N Engl J Med 2015;372:2285-95.
  5. Jovin TG, Chamorro A, Cobo E, de Miquel MA, Molina CA et al. Thrombectomy within 8 Hours after symptom onset in ischemic stroke. N Engl J Med 2015;372:2296-306
  6. Goyal M, Demchuk AM, Menon BK, Eesa M, Rempel JL et al. Randomized assessment of rapid endovascular treatment of ischemic stroke. N Engl J Med 2015;372:1019-30.
  7. Albers GW, Marks MP, Kemp S, Christensen S, Tsai JP et al. Thrombectomy for stroke at 6 to 16 hours with selection by perfusion imaging. N Engl J Med 2018;378:708-18.
  8. Nogueira RG, Jadhav AP, Haussen DC, Bonafe A, Budzik RF, et al. Thrombectomy 6 to 24 Hours after stroke with a Mismatch between deficit and infarct. N Engl J Med 2018;378:11-21.
Nurse Corner Ms Leung Shun Wah Ava
  Ward Manager, KWH A&E
Down-to-earth Experience in the Journey Walking Through the Year of 2020 – Challenges for Nurse Manager during COVID-19 Pandemic in an Accident & Emergency Department in Hong Kong
Since late December 2019, COVID-19, a deadly contagious respiratory disease which was new to all health care professionals,
started spreading rapidly across the world. The impact of this pandemic to the whole city, as well as to our health care system was catastrophic. Nurse managers always shoulder the key responsibilities in maintaining smooth delivery of patient care services. However, in the midst of such uncertain and unprecedented crisis, the real challenge was to ensure that our working team could respond effectively towards the change, and expand care capacity accordingly. What we had learnt about quality management in our school time, looking into the essential quality elements (i.e. man/machine/method/material/environment), had bridged us over the troubled waters.
New information related to infection control, case reporting, working logistics and treatment flooded in from time to time. In order to support our staff (man) under this stressful situation, managers need to disseminate up-to-date information to staff in an organized manner so that key messages could be translated into actions. When new practice was introduced (e.g. enhanced surveillance for patients who need admission…etc.), key actions were illustrated in a 1-page flowchart (material). Repeated briefing sessions (~15 minutes) were arranged for frontline nurses; during which new workflows were introduced by nurse managers so as to achieve standardization of practice, and to minimize misunderstanding (method).
To protect frontline staff from cross infection, manager had reviewed all infection control measures in a proactive way. As the pandemic was so overwhelming, all sorts of personal protective equipment (PPE) became scarce resources. Through timely sharing of the latest infection control messages, early identification and handling of queries and sentiments (man), and securing the accessibility and sustainability of PPE supply in workplace, we could dispel doubts and misunderstanding (material). Designated areas were reserved next to negative pressure rooms to allow convenient DON & DOFF of PPE (environment). Partitions were installed in staff common room to reduce chance of disease transmission (machine). Contact log record was created once the staff was engaged in the care of suspected COVID-19 patient, and this could facilitate easy and accurate contact tracing (method). Trainer team was assembled to expedite efficient and effective staff training on the revised practice, and the application of new devices (e.g. mobile protective screen specifically designed to further reduce the risk of droplet splash in aerosol generating procedures…etc.).
Early recognition and proper segregation of patients who were suspected or confirmed to have SARS-CoV-2 infection could safeguard other A&E patients and the public from cross infection. Triage nurse shouldered a crucial role in patient assessment, as well for risk stratification, at the very beginning of the patient journey. Prompt identification of patients with risk of contracting infection relied on accurate information extraction and analysis from profuse amount of fragmented details (e.g. presenting signs and symptoms, FTOCC, history of visit to specific venue, residential address…etc.). To enable efficient extraction of such crucial information, checklists were developed as a guide for staff to collect information, identify and stratify risk, reaching clinical decision, and deliver appropriate instructions to patients (material).
