Machine learning as a supportive tool to recognize cardiac arrest in emergency calls.

Artificial Intelligence (AI) has the potential to greatly improve the delivery of healthcare and other services that advance population health and wellbeing. However, the use of AI in healthcare also brings potential risks that may cause unintended harm. To guide future developments in AI, the High-Level Expert Group on AI set up by the European Commission (EC), recently published ethics guidelines for what it terms “trustworthy” AI. These guidelines are aimed at a variety of stakeholders, especially guiding practitioners toward more ethical and more robust applications of AI. In line with efforts of the EC, AI ethics scholarship focuses increasingly on converting abstract principles into actionable recommendations. However, the interpretation, relevance, and implementation of trustworthy AI depend on the domain and the context in which the AI system is used. The main contribution of this paper is to demonstrate how to use the general AI HLEG trustworthy AI guidelines in practice in the healthcare domain. To this end, we present a best practice of assessing the use of machine learning as a supportive tool to recognize cardiac arrest in emergency calls. The AI system under assessment is currently in use in the city of Copenhagen in Denmark. The assessment is accomplished by an independent team composed of philosophers, policy makers, social scientists, technical, legal, and medical experts. By leveraging an interdisciplinary team, we aim to expose the complex trade-offs and the necessity for such thorough human review when tackling socio-technical applications of AI in healthcare. For the assessment, we use a process to assess trustworthy AI, called 1 Z-Inspection® to identify specific challenges and potential ethical trade-offs when we consider AI in practice.

[1]  Roberto V. Zicari,et al.  Ethical maintenance of artificial intelligence systems , 2021, Artificial Intelligence for Sustainable Value Creation.

[2]  Jesmin Jahan Tithi,et al.  Z-Inspection®: A Process to Assess Trustworthy AI , 2021, IEEE Transactions on Technology and Society.

[3]  P. Kudenchuk,et al.  Effect of Machine Learning on Dispatcher Recognition of Out-of-Hospital Cardiac Arrest During Calls to Emergency Medical Services , 2021, JAMA network open.

[4]  Kristof Meding,et al.  Ethical considerations and statistical analysis of industry involvement in machine learning research , 2020, AI & SOCIETY.

[5]  Eleanore Hickman,et al.  Trustworthy AI and Corporate Governance: The EU’s Ethics Guidelines for Trustworthy Artificial Intelligence from a Company Law Perspective , 2020, European Business Organization Law Review.

[6]  Margherita Pagani,et al.  Artificial Intelligence for Sustainable Value Creation , 2021 .

[7]  W. Price,et al.  How Much Can Potential Jurors Tell Us About Liability for Medical Artificial Intelligence? , 2021, The Journal of Nuclear Medicine.

[8]  Jeffrey L Pellegrino,et al.  Diagnosis of Out-of-Hospital Cardiac Arrest by Emergency Medical Dispatch: A Diagnostic Systematic Review. , 2020, Resuscitation.

[9]  W. Price,et al.  Invited Perspective: How Much Can Potential Jurors Tell Us about Liability for Medical AI? , 2020, Journal of nuclear medicine : official publication, Society of Nuclear Medicine.

[10]  J. Herlitz,et al.  Shortening Ambulance Response Time Increases Survival in Out‐of‐Hospital Cardiac Arrest , 2020, Journal of the American Heart Association.

[11]  Kellie Owens,et al.  Those designing healthcare algorithms must become actively anti-racist , 2020, Nature Medicine.

[12]  J. Biddle.  On Predicting Recidivism: Epistemic Risk, Tradeoffs, and Values in Machine Learning , 2020, Canadian Journal of Philosophy.

[13]  Ryan Abbott,et al.  The Reasonable Robot: Artificial Intelligence and the Law , 2020 .

[14]  Ryan Abbott.  The Reasonable Robot , 2020 .

[15]  Diego H. Milone,et al.  Gender imbalance in medical imaging datasets produces biased classifiers for computer-aided diagnosis , 2020, Proceedings of the National Academy of Sciences.

[16]  Lars Maaloe,et al.  MultiQT: Multimodal learning for real-time question tracking in speech , 2020, ACL.

[17]  Peter Henderson,et al.  Toward Trustworthy AI Development: Mechanisms for Supporting Verifiable Claims , 2020, ArXiv.

