Revolutionizing healthcare: the role of artificial intelligence in clinical practice.

Healthcare systems are complex and challenging for all stakeholders, but artificial intelligence (AI) has transformed various fields, including healthcare, with the potential to improve patient care and quality of life. Rapid AI advancements can revolutionize healthcare by integrating it into clinical practice. Reporting AI's role in clinical practice is crucial for successful implementation by equipping healthcare providers with essential knowledge and tools. This review article provides a comprehensive and up-to-date overview of the current state of AI in clinical practice, including its potential applications in disease diagnosis, treatment recommendations, and patient engagement. It also discusses the associated challenges, covering ethical and legal considerations and the need for human expertise. By doing so, it enhances understanding of AI's significance in healthcare and supports healthcare organizations in effectively adopting AI technologies. The current investigation analyzed the use of AI in the healthcare system with a comprehensive review of relevant indexed literature, such as PubMed/Medline, Scopus, and EMBASE, with no time constraints but limited to articles published in English. The focused question explores the impact of applying AI in healthcare settings and the potential outcomes of this application. Integrating AI into healthcare holds excellent potential for improving disease diagnosis, treatment selection, and clinical laboratory testing. AI tools can leverage large datasets and identify patterns to surpass human performance in several healthcare aspects. AI offers increased accuracy, reduced costs, and time savings while minimizing human errors. It can revolutionize personalized medicine, optimize medication dosages, enhance population health management, establish guidelines, provide virtual health assistants, support mental health care, improve patient education, and influence patient-physician trust. AI can be used to diagnose diseases, develop personalized treatment plans, and assist clinicians with decision-making. Rather than simply automating tasks, AI is about developing technologies that can enhance patient care across healthcare settings. However, challenges related to data privacy, bias, and the need for human expertise must be addressed for the responsible and effective implementation of AI in healthcare.

Alowais SA ，Alghamdi SS ，Alsuhebany N ，Alqahtani T ，Alshaya AI ，Almohareb SN ，Aldairem A ，Alrashed M ，Bin Saleh K ，Badreldin HA ，Al Yami MS ，Al Harbi S ，Albekairy AM ... -  《BMC Medical Education》

Generative AI in healthcare: an implementation science informed translational path on application, integration and governance.

Artificial intelligence (AI), particularly generative AI, has emerged as a transformative tool in healthcare, with the potential to revolutionize clinical decision-making and improve health outcomes. Generative AI, capable of generating new data such as text and images, holds promise in enhancing patient care, revolutionizing disease diagnosis and expanding treatment options. However, the utility and impact of generative AI in healthcare remain poorly understood, with concerns around ethical and medico-legal implications, integration into healthcare service delivery and workforce utilisation. Also, there is not a clear pathway to implement and integrate generative AI in healthcare delivery. This article aims to provide a comprehensive overview of the use of generative AI in healthcare, focusing on the utility of the technology in healthcare and its translational application highlighting the need for careful planning, execution and management of expectations in adopting generative AI in clinical medicine. Key considerations include factors such as data privacy, security and the irreplaceable role of clinicians' expertise. Frameworks like the technology acceptance model (TAM) and the Non-Adoption, Abandonment, Scale-up, Spread and Sustainability (NASSS) model are considered to promote responsible integration. These frameworks allow anticipating and proactively addressing barriers to adoption, facilitating stakeholder participation and responsibly transitioning care systems to harness generative AI's potential. Generative AI has the potential to transform healthcare through automated systems, enhanced clinical decision-making and democratization of expertise with diagnostic support tools providing timely, personalized suggestions. Generative AI applications across billing, diagnosis, treatment and research can also make healthcare delivery more efficient, equitable and effective. However, integration of generative AI necessitates meticulous change management and risk mitigation strategies. Technological capabilities alone cannot shift complex care ecosystems overnight; rather, structured adoption programs grounded in implementation science are imperative. It is strongly argued in this article that generative AI can usher in tremendous healthcare progress, if introduced responsibly. Strategic adoption based on implementation science, incremental deployment and balanced messaging around opportunities versus limitations helps promote safe, ethical generative AI integration. Extensive real-world piloting and iteration aligned to clinical priorities should drive development. With conscientious governance centred on human wellbeing over technological novelty, generative AI can enhance accessibility, affordability and quality of care. As these models continue advancing rapidly, ongoing reassessment and transparent communication around their strengths and weaknesses remain vital to restoring trust, realizing positive potential and, most importantly, improving patient outcomes.