Limitation in physical space in the A&E department not only restricted the expansion of its service capacity, but also carried increased risks of cross infections. Patients classified in different tiers according to HA’s Enhanced Laboratory Surveillance (ELS) stratification were allocated to designated waiting areas (environment). Although limitation in physical space is a deadlock, especially during the painful stage of hospital major redevelopment, we applied the wisdom of our ancestors: “If Poor, get to Change; With Change, get to Gain” (「窮則變,變則通」). Air purifiers and High-Efficiency Particulate Air (HEPA) Filters were provided by hospital administration. Additional transparent mobile partition screens were acquired to be placed between patients as physical barrier to further reduce the risk of cross infection. These screens facilitated patient monitoring from a distance, despite the physical barrier. Signal lights were installed outside each negative pressure room so that real time monitoring of environmental ventilation status was made easy (machine).
Active management in admission block is a key strategy in maintaining safe and efficient patient care. A web-based dashboard bed management system (machine) was developed for real time monitoring of hospital wide inpatient bed utilization and admission block situation in A&E since 2017. This system was further enhanced to facilitate transparent communication between hospital management, clinical departments and frontline managers, without repetitive workload of counting and reporting. Key features included display of information:
  1. The number, gender, Accident & Emergency Record (AE) number, waiting time for admission, and specialties of queuing patients under admission block;
  2. Bed utilization status of every cubicle within each inpatient ward, showing the availability of the various types of beds at one glance;
  3. Number of overflow inpatient cases amongst different clinical specialties;
  4. Different symbols remarking the specific type of bed that the A&E patients were waiting for (e.g. ordinary patients, patients who need isolation, pending test results of SARS-CoV-2), and the stage of care (e.g. pending or already received specialist consultation…etc.).
Moreover, designated A&E nurse duty in charge in every shift was appointed to closely communicate with surveillance ward for coordination of patient admission (method).
In order to improve case tracking and recording of COVID-19 related workload, a daily record database with automated statistical function was designed (material). Nurse duty in charge only needed to input the patients’ AE number, tier classification, type of SARS-CoV-2 test done and final clinical specialty. Thereafter, a one-page statistics showing a breakdown of all cases will be ready for reporting. Managers could also get a quick idea about the situation and facilitate resources planning.
To conclude our experience in struggling with the COVID-19 Pandemic, some important ideas have proven helpful in leading the nursing team to get over the difficult days:
  1. Sharing of up-to-date information with staff in a timely, “bite-sized” and “easy to digest” way could much improve the staff’s adaptability and compliance to the ever-changing policy and practice;
  2. Maintain sensitivity, and react promptly to staffs’ concern;
  3. Engage and empower frontline supervisors in the inter-departmental coordination of patient care could improve staffs’ sense of control over their work;
  4. Be accessible to deal with staff uncertainties, and exercise decision making timely;
  5. Keep relevant records (e.g. contact history, statistics…etc.) so that tracking of information and decision making could be much easier.
The war against COVID-19 is not easy for everyone. With the dedicated staff of our nursing team and under the down-to-earth leadership of our DOM, we got through the difficult year of 2020. In the new year of 2021, we sincerely hope that this COVID-19 crisis will become history of Mankind soon.

Mobile protection screen for AGP
Partition in staff common room
 Transparent mobile partition screen
Web based dashboard bed management system
Advance in EM  Dr Kwun-bun Wong   
  Associate Consultant, Prince of Wales Hospital
Artificial intelligence in Emergency Medicine ----- Are we prepared?

How could artificial intelligence (AI) benefit Emergency Medicine (EM)?
The tasks of EM physicians are not easy. We have to take care of multiple critical patients, and make rapid access to patient and clinical information for the necessary acute decision making. The unpredictability of patient influx, criticality of the conditions, 24/7 service, resource intense evaluations etc., all lead to serious strain on care providers and resources. These in turn lead to the high-risk of human error and exhaust human capabilities. It is worthwhile to explore how AI can help enhance and complement the Emergency Department (ED) functions and improve EM physicians’ tasks.