[18]  T. Evgeniou,et al.  The need for a system view to regulate artificial intelligence/machine learning-based software as medical device , 2020, npj Digital Medicine.

[19]  S. Gerke,et al.  Ethical and legal challenges of artificial intelligence-driven healthcare , 2020, Artificial Intelligence in Healthcare.

[20]  David Flynn,et al.  A Safety Framework for Critical Systems Utilising Deep Neural Networks , 2020, SAFECOMP.

[21]  G. Dukic,et al.  European , 2020, Definitions.

[22]  G. Perkins,et al.  Survival after out-of-hospital cardiac arrest in Europe - results of the EuReCa TWO study. , 2020, Resuscitation.

[23]  Francesca N. Delling,et al.  Heart Disease and Stroke Statistics—2020 Update: A Report From the American Heart Association , 2020, Circulation.

[24]  Kristine Bærøe,et al.  How to achieve trustworthy artificial intelligence for health , 2020, Bulletin of the World Health Organization.

[25]  N. Palladino,et al.  The role of the epistemic communities in the 'constitutionalization' of the Internet Governance: the case of the EU High-Level Expert Group on Artificial Intelligence. , 2020 .

[26]  Anastasiya Kiseleva,et al.  AI as a Medical Device: Is it Enough to Ensure Performance Transparency and Accountability? , 2020 .

[27]  Philipp Berens,et al.  On the ethics of algorithmic decision-making in healthcare , 2019, Journal of Medical Ethics.

[28]  W. Price,et al.  Potential Liability for Physicians Using Artificial Intelligence. , 2019, JAMA.

[29]  Brian W. Powers,et al.  Dissecting racial bias in an algorithm used to manage the health of populations , 2019, Science.

[30]  Julia M. Puaschunder.  The Legal and International Situation of AI, Robotics and Big Data With Attention to Healthcare , 2019, SSRN Electronic Journal.

[31]  K. Nugent,et al.  Cardiogenic Pulmonary Edema. , 2019, The American journal of the medical sciences.

[32]  F. C. Müller,et al.  Reply to: "Machine learning as a supportive tool to recognize cardiac arrest in emergency calls". , 2019, Resuscitation.

[33]  Andrew McCallum,et al.  Energy and Policy Considerations for Deep Learning in NLP , 2019, ACL.

[34]  Luciano Floridi,et al.  From What to How: An Initial Review of Publicly Available AI Ethics Tools, Methods and Research to Translate Principles into Practices , 2019, Science and Engineering Ethics.

[35]  I. Kerridge,et al.  Conflicts of interest: new thinking, new processes , 2019, Internal medicine journal.

[36]  Raija Koivisto,et al.  Ethical Framework for Designing Autonomous Intelligent Systems , 2019, Journal of Open Innovation: Technology, Market, and Complexity.

[37]  Ole Winther,et al.  BIVA: A Very Deep Hierarchy of Latent Variables for Generative Modeling , 2019, NeurIPS.

[38]  High-Level Expert Group on Artificial Intelligence – Draft Ethics Guidelines for Trustworthy AI , 2019 .

[39]  David C. Kale,et al.  Do no harm: a roadmap for responsible machine learning for health care , 2019, Nature Medicine.

[40]  P. Pek,et al.  Global resuscitation alliance utstein recommendations for developing emergency care systems to improve cardiac arrest survival. , 2018, Resuscitation.

[41]  Geraint Rees,et al.  Clinically applicable deep learning for diagnosis and referral in retinal disease , 2018, Nature Medicine.

[42]  Seung Chul Lee,et al.  Effect of detection time interval for out-of-hospital cardiac arrest on outcomes in dispatcher-assisted cardiopulmonary resuscitation: A nationwide observational study. , 2018, Resuscitation.

[43]  Peter Sprivulis,et al.  Arti fi cial intelligence and machine learning in emergency medicine , 2022 .

[44]  Quoc V. Le,et al.  Scalable and accurate deep learning with electronic health records , 2018, npj Digital Medicine.

[45]  Timnit Gebru,et al.  Gender Shades: Intersectional Accuracy Disparities in Commercial Gender Classification , 2018, FAT.

[46]  Ian R. Kerr,et al.  Robots and Artificial Intelligence in Health Care , 2017 .

