Frameworks for procurement, integration, monitoring, and evaluation of artificial intelligence tools in clinical settings: A systematic review.

Research on the applications of artificial intelligence (AI) tools in medicine has increased exponentially over the last few years but its implementation in clinical practice has not seen a commensurate increase with a lack of consensus on implementing and maintaining such tools. This systematic review aims to summarize frameworks focusing on procuring, implementing, monitoring, and evaluating AI tools in clinical practice. A comprehensive literature search, following PRSIMA guidelines was performed on MEDLINE, Wiley Cochrane, Scopus, and EBSCO databases, to identify and include articles recommending practices, frameworks or guidelines for AI procurement, integration, monitoring, and evaluation. From the included articles, data regarding study aim, use of a framework, rationale of the framework, details regarding AI implementation involving procurement, integration, monitoring, and evaluation were extracted. The extracted details were then mapped on to the Donabedian Plan, Do, Study, Act cycle domains. The search yielded 17,537 unique articles, out of which 47 were evaluated for inclusion based on their full texts and 25 articles were included in the review. Common themes extracted included transparency, feasibility of operation within existing workflows, integrating into existing workflows, validation of the tool using predefined performance indicators and improving the algorithm and/or adjusting the tool to improve performance. Among the four domains (Plan, Do, Study, Act) the most common domain was Plan (84%, n = 21), followed by Study (60%, n = 15), Do (52%, n = 13), & Act (24%, n = 6). Among 172 authors, only 1 (0.6%) was from a low-income country (LIC) and 2 (1.2%) were from lower-middle-income countries (LMICs). Healthcare professionals cite the implementation of AI tools within clinical settings as challenging owing to low levels of evidence focusing on integration in the Do and Act domains. The current healthcare AI landscape calls for increased data sharing and knowledge translation to facilitate common goals and reap maximum clinical benefit.

AI as a Medical Device for Ophthalmic Imaging in Europe, Australia, and the United States: Protocol for a Systematic Scoping Review of Regulated Devices.

BACKGROUND: Artificial intelligence as a medical device (AIaMD) has the potential to transform many aspects of ophthalmic care, such as improving accuracy and speed of diagnosis, addressing capacity issues in high-volume areas such as screening, and detecting novel biomarkers of systemic disease in the eye (oculomics). In order to ensure that such tools are safe for the target population and achieve their intended purpose, it is important that these AIaMD have adequate clinical evaluation to support any regulatory decision. Currently, the evidential requirements for regulatory approval are less clear for AIaMD compared to more established interventions such as drugs or medical devices. There is therefore value in understanding the level of evidence that underpins AIaMD currently on the market, as a step toward identifying what the best practices might be in this area. In this systematic scoping review, we will focus on AIaMD that contributes to clinical decision-making (relating to screening, diagnosis, prognosis, and treatment) in the context of ophthalmic imaging. OBJECTIVE: This study aims to identify regulator-approved AIaMD for ophthalmic imaging in Europe, Australia, and the United States; report the characteristics of these devices and their regulatory approvals; and report the available evidence underpinning these AIaMD. METHODS: The Food and Drug Administration (United States), the Australian Register of Therapeutic Goods (Australia), the Medicines and Healthcare products Regulatory Agency (United Kingdom), and the European Database on Medical Devices (European Union) regulatory databases will be searched for ophthalmic imaging AIaMD through a snowballing approach. PubMed and clinical trial registries will be systematically searched, and manufacturers will be directly contacted for studies investigating the effectiveness of eligible AIaMD. Preliminary regulatory database searches, evidence searches, screening, data extraction, and methodological quality assessment will be undertaken by 2 independent review authors and arbitrated by a third at each stage of the process. RESULTS: Preliminary searches were conducted in February 2023. Data extraction, data synthesis, and assessment of methodological quality commenced in October 2023. The review is on track to be completed and submitted for peer review by April 2024. CONCLUSIONS: This systematic review will provide greater clarity on ophthalmic imaging AIaMD that have achieved regulatory approval as well as the evidence that underpins them. This should help adopters understand the range of tools available and whether they can be safely incorporated into their clinical workflow, and it should also support developers in navigating regulatory approval more efficiently. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/52602.

Development and validation of a novel stemness-related prognostic model for neuroblastoma using integrated machine learning and bioinformatics analyses.

Background: Neuroblastoma (NB) is a common solid tumor in children, with a dismal prognosis in high-risk cases. Despite advancements in NB treatment, the clinical need for precise prognostic models remains critical, particularly to address the heterogeneity of cancer stemness which plays a pivotal role in tumor aggressiveness and patient outcomes. By utilizing machine learning (ML) techniques, we aimed to explore the cancer stemness features in NB and identify stemness-related hub genes for future investigation and potential targeted therapy. Methods: The public dataset GSE49710 was employed as the training set for acquire gene expression data and NB sample information, including age, stage, and MYCN amplification status and survival. The messenger RNA (mRNA) expression-based stemness index (mRNAsi) was calculated and patients were grouped according to their mRNAsi value. Stemness-related hub genes were identified from the differentially expressed genes (DEGs) to construct a gene signature. This was followed by evaluating the relationship between cancer stemness and the NB immune microenvironment, and the development of a predictive nomogram. We assessed the prognostic outcomes including overall survival (OS) and event-free survival, employing machine learning methods to measure predictive accuracy through concordance indices and validation in an independent cohort E-MTAB-8248. Results: Based on mRNAsi, we categorized NB patients into two groups to explore the association between varying levels of stemness and their clinical outcomes. High mRNAsi was linked to the advanced International Neuroblastoma Staging System (INSS) stage, amplified MYCN, and elder age. High mRNAsi patients had a significantly poorer prognosis than low mRNAsi cases. According to the multivariate Cox analysis, the mRNAsi was an independent risk factor of prognosis in NB patients. After least absolute shrinkage and selection operator (LASSO) regression analysis, four key genes (ERCC6L, DUXAP10, NCAN, DIRAS3) most related to mRNAsi scores were discovered and a risk model was built. Our model demonstrated a significant prognostic capacity with hazard ratios (HR) ranging from 18.96 to 41.20, P values below 0.0001, and area under the receiver operating characteristic curve (AUC) values of 0.918 in the training set, suggesting high predictive accuracy which was further confirmed by external verification. Individuals with a low four-gene signature score had a favorable outcome and better immune responses. Finally, a nomogram for clinical practice was constructed by integrating the four-gene signature and INSS stage. Conclusions: Our findings confirm the influence of CSC features in NB prognosis. The newly developed NB stemness-related four-gene signature prognostic signature could facilitate the prognostic prediction, and the identified hub genes may serve as promising targets for individualized treatments.

Commentary: Patient Perspectives on Artificial Intelligence; What have We Learned and How Should We Move Forward?

Artificial intelligence (AI) in healthcare has now begun to make its contributions to real-world patient care with varying degrees of both public and clinical acceptability around it. The heavy investment from governments, industry and academia needed to reach this point has helped to surface different perspectives on AI. As clinical AI applications become a reality, however, there is an increasing need to harness and integrate patient perspectives, which address the distinct needs of different populations, healthcare systems and clinical problems more closely. Despite this need, patient perspectives on AI implementation have little presence in academic literature and within implementation science and are not sufficiently considered throughout the MedTech and eHealthtech product development cycle, which brings its own challenges and opportunities. This joint patient expert/clinician commentary aims to briefly summarise views on AI. It reflects upon recommendations on how stakeholders such as clinicians and Health & MedTech small and medium-sized enterprises (SMEs) can make practical usage of these views. The recommendations of the authors centre around how to work better with patients to enable both product centric and patient centric innovation and person-centred care.

Safety and efficacy of an artificial intelligence-enabled decision tool for treatment decisions in neovascular age-related macular degeneration and an exploration of clinical pathway integration and implementation: protocol for a multi-methods validation study.