Definition of artificial intelligence (AI), machine learning (ML), and deep learning (DL)
The term "artificial intelligence" is often used to describe machines (or computers) that mimic "cognitive" functions that humans associate with the human mind, such as "learning" and "problem solving" (1). Machine learning is a subset of AI about computer algorithms that improve automatically through experience. (2) Deep learning, or deep neural learning, is a subset of machine learning, which uses the neural networks to analyze different factors, with a structure that is similar to the human neural system.
We are in the state of using narrow AI, i.e. trying to achieve specific goals. One of the famous example is AlphaGo, a computer program that plays the board game Go. In future, general AI and superintelligence will emerge when machines become human-like. They make their own decisions and learn without any human input.
AI in Emergency Medicine
Machine learning systems have the potential of prediction and early detection of diseases in ED. Some of the most important algorithms in this field are logistic regression, support vector machines (SVM), Naive Bayes algorithm, decision trees, random forest, gradient boosting and deep learning (4). It was proven to be effective in predicting AKI (5), Influenza (6), UTI (7), Sepsis (8), COPD and asthma (9), and appendicitis (10). AI detected patients requiring urgent coronary revascularization within 48 hours from only 12 leads electrocardiogram (11); and improve detection of large vessel occlusion stroke and rapid triage necessary for expedited treatment (12). By early prediction and diagnosis of high risk diseases, necessary interventions can be performed more rapidly in ED to prevent multiple complications from disease progression.
One of the main applications of AI is triage. An efficient triage can significantly enhance patient flow, reduce length of stay, optimize resource utilization and allocations, and risk stratifications. A retrospective, cross-sectional study of 172,726 ED visits was conducted (13) which showed that E-triage more accurately classified Emergency Severity Index level 3 patients compared with human, and highlighted opportunities to use predictive analytics to support triage decision making.
Another Application of AI in ED is diagnostic imaging. A systematic review and meta-analysis showed that diagnostic performance of deep learning models to be equivalent to that of healthcare professionals. Study showed that software detection of the presence of at least one non-contrast CT feature of acute TBI demonstrated high sensitivity of 98% and high negative predictive value of 99% (15). AI was also shown to be effective in x-ray diagnosis of skeletal pathology (16), and detection of abdominal free fluid in ultrasound (17). Quick and accurate identification of these diagnoses is extremely valuable, both in high volume tertiary centers with potentially lengthy waiting periods for radiology consultations, and in smaller centers with limited staffing of diagnostic radiologists. However, the current AI imaging studies are still focusing on the non-patient-focused radiographic and pathological endpoints. We need further development and translate the results to clinically meaningful endpoints such as survival, symptoms, and need for treatment.
Thirdly, the application of AI can be extended to “before” and “after” ED visits. One model achieved an accuracy of 71.8 %, a sensitivity of 73.8 %, and a specificity of 71.4 % to predict a child’s asthma control level one week ahead (18). 75.8% exacerbations were detected early, with an average of 5 ± 1.9 days in advance for medical attention (19). Hopefully, the use of these applications helps home monitoring after ED visits. At the same time, it may help to reduce the burden of ED visit. In prehospital setting, Danish startup Corti deployed an AI-enabled system that helps to deliver up to 95% accuracy at detecting out of hospital cardiac arrests, reduce error rates of human call-takers and dispatchers, and make critical diagnoses faster to save lives (20).
AI is a newly evolving and disruptive technology; it will definitely change our practice in the future. However, ethical discussions, safety and regulations are barriers hindering the integration of AI into clinical practice. Further external validation and well defined cohorts to augment the quality and interpretability of AI studies are needed.
  1. Russell, Stuart J.; Norvig, Peter (2009). Artificial Intelligence: A Modern Approach (3rd ed.). Upper Saddle River, New Jersey: Prentice Hall. ISBN 978-0-13-604259-4.
  2. Mitchell, Tom (1997). Machine Learning. New York: McGraw Hill. ISBN 0-07-042807-7. OCLC 36417892.
  3. Bengio, Y.; Courville, A.; Vincent, P. (2013). "Representation Learning: A Review and New Perspectives". IEEE Transactions on Pattern Analysis and Machine Intelligence35 (8): 1798–1828.