[47]  A. Claesson,et al.  Recognising out-of-hospital cardiac arrest during emergency calls increases bystander cardiopulmonary resuscitation and survival. , 2017, Resuscitation.

[48]  Scott Lundberg,et al.  A Unified Approach to Interpreting Model Predictions , 2017, NIPS.

[49]  L. Køber,et al.  Bystander Efforts and 1‐Year Outcomes in Out‐of‐Hospital Cardiac Arrest , 2017, The New England journal of medicine.

[50]  Marco Sciandrone,et al.  Prediction of chronic damage in systemic lupus erythematosus by using machine-learning models , 2017, PloS one.

[51]  M. Fornage,et al.  Heart Disease and Stroke Statistics—2017 Update: A Report From the American Heart Association , 2017, Circulation.

[52]  Rosangela Barcaro.  Ethical Assessment of Emerging Technologies. Appraising the Moral Plausibility of Technological Visions , 2017 .

[53]  F. Lippert,et al.  Recognition of out-of-hospital cardiac arrest during emergency calls — a systematic review of observational studies , 2017, Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine.

[54]  Subhashini Venugopalan,et al.  Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs. , 2016, JAMA.

[55]  H. Friberg,et al.  Recognition of out-of-hospital cardiac arrest by medical dispatchers in emergency medical dispatch centres in two countries. , 2016, Resuscitation.

[56]  Roger Allan Ford,et al.  Privacy and Accountability in Black-Box Medicine , 2016 .

[57]  Tamara Kredo,et al.  Guide to clinical practice guidelines: the current state of play , 2016, International journal for quality in health care : journal of the International Society for Quality in Health Care.

[58]  Federica Lucivero,et al.  Ethical Assessments of Emerging Technologies , 2016 .

[59]  Mary Ann Peberdy,et al.  Part 8: Post–Cardiac Arrest Care 2015 American Heart Association Guidelines Update for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care , 2015, Circulation.

[60]  Gavin D Perkins,et al.  European Resuscitation Council Guidelines for Resuscitation 2015: Section 1. Executive summary. , 2015, Resuscitation.

[61]  J. Soar,et al.  Adult basic life support and automated external defibrillation , 2015 .

[62]  M. Rosenqvist,et al.  Early cardiopulmonary resuscitation in out-of-hospital cardiac arrest. , 2015, The New England journal of medicine.

[63]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[64]  I. Stiell,et al.  Cardiac arrest diagnostic accuracy of 9-1-1 dispatchers: a prospective multi-center study. , 2015, Resuscitation.

[65]  L. Bossaert,et al.  EuReCa ONE – 27 Nations, ONE Europe, ONE Registry: a prospective observational analysis over one month in 27 resuscitation registries in Europe – the EuReCa ONE study protocol , 2015, Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine.

[66]  Gavin D Perkins,et al.  European Resuscitation Council Guidelines for Resuscitation 2015: Section 11. The ethics of resuscitation and end-of-life decisions. , 2015, Resuscitation.

[67]  L. Wik,et al.  Comparison of Medical Priority Dispatch (MPD) and Criteria Based Dispatch (CBD) relating to cardiac arrest calls. , 2014, Resuscitation.

[68]  Fredrik Folke,et al.  Association of national initiatives to improve cardiac arrest management with rates of bystander intervention and patient survival after out-of-hospital cardiac arrest. , 2013, JAMA.

[69]  E. Christensen,et al.  Implementing a nationwide criteria-based emergency medical dispatch system: A register-based follow-up study , 2013, Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine.

[70]  Matthew Taecker.  Hospitalizations, Costs, and Outcomes of Severe Sepsis in the United States 2003 to 2007 , 2012 .

[71]  T. Lagu,et al.  Hospitalizations, costs, and outcomes of severe sepsis in the United States 2003 to 2007 , 2012, Critical care medicine.

[72]  Y. Tahara,et al.  Complications of bystander cardiopulmonary resuscitation for unconscious patients without cardiopulmonary arrest , 2012, Journal of emergencies, trauma, and shock.

[73]  E. Lerner,et al.  The Frequency and Consequences of Cardiopulmonary Resuscitation Performed by Bystanders on Patients Who Are Not in Cardiac Arrest , 2011, Prehospital emergency care : official journal of the National Association of EMS Physicians and the National Association of State EMS Directors.