INTRODUCTION: Neovascular age-related macular degeneration (nAMD) management is one of the largest single-disease contributors to hospital outpatient appointments. Partial automation of nAMD treatment decisions could reduce demands on clinician time. Established artificial intelligence (AI)-enabled retinal imaging analysis tools, could be applied to this use-case, but are not yet validated for it. A primary qualitative investigation of stakeholder perceptions of such an AI-enabled decision tool is also absent. This multi-methods study aims to establish the safety and efficacy of an AI-enabled decision tool for nAMD treatment decisions and understand where on the clinical pathway it could sit and what factors are likely to influence its implementation. METHODS AND ANALYSIS: Single-centre retrospective imaging and clinical data will be collected from nAMD clinic visits at a National Health Service (NHS) teaching hospital ophthalmology service, including judgements of nAMD disease stability or activity made in real-world consultant-led-care. Dataset size will be set by a power calculation using the first 127 randomly sampled eligible clinic visits. An AI-enabled retinal segmentation tool and a rule-based decision tree will independently analyse imaging data to report nAMD stability or activity for each of these clinic visits. Independently, an external reading centre will receive both clinical and imaging data to generate an enhanced reference standard for each clinic visit. The non-inferiority of the relative negative predictive value of AI-enabled reports on disease activity relative to consultant-led-care judgements will then be tested. In parallel, approximately 40 semi-structured interviews will be conducted with key nAMD service stakeholders, including patients. Transcripts will be coded using a theoretical framework and thematic analysis will follow. ETHICS AND DISSEMINATION: NHS Research Ethics Committee and UK Health Research Authority approvals are in place (21/NW/0138). Informed consent is planned for interview participants only. Written and oral dissemination is planned to public, clinical, academic and commercial stakeholders.

Stakeholder Perspectives of Clinical Artificial Intelligence Implementation: Systematic Review of Qualitative Evidence.

BACKGROUND: The rhetoric surrounding clinical artificial intelligence (AI) often exaggerates its effect on real-world care. Limited understanding of the factors that influence its implementation can perpetuate this. OBJECTIVE: In this qualitative systematic review, we aimed to identify key stakeholders, consolidate their perspectives on clinical AI implementation, and characterize the evidence gaps that future qualitative research should target. METHODS: Ovid-MEDLINE, EBSCO-CINAHL, ACM Digital Library, Science Citation Index-Web of Science, and Scopus were searched for primary qualitative studies on individuals' perspectives on any application of clinical AI worldwide (January 2014-April 2021). The definition of clinical AI includes both rule-based and machine learning-enabled or non-rule-based decision support tools. The language of the reports was not an exclusion criterion. Two independent reviewers performed title, abstract, and full-text screening with a third arbiter of disagreement. Two reviewers assigned the Joanna Briggs Institute 10-point checklist for qualitative research scores for each study. A single reviewer extracted free-text data relevant to clinical AI implementation, noting the stakeholders contributing to each excerpt. The best-fit framework synthesis used the Nonadoption, Abandonment, Scale-up, Spread, and Sustainability (NASSS) framework. To validate the data and improve accessibility, coauthors representing each emergent stakeholder group codeveloped summaries of the factors most relevant to their respective groups. RESULTS: The initial search yielded 4437 deduplicated articles, with 111 (2.5%) eligible for inclusion (median Joanna Briggs Institute 10-point checklist for qualitative research score, 8/10). Five distinct stakeholder groups emerged from the data: health care professionals (HCPs), patients, carers and other members of the public, developers, health care managers and leaders, and regulators or policy makers, contributing 1204 (70%), 196 (11.4%), 133 (7.7%), 129 (7.5%), and 59 (3.4%) of 1721 eligible excerpts, respectively. All stakeholder groups independently identified a breadth of implementation factors, with each producing data that were mapped between 17 and 24 of the 27 adapted Nonadoption, Abandonment, Scale-up, Spread, and Sustainability subdomains. Most of the factors that stakeholders found influential in the implementation of rule-based clinical AI also applied to non-rule-based clinical AI, with the exception of intellectual property, regulation, and sociocultural attitudes. CONCLUSIONS: Clinical AI implementation is influenced by many interdependent factors, which are in turn influenced by at least 5 distinct stakeholder groups. This implies that effective research and practice of clinical AI implementation should consider multiple stakeholder perspectives. The current underrepresentation of perspectives from stakeholders other than HCPs in the literature may limit the anticipation and management of the factors that influence successful clinical AI implementation. Future research should not only widen the representation of tools and contexts in qualitative research but also specifically investigate the perspectives of all stakeholder HCPs and emerging aspects of non-rule-based clinical AI implementation. TRIAL REGISTRATION: PROSPERO (International Prospective Register of Systematic Reviews) CRD42021256005; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=256005. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): RR2-10.2196/33145.