  4. Negin Shafaf1 and Hamed Malek Applications of Machine Learning Approaches in Emergency Medicine; a Review Article Arch Acad Emerg Med. 2019; 7(1): 34.
  5. Hamid Mohamadlou, Anna Lynn-Palevsky, Christopher Barton et al  Prediction of Acute Kidney Injury with a Machine Learning Algorithm Using Electronic Health Record Data Can J Kidney Health Dis 2018 Jun 8;
  6. Ye Y, Tsui F, Wagner M, Espino JU, Li Q. Influenza detection from emergency department reports using natural language processing and Bayesian network classifiers. Journal of the American Medical Informatics Association. 2014;21(5):815–23
  7.  Taylor RA, Moore CL, Cheung K-H, Brandt C. Predicting urinary tract infections in the emergency department with machine learning. PLoS One. 2018;13(3):e0194085
  8. Mao Q, Jay M, Hoffman JL, Calvert J, Barton C, Shimabukuro D, et al. Multicentre validation of a sepsis prediction algorithm using only vital sign data in the emergency department, general ward and ICU. BMJ open. 2018;8(1):e017833
  9. Goto T, Camargo Jr CA, Faridi MK, Yun BJ, Hasegawa K. Machine learning approaches for predicting disposition of asthma and COPD exacerbations in the ED. The American journal of emergency medicine. 2018;36(9):1650–4
  10. Elikashvili I, Spina L, Ayalin T, Cheng J, Morley EJ, Singh J. An Evidence-based Review of Acute Appendicitis in Childhood. Pediatric Emergency Medicine Practice. 2012;9(3):1–11.
  11. Shinichi Goto, Mai Kimura, Yoshinori Katsumata Artificial intelligence to predict needs for urgent revascularization from 12-leads electrocardiography in emergency patients PLOS One January 9, 2019
  12. Murray NM, Unberath M, Hager GD et al Artificial intelligence to diagnose ischemic stroke and identify large vessel occlusions: a systematic review J Neurointerv Surg 2020 Feb;12(2):156-164
  13. Levin S, Toerper M, Hamrock E, et al. Machine-Learning-Based Electronic Triage More Accurately Differentiates Patients with Respect to Clinical Outcomes Compared with the Emergency Severity Index. Ann Emerg Med. 2018;71(5):565‐574.e2. doi:10.1016/j.annemergmed.2017.08.005
  14. Liu X Faes L Kale AU et al. A comparison of deep learning performance against health-care professionals in detecting diseases from medical imaging: a systematic review and meta-analysis Lancet Digital Health. 2019; 1: e271-e297
  15. Yuh EL, Gean AD, Manley GT, et al. Computer-aided assessment of head computed tomography (CT) studies in patients with suspected traumatic brain injury. J Neurotrauma. 2008;25(10):1163–72.
  16. Jakub Olczak Niklas Fahlberg Atsuto Maki et al. Artificial intelligence for analyzing orthopedic trauma radiographs Acta Orthop. 2017 Dec;88(6):581-586
  17. Sjogren AR, Leo MM, Feldman J, et al. Image segmentation and machine learning for detection of abdominal free fluid in focused assessment with sonography for trauma examinations: A pilot study. J Am Inst Ultrasound Med. 2016;35(11):2501–9
  18. Luo G, Stone BL, Fassl B, et al. Predicting asthma control deterioration in children. BMC Med Inform Decis Mak. 2015;15(1):84
  19. Fernandez-Granero MA, Sanchez-Morillo D, Leon-Jimenez A. Computerized analysis of telemonitored respiratory sounds for predicting acute exacerbations of COPD. Sensors. 2015;15(10):26978–96
  20. Corti – AI that saves lives [Internet]. Corti.  [cited 2018 Oct 15]
Copyright © 2021 non-profit, All rights reserved.

Want to change how you receive these emails?
You can update your preferences or unsubscribe from this list

Email Marketing Powered by Mailchimp