[74]  R. Berg,et al.  Global incidences of out-of-hospital cardiac arrest and survival rates: Systematic review of 67 prospective studies. , 2010, Resuscitation.

[75]  Arthur L. Kellermann,et al.  Predictors of Survival From Out-of-Hospital Cardiac Arrest A Systematic Review and Meta-Analysis , 2013 .

[76]  P. Pepe,et al.  Dispatcher assessments for agonal breathing improve detection of cardiac arrest. , 2009, Resuscitation.

[77]  A. Zwinderman,et al.  Importance of the First Link: Description and Recognition of an Out-of-Hospital Cardiac Arrest in an Emergency Call , 2009, Circulation.

[78]  D. Wade,et al.  Cognitive impairments in survivors of out-of-hospital cardiac arrest: a systematic review. , 2009, Resuscitation.

[79]  Robert A. Berg,et al.  Gasping During Cardiac Arrest in Humans Is Frequent and Associated With Improved Survival , 2008, Circulation.

[80]  J. Morrison,et al.  Ethical and legal challenges. , 2008, Fertility and sterility.

[81]  A H Marshall,et al.  The obstacles to maximising the impact of public access defibrillation: an assessment of the dispatch mechanism for out-of-hospital cardiac arrest , 2007, Heart.

[82]  D. Wade,et al.  Life after survival: long-term daily life functioning and quality of life of patients with hypoxic brain injury as a result of a cardiac arrest , 2007, Clinical rehabilitation.

[83]  Shyr-Chyr Chen,et al.  Evaluation of emergency medical dispatch in out-of-hospital cardiac arrest in Taipei. , 2007, Resuscitation.

[84]  V. Pettilä,et al.  Effect of protocol compliance to cardiac arrest identification by emergency medical dispatchers. , 2006, Resuscitation.

[85]  Frank Archer,et al.  Sensitivity and Specificity of the Medical Priority Dispatch System in Detecting Cardiac Arrest Emergency Calls in Melbourne , 2006, Prehospital and Disaster Medicine.

[86]  T. Rea,et al.  Incidence of EMS-treated out-of-hospital cardiac arrest in Europe. , 2005, Resuscitation.

[87]  M. Gratton,et al.  The accuracy of predicting cardiac arrest by emergency medical services dispatchers: the calling party effect. , 2003, Academic emergency medicine : official journal of the Society for Academic Emergency Medicine.

[88]  Michiel Korthals,et al.  Pragmatist Ethics for a Technological Culture , 2002 .

[89]  J. Sexton,et al.  Error, stress, and teamwork in medicine and aviation: cross sectional surveys , 2000, BMJ : British Medical Journal.

[90]  J. Bache.  European Resuscitation Council Guidelines for Resuscitation , 1998 .

[91]  S Tenner,et al.  The influence of the probability of survival on patients' preferences regarding cardiopulmonary resuscitation. , 1994, The New England journal of medicine.

[92]  A. Hallstrom,et al.  Predicting survival from out-of-hospital cardiac arrest: a graphic model. , 1993, Annals of emergency medicine.

[93]  M S Eisenberg,et al.  Incidence of agonal respirations in sudden cardiac arrest. , 1992, Annals of emergency medicine.

[94]  Anthony M. Smith,et al.  Reliability-Centered Maintenance , 1992 .

[95]  J. Ornato,et al.  Improving survival from sudden cardiac arrest: the "chain of survival" concept. A statement for health professionals from the Advanced Cardiac Life Support Subcommittee and the Emergency Cardiac Care Committee, American Heart Association. , 1991, Circulation.

[96]  Lawrence D. Jackel,et al.  Backpropagation Applied to Handwritten Zip Code Recognition , 1989, Neural Computation.

[97]  P. Safar,et al.  Resuscitation from clinical death: Pathophysiologic limits and therapeutic potentials , 1988, Critical care medicine.

[98]  D. Altman,et al.  STATISTICAL METHODS FOR ASSESSING AGREEMENT BETWEEN TWO METHODS OF CLINICAL MEASUREMENT , 1986, The Lancet.

[99]  Josef Kittler,et al.  Pattern recognition : a statistical approach , 1982 .

[100]  T. Cullen,et al.  Global existence of solutions for the relativistic Boltzmann equation on the flat Robertson-Walker space-time for arbitrarily large intial data , 2005, gr-qc/0507035.

[101]  and the survival , 2022 .