Reddy S 《Implementation Science》

Artificial Intelligence, the Digital Surgeon: Unravelling Its Emerging Footprint in Healthcare - The Narrative Review.

Artificial Intelligence (AI) holds transformative potential for the healthcare industry, offering innovative solutions for diagnosis, treatment planning, and improving patient outcomes. As AI continues to be integrated into healthcare systems, it promises advancements across various domains. This review explores the diverse applications of AI in healthcare, along with the challenges and limitations that need to be addressed. The aim is to provide a comprehensive overview of AI's impact on healthcare and to identify areas for further development and focus. The review discusses the broad range of AI applications in healthcare. In medical imaging and diagnostics, AI enhances the accuracy and efficiency of diagnostic processes, aiding in early disease detection. AI-powered clinical decision support systems assist healthcare professionals in patient management and decision-making. Predictive analytics using AI enables the prediction of patient outcomes and identification of potential health risks. AI-driven robotic systems have revolutionized surgical procedures, improving precision and outcomes. Virtual assistants and chatbots enhance patient interaction and support, providing timely information and assistance. In the pharmaceutical industry, AI accelerates drug discovery and development by identifying potential drug candidates and predicting their efficacy. Additionally, AI improves administrative efficiency and operational workflows in healthcare, streamlining processes and reducing costs. AI-powered remote monitoring and telehealth solutions expand access to healthcare, particularly in underserved areas. Despite the significant promise of AI in healthcare, several challenges persist. Ensuring the reliability and consistency of AI-driven outcomes is crucial. Privacy and security concerns must be navigated carefully, particularly in handling sensitive patient data. Ethical considerations, including bias and fairness in AI algorithms, need to be addressed to prevent unintended consequences. Overcoming these challenges is critical for the ethical and successful integration of AI in healthcare. The integration of AI into healthcare is advancing rapidly, offering substantial benefits in improving patient care and operational efficiency. However, addressing the associated challenges is essential to fully realize the transformative potential of AI in healthcare. Future efforts should focus on enhancing the reliability, transparency, and ethical standards of AI technologies to ensure they contribute positively to global health outcomes.

Shang Z ，Chauhan V ，Devi K ，Patil S ... -  《Journal of Multidisciplinary Healthcare》

The future of Cochrane Neonatal.

