1
|
Alami H, Lehoux P, Papoutsi C, Shaw SE, Fleet R, Fortin JP. Understanding the integration of artificial intelligence in healthcare organisations and systems through the NASSS framework: a qualitative study in a leading Canadian academic centre. BMC Health Serv Res 2024; 24:701. [PMID: 38831298 PMCID: PMC11149257 DOI: 10.1186/s12913-024-11112-x] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/03/2023] [Accepted: 05/14/2024] [Indexed: 06/05/2024] Open
Abstract
BACKGROUND Artificial intelligence (AI) technologies are expected to "revolutionise" healthcare. However, despite their promises, their integration within healthcare organisations and systems remains limited. The objective of this study is to explore and understand the systemic challenges and implications of their integration in a leading Canadian academic hospital. METHODS Semi-structured interviews were conducted with 29 stakeholders concerned by the integration of a large set of AI technologies within the organisation (e.g., managers, clinicians, researchers, patients, technology providers). Data were collected and analysed using the Non-Adoption, Abandonment, Scale-up, Spread, Sustainability (NASSS) framework. RESULTS Among enabling factors and conditions, our findings highlight: a supportive organisational culture and leadership leading to a coherent organisational innovation narrative; mutual trust and transparent communication between senior management and frontline teams; the presence of champions, translators, and boundary spanners for AI able to build bridges and trust; and the capacity to attract technical and clinical talents and expertise. Constraints and barriers include: contrasting definitions of the value of AI technologies and ways to measure such value; lack of real-life and context-based evidence; varying patients' digital and health literacy capacities; misalignments between organisational dynamics, clinical and administrative processes, infrastructures, and AI technologies; lack of funding mechanisms covering the implementation, adaptation, and expertise required; challenges arising from practice change, new expertise development, and professional identities; lack of official professional, reimbursement, and insurance guidelines; lack of pre- and post-market approval legal and governance frameworks; diversity of the business and financing models for AI technologies; and misalignments between investors' priorities and the needs and expectations of healthcare organisations and systems. CONCLUSION Thanks to the multidimensional NASSS framework, this study provides original insights and a detailed learning base for analysing AI technologies in healthcare from a thorough socio-technical perspective. Our findings highlight the importance of considering the complexity characterising healthcare organisations and systems in current efforts to introduce AI technologies within clinical routines. This study adds to the existing literature and can inform decision-making towards a judicious, responsible, and sustainable integration of these technologies in healthcare organisations and systems.
Collapse
Affiliation(s)
- Hassane Alami
- Department of Health Management, Evaluation and Policy, School of Public Health, University of Montreal, P.O. Box 6128, Branch Centre-Ville, Montreal, QC, H3C 3J7, Canada.
- Center for Public Health Research of the University of Montreal, Montreal, QC, Canada.
- Institute for Data Valorization (IVADO), Montreal, QC, Canada.
- Nuffield Department of Primary Care Health Sciences, University of Oxford, Oxford, UK.
| | - Pascale Lehoux
- Department of Health Management, Evaluation and Policy, School of Public Health, University of Montreal, P.O. Box 6128, Branch Centre-Ville, Montreal, QC, H3C 3J7, Canada
- Center for Public Health Research of the University of Montreal, Montreal, QC, Canada
| | - Chrysanthi Papoutsi
- Nuffield Department of Primary Care Health Sciences, University of Oxford, Oxford, UK
| | - Sara E Shaw
- Nuffield Department of Primary Care Health Sciences, University of Oxford, Oxford, UK
| | - Richard Fleet
- Faculty of Medicine, Laval University, Quebec, QC, Canada
- VITAM Research Centre on Sustainable Health, Faculty of Medicine, Laval University, Quebec, QC, Canada
| | - Jean-Paul Fortin
- Faculty of Medicine, Laval University, Quebec, QC, Canada
- VITAM Research Centre on Sustainable Health, Faculty of Medicine, Laval University, Quebec, QC, Canada
| |
Collapse
|
2
|
Gillner S. We're implementing AI now, so why not ask us what to do? - How AI providers perceive and navigate the spread of diagnostic AI in complex healthcare systems. Soc Sci Med 2024; 340:116442. [PMID: 38029666 DOI: 10.1016/j.socscimed.2023.116442] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/09/2023] [Revised: 10/14/2023] [Accepted: 11/17/2023] [Indexed: 12/01/2023]
Abstract
Despite high expectations of artificial intelligence (AI) in medical diagnostics, predictions of its extensive and rapid adoption have so far not been matched by reality. AI providers seeking to promote and perpetuate the use of this technology are faced with the complex reality of embedding AI-enabled diagnostics across variable implementation contexts. In this study, we draw upon a complexity science approach and qualitative methodology to understand how AI providers perceive and navigate the spread of AI in complex healthcare systems. Using semi-structured, one-to-one interviews, we collected qualitative data from 14 providers of AI-enabled diagnostics. We triangulated the data by complementing the interviews with multiple sources, including a focus group of physicians with experience using these technologies. The notion of embedding allowed us to connect local implementation efforts with systemic diffusion. Our study reveals that AI providers self-organise to increase their adaptability when navigating the variable conditions and unpredictability of complex healthcare contexts. In addition to the tensions perceived by AI providers within the sociocultural, technological, and institutional subsystems of healthcare, we illustrate the practices emerging among them to mitigate these tensions: stealth science, agility, and digital ambidexterity. Our study contributes to the growing body of literature on the spread of AI in healthcare by capturing the view of technology providers and adding a new theoretical perspective through the lens of complexity science.
Collapse
Affiliation(s)
- Sandra Gillner
- KPM Center for Public Management, University of Bern, Freiburgstr. 3, 3010, Bern, Switzerland; Swiss Institute for Translational and Entrepreneurial Medicine (sitem-insel), Freiburgstr. 3, 3010, Bern, Switzerland.
| |
Collapse
|
3
|
Amann J, Vayena E, Ormond KE, Frey D, Madai VI, Blasimme A. Expectations and attitudes towards medical artificial intelligence: A qualitative study in the field of stroke. PLoS One 2023; 18:e0279088. [PMID: 36630325 PMCID: PMC9833517 DOI: 10.1371/journal.pone.0279088] [Citation(s) in RCA: 5] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/01/2022] [Accepted: 11/01/2022] [Indexed: 01/12/2023] Open
Abstract
INTRODUCTION Artificial intelligence (AI) has the potential to transform clinical decision-making as we know it. Powered by sophisticated machine learning algorithms, clinical decision support systems (CDSS) can generate unprecedented amounts of predictive information about individuals' health. Yet, despite the potential of these systems to promote proactive decision-making and improve health outcomes, their utility and impact remain poorly understood due to their still rare application in clinical practice. Taking the example of AI-powered CDSS in stroke medicine as a case in point, this paper provides a nuanced account of stroke survivors', family members', and healthcare professionals' expectations and attitudes towards medical AI. METHODS We followed a qualitative research design informed by the sociology of expectations, which recognizes the generative role of individuals' expectations in shaping scientific and technological change. Semi-structured interviews were conducted with stroke survivors, family members, and healthcare professionals specialized in stroke based in Germany and Switzerland. Data was analyzed using a combination of inductive and deductive thematic analysis. RESULTS Based on the participants' deliberations, we identified four presumed roles that medical AI could play in stroke medicine, including an administrative, assistive, advisory, and autonomous role AI. While most participants held positive attitudes towards medical AI and its potential to increase accuracy, speed, and efficiency in medical decision making, they also cautioned that it is not a stand-alone solution and may even lead to new problems. Participants particularly emphasized the importance of relational aspects and raised questions regarding the impact of AI on roles and responsibilities and patients' rights to information and decision-making. These findings shed light on the potential impact of medical AI on professional identities, role perceptions, and the doctor-patient relationship. CONCLUSION Our findings highlight the need for a more differentiated approach to identifying and tackling pertinent ethical and legal issues in the context of medical AI. We advocate for stakeholder and public involvement in the development of AI and AI governance to ensure that medical AI offers solutions to the most pressing challenges patients and clinicians face in clinical care.
Collapse
Affiliation(s)
- Julia Amann
- Health Ethics and Policy Lab, Department of Health Sciences and Technology, ETH Zurich, Zurich, Switzerland
| | - Effy Vayena
- Health Ethics and Policy Lab, Department of Health Sciences and Technology, ETH Zurich, Zurich, Switzerland
| | - Kelly E. Ormond
- Health Ethics and Policy Lab, Department of Health Sciences and Technology, ETH Zurich, Zurich, Switzerland
| | - Dietmar Frey
- CLAIM—Charité Lab for AI in Medicine, Charité Universitätsmedizin Berlin, Berlin, Germany
| | - Vince I. Madai
- CLAIM—Charité Lab for AI in Medicine, Charité Universitätsmedizin Berlin, Berlin, Germany
- QUEST Center for Responsible Research, Berlin Institute of Health (BIH), Charité Universitätsmedizin Berlin, Berlin, Germany
- School of Computing and Digital Technology, Faculty of Computing, Engineering and the Built Environment, Birmingham City University, Birmingham, United Kingdom
| | - Alessandro Blasimme
- Health Ethics and Policy Lab, Department of Health Sciences and Technology, ETH Zurich, Zurich, Switzerland
| |
Collapse
|
4
|
Taira RK, Garlid AO, Speier W. Design considerations for a hierarchical semantic compositional framework for medical natural language understanding. PLoS One 2023; 18:e0282882. [PMID: 36928721 PMCID: PMC10019629 DOI: 10.1371/journal.pone.0282882] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/16/2020] [Accepted: 02/24/2023] [Indexed: 03/18/2023] Open
Abstract
Medical natural language processing (NLP) systems are a key enabling technology for transforming Big Data from clinical report repositories to information used to support disease models and validate intervention methods. However, current medical NLP systems fall considerably short when faced with the task of logically interpreting clinical text. In this paper, we describe a framework inspired by mechanisms of human cognition in an attempt to jump the NLP performance curve. The design centers on a hierarchical semantic compositional model (HSCM), which provides an internal substrate for guiding the interpretation process. The paper describes insights from four key cognitive aspects: semantic memory, semantic composition, semantic activation, and hierarchical predictive coding. We discuss the design of a generative semantic model and an associated semantic parser used to transform a free-text sentence into a logical representation of its meaning. The paper discusses supportive and antagonistic arguments for the key features of the architecture as a long-term foundational framework.
Collapse
Affiliation(s)
- Ricky K. Taira
- Medical and Imaging Informatics (MII) Group, Department of Radiological Sciences, University of California, Los Angeles, Los Angeles, California, United States of America
- * E-mail:
| | - Anders O. Garlid
- Medical and Imaging Informatics (MII) Group, Department of Radiological Sciences, University of California, Los Angeles, Los Angeles, California, United States of America
| | - William Speier
- Medical and Imaging Informatics (MII) Group, Department of Radiological Sciences, University of California, Los Angeles, Los Angeles, California, United States of America
- Department of Bioengineering, University of California, Los Angeles, Los Angeles, California, United States of America
| |
Collapse
|
5
|
Létourneau J, Bélanger E, Sia D, Beogo I, Robins S, Kruglova K, Jubinville M, Tchouaket EN. Identifying performance factors of long-term care facilities in the context of the COVID-19 pandemic: a scoping review protocol. Syst Rev 2022; 11:203. [PMID: 36151556 PMCID: PMC9502645 DOI: 10.1186/s13643-022-02069-1] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 09/20/2021] [Accepted: 09/03/2022] [Indexed: 11/10/2022] Open
Abstract
BACKGROUND Long-term care facilities (LTCFs) have been severely affected by the COVID-19 pandemic with serious consequences for the residents. Some LTCFs performed better than others, experiencing lower case and death rates due to COVID-19. A comprehensive understanding of the factors that have affected the transmission of COVID-19 in LTCFs is lacking, as no published studies have applied a multidimensional conceptual framework to evaluate the performance of LTCFs during the pandemic. Much research has focused on infection prevention and control strategies or specific disease outcomes (e.g., death rates). To address these gaps, our scoping review will identify and analyze the performance factors that have influenced the management of COVID-19 in LTCFs by adopting a multidimensional conceptual framework. METHODS We will query the CINAHL, MEDLINE (Ovid), CAIRN, Science Direct, and Web of Science databases for peer-reviewed articles written in English or French and published between January 1, 2020 and December 31, 2021. We will include articles that focus on the specified context (COVID-19), population (LTCFs), interest (facilitators and barriers to performance of LTCFs), and outcomes (dimensions of performance according to a modified version of the Ministère de la santé et des services sociaux du Québec conceptual framework). Each article will be screened by at least two co-authors independently followed by data extraction of the included articles by one co-author and a review by the principal investigator. RESULTS We will present the results both narratively and with visual aids (e.g., flowcharts, tables, conceptual maps). DISCUSSION Our scoping review will provide a comprehensive understanding of the factors that have affected the performance of LTCFs during the COVID-19 pandemic. This knowledge can help inform the development of more effective infection prevention and control measures for future pandemics and outbreaks. The results of our review may lead to improvements in the care and safety of LTCF residents and staff. SCOPING REVIEW REGISTRATION: Research Registry researchregistry7026.
Collapse
Affiliation(s)
- Josiane Létourneau
- Department of Nursing, Université du Québec en Outaouais, St-Jérôme Campus, 5, rue Saint-Joseph, Office J-2204, Saint-Jérôme, Québec, Canada
| | - Emilie Bélanger
- Department of Nursing, Université du Québec en Outaouais, St-Jérôme Campus, 5, rue Saint-Joseph, Office J-2204, Saint-Jérôme, Québec, Canada
| | - Drissa Sia
- Department of Nursing, Université du Québec en Outaouais, St-Jérôme Campus, 5, rue Saint-Joseph, Office J-2204, Saint-Jérôme, Québec, Canada
| | - Idrissa Beogo
- School of Nursing, Faculty of Health Sciences, University of Ottawa, Ottawa, Ontario, Canada
| | - Stephanie Robins
- Department of Nursing, Université du Québec en Outaouais, St-Jérôme Campus, 5, rue Saint-Joseph, Office J-2204, Saint-Jérôme, Québec, Canada
| | - Katya Kruglova
- Department of Nursing, Université du Québec en Outaouais, St-Jérôme Campus, 5, rue Saint-Joseph, Office J-2204, Saint-Jérôme, Québec, Canada
| | - Maripier Jubinville
- Department of Nursing, Université du Québec en Outaouais, St-Jérôme Campus, 5, rue Saint-Joseph, Office J-2204, Saint-Jérôme, Québec, Canada
| | - Eric Nguemeleu Tchouaket
- Department of Nursing, Université du Québec en Outaouais, St-Jérôme Campus, 5, rue Saint-Joseph, Office J-2204, Saint-Jérôme, Québec, Canada.
| |
Collapse
|
6
|
Held LA, Wewetzer L, Steinhäuser J. Determinants of the implementation of an artificial intelligence-supported device for the screening of diabetic retinopathy in primary care - a qualitative study. Health Informatics J 2022; 28:14604582221112816. [PMID: 35921547 DOI: 10.1177/14604582221112816] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/15/2022]
Abstract
Diabetic retinopathy is a microvascular complication of diabetes mellitus that is usually asymptomatic in the early stages. Therefore, its timely detection and treatment are essential. First pilot projects exist to establish a smartphone-based and AI-supported screening of DR in primary care. This study explored health professionals' perceptions of potential barriers and enablers of using a screening such as this in primary care to understand the mechanisms that could influence implementation into routine clinical practice. Semi-structured telephone interviews were conducted and analysed with the help of qualitative analysis of Mayring. The following main influencing factors to implementation have been identified: personal attitude, organisation, time, financial factors, education, support, technical requirement, influence on profession and patient welfare. Most determinants could be relocated in the behaviour change wheel, a validated implementation model. Further research on the patients' perspective and a ranking of the determinants found is needed.
Collapse
Affiliation(s)
- Linda A Held
- Institute of Family Medicine, 54360University Medical Center Schleswig-Holstein, Campus Lübeck, Germany
| | - Larisa Wewetzer
- Institute of Family Medicine, 54360University Medical Center Schleswig-Holstein, Campus Lübeck, Germany
| | - Jost Steinhäuser
- Institute of Family Medicine, 54360University Medical Center Schleswig-Holstein, Campus Lübeck, Germany
| |
Collapse
|
7
|
Saw SN, Ng KH. Current challenges of implementing artificial intelligence in medical imaging. Phys Med 2022; 100:12-17. [PMID: 35714523 DOI: 10.1016/j.ejmp.2022.06.003] [Citation(s) in RCA: 23] [Impact Index Per Article: 11.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 12/06/2021] [Revised: 04/26/2022] [Accepted: 06/11/2022] [Indexed: 12/31/2022] Open
Abstract
The idea of using artificial intelligence (AI) in medical practice has gained vast interest due to its potential to revolutionise healthcare systems. However, only some AI algorithms are utilised due to systems' uncertainties, besides the never-ending list of ethical and legal concerns. This paper intends to provide an overview of current AI challenges in medical imaging with an ultimate aim to foster better and effective communication among various stakeholders to encourage AI technology development. We identify four main challenges in implementing AI in medical imaging, supported with consequences and past events when these problems fail to mitigate. Among them is the creation of a robust AI algorithm that is fair, trustable and transparent. Another issue is on data governance, in which best practices in data sharing must be established to promote trust and protect the patients' privacy. Next, stakeholders, such as the government, technology companies and hospital management, should come to a consensus in creating trustworthy AI policies and regulatory frameworks, which is the fourth challenge, to support, encourage and spur innovation in digital AI healthcare technology. Lastly, we discussed the efforts of various organizations such as the World Health Organisation (WHO), American College of Radiology (ACR), European Society of Radiology (ESR) and Radiological Society of North America (RSNA), who are already actively pursuing ethical developments in AI. The efforts by various stakeholders will eventually overcome hurdles and the deployment of AI-driven healthcare applications in clinical practice will become a reality and hence lead to better healthcare services and outcomes.
Collapse
Affiliation(s)
- Shier Nee Saw
- Department of Artificial Intelligence, Faculty of Computer Science and Information Technology, Universiti Malaya, 50603 Kuala Lumpur, Malaysia.
| | - Kwan Hoong Ng
- Department of Biomedical Imaging, Universiti Malaya, 50603 Kuala Lumpur, Malaysia; Department of Medical Imaging and Radiological Sciences, College of Health Sciences, Kaohsiung Medical University, Kaohsiung, Taiwan
| |
Collapse
|