Cochrane Neonatal was first established in 1993, as one of the original review groups of the Cochrane Collaboration. In fact, the origins of Cochrane Neonatal precede the establishment of the collaboration. In the 1980's, the National Perinatal Epidemiology Unit at Oxford, led by Dr. Iain Chalmers, established the "Oxford Database of Perinatal Trials" (ODPT), a register of virtually all randomized controlled trials in perinatal medicine to provide a resource for reviews of the safety and efficacy of interventions used in perinatal care and to foster cooperative and coordinated research efforts in the perinatal field [1]. An effort that was clearly ahead of its time, ODPT comprised four main elements: a register of published reports of trials; a register of unpublished trials; a register of ongoing and planned trials; and data derived from pooled overviews (meta-analyses) of trials. This core effort grew into the creation of the seminal books, "Effective Care in Pregnancy and Childbirth" as well as "Effective Care of the Newborn Infant" [2,3]. As these efforts in perinatal medicine grew, Iain Chalmers thought well beyond perinatal medicine into the creation of a worldwide collaboration that became Cochrane [4]. The mission of the Cochrane Collaboration is to promote evidence-informed health decision-making by producing high-quality, relevant, accessible systematic reviews and other synthesized research evidence (www.cochrane.org). Cochrane Neonatal has continued to be one of the most productive review groups, publishing between 25 tpo to 40 new or updated systematic reviews each year. The impact factor has been steadily increasing over four years and now rivals most of the elite journals in pediatric medicine. Cochrane Neonatal has been a worldwide effort. Currently, there are 404 reviews involving 1206 authors from 52 countries. What has Cochrane done for babies? Reviews from Cochrane Neonatal have informed guidelines and recommendations worldwide. From January 2018 through June 2020, 77 international guidelines cited 221 Cochrane Neonatal reviews. These recommendations have included recommendations of the use of postnatal steroids, inhaled nitric oxide, feeding guidelines for preterm infants and other core aspects of neonatal practice. In addition, Cochrane Reviews has been the impetus for important research, including the large-scale trial of prophylactic indomethacin therapy, a variety of trials of postnatal steroids, trials of emollient ointment and probiotic trials [6]. While justifiably proud of these accomplishments, one needs to examine the future contribution of Cochrane Neonatal to the neonatal community. The future of Cochrane Neonatal is inexorably linked to the future of neonatal research. Obviously, there is no synthesis of trials data if, as a community, we fail to provide the core substrate for that research. As we look at the current trials' environment, fewer randomized controlled trial related to neonates are being published in recent years. A simple search of PubMed, limiting the search to "neonates" and "randomized controlled trials" shows that in the year 2000, 321 randomized controlled trials were published. These peaked five years ago, in 2015, with close to 900 trials being published. However, in 2018, only 791 studies are identified. Does this decrease represent a meaningful change in the neonatal research environment? Quite possibly. There are shifting missions of clinical neonatology at academic medical institutions, at least in the United States, with a focus on business aspects as well as other important competing clinical activities. Quality improvement has taken over as one of the major activities at both private and academic neonatal practices. Clearly, this is a needed improvement. All units at levels need to be dedicated to improving the outcomes of the sick and fragile population we care for. However, this need not be at the expense of formal clinical trials. It is understandable that this approach would be taken. Newer interventions frequently relate to complex systems of care and not the simple single interventions. Even trials that might traditionally have been done as randomized controlled trials, such as the introduction of a new mode of ventilation, are in reality complex challenges to the ability of institutions to create systems to adapt to these new technologies. Cost of doing trials has always been a barrier. The challenging regulatory and ethical environment contributes to these problems as well [7]. Despite these barriers, how does the research agenda of the neonatal community move forward in the 21st Century? We need to reassess how we create and disseminate our research findings. Innovative trial designs will allow us to address complex issues that we may not have tackled with conventional trials. Adaptive designs may allow us to look at potentially life-saving therapies in a way that feel more efficient and more ethical [8]. Clarifying issues such as the use of inhaled nitric oxide in preterm infants would be greatly served if we even knew whether or not there are hypoxemic preterm infant who would benefit from this therapy [9]. Current trials do not suggest so, yet current practice tells us that a significant number of these babies will receive inhaled nitric oxide [10-13]. Adaptive design, such as those done with trials of extracorporeal membrane oxygenation (ECMO), would allow us to quickly assess whether, in fact, these therapies are life-saving and allow us to consider whether or not further trials are needed [14,15]. Our understanding that many interventions involve entire systems approaches does not relegate us only to doing quality improvement work. Cluster designs may allow us to test more complex interventions that have usually been under the purview of quality improvement [16-18]. Cluster trials are well suited for such investigations and can be done with the least interruption to ongoing care. Ultimately, quality improvement is the application of the best evidence available (evidence-based medicine is "what to do" and evidence-based practice is "how to do"). [19,20]. Nascent efforts, such as the statement on "embedding necessary research into culture and health" (the ENRICH statement) call for the conduct of large, efficient pragmatic trials to evaluate neonatal outcomes, as in part called for in the ALPHA Collaboration [21,22]. This statement envisions an international system to identify important research questions by consulting regularly with all stakeholders, including patients, public health professionals, researchers, providers, policy makers, regulators, funders of industry. The ENRICH statement envisions a pathway to enable individuals, educational institutions, hospitals and health-care facilities to confirm their status as research-friendly by integrating an understanding of trials, other research and critical thinking and to teaching learning and culture, as well as an engagement with funders, professional organizations and regulatory bodies and other stake holders to raise awareness of the value of efficient international research to reduce barriers to large international pragmatic trials and other collaborative studies. In the future, if trials are to be done on this scale or trials are prospectively designed to be analyzed together, core outcome measures must be identified and standardized. That clinical trials supply estimates of outcomes that are relevant to patients and their families is critical. In addition, current neonatal research evaluates many different outcomes using multiple measures. A given measure can have multiple widely used definitions. Bronchopulmonary dysplasia (or chronic lung disease just to add to the confusion) quickly comes to mind [23,24]. The use of multiple definitions when attempting to measure the same outcome prevents synthesis of trial results and meta-analysis and hinders efforts to refine our estimates of effects. Towards that end, Webbe and colleagues have set out to develop a core outcome set for neonatal research [25]. Key stakeholders in the neonatal community reviewed multiple outcomes reported in neonatal trials and qualitative studies. Based on consensus, key outcome measures were identified, including survival, sepsis, necrotizing enterocolitis, brain injury on imaging, retinopathy or prematurity, gross motor ability, general cognitive ability, quality of life, adverse events, visual impairment or blindness, hearing impairment or deafness, chronic lung disease/bronchopulmonary dysplasia. Trials registration has to be a continued focus of the neonatal community. Trials registration allows for systematic reviewers to understand whether or not reporting bias has occurred [26]. It also allows for transparent incorporation of these core outcome measures. Ultimately, trials registration should include public reporting of all of these core outcomes and, in the future, access to data on an individual level such that more sophisticated individual patient data meta-analysis could occur. Lastly, there is no reason to see clinical trials and quality improvement as separate or exclusive activities. In fact, in the first NICQ Collaborative, conducted by Vermont Oxford Network, participation in a trial of postnatal steroids was considered part of the quality improvement best practices as opposed to simply choosing an as-of-yet unproven approach to use of this potent drug [27]. What role will Cochrane Neonatal play as we move forward in the 21st Century? As the neonatal community moves forward with its' research agenda, Cochrane Neonatal must not only follow but also lead with innovative approaches to synthesizing research findings. Cochrane Neonatal must continue to work closely with guideline developers. The relationship between systematic review production and guideline development is clearly outlined in reports from the Institute of Medicine [28,29]. Both are essential to guideline development; the systematic review group culling the evidence for the benefits and harms of a given intervention and the guideline group addressing the contextual issues of cost, feasibility, implementation and the values and preferences of individuals and societies. Most national and international guidelines groups now routinely use systematic reviews as the evidence basis for their guidelines and recommendations. Examples of the partnership between Cochrane Neonatal and international guideline development can be seen in our support of the World Health Organization (WHO) guidelines on the use of vitamin A or the soon to be published recommendations from the International Liaison Committee on Resuscitation (ILCOR) on cord management in preterm and term infants [30]. In the future, we need to collaborate early in the guideline development process so that the reviews are fit for purpose and meet the needs of the guideline developers and the end users. Towards this end, all Cochrane Neonatal reviews now contain GRADE assessments of the key clinical findings reported in the systematic review [31]. Addition of these assessments addresses the critical issue of our confidence in the findings. We are most confident in evidence provided by randomized controlled trials but this assessment can be can be downgraded if the studies that reported on the outcome in question had a high risk of bias, indirectness, inconsistency of results, or imprecision, or where there is evidence of reporting bias. Information provided by GRADE assessments is seen as critical in the process of moving from the evidence to formal recommendations [32]. We need to explore complex reviews, such as network (NMA) or multiple treatment comparison (MCT) meta-analyses, to address issues not formally addressed in clinical trials [33]. In conditions where there are multiple effective interventions, it is rare for all possible interventions to have been tested against each other [34]. A solution could be provided by network meta-analysis, which allows for comparing all treatments with each other, even if randomized controlled trials are not available for some treatment comparisons [34]. Network meta-analysis uses both direct (head-to-head) randomized clinical trial (RCT) evidence as well as indirect evidence from RCTs to compare the relative effectiveness of all included interventions [35]. However, Mills and colleagues note that the methodological quality of MTCs may be difficult for clinicians to interpret because the number of interventions evaluated may be large and the methodological approaches may be complex [35]. Cochrane Neonatal must take a role in both the creation of such analyses and the education of the neonatal community regarding the pitfalls of such an approach. The availability of individual patient data will make more sophisticated analyses more available to the community. Although the current crop of individual patient data meta-analyses (including the reviews of elective high frequency ventilation, inhaled nitric oxide and oxygen targets) have not differed substantially from the findings of the trials level reviews (suggesting that, in fact, sick neonates are more alike that unalike), there still will be a large role for individual patient data meta-analysis, at least to end the unfound conclusions that these therapies are effective in various subgroups (be it issues of sex, disease severity, or clinical setting) [36-39]. Future trials should take a lesson from the NeOProM Collaborative [37,39]. Given the difficulty in generating significant sample size and creating funding in any single environment, trials with similar protocols should be conducted in a variety of healthcare settings with an eye towards both study level and individual patient level meta-analysis at the conclusion of those trials, allowing for broader contribution to the trials data, more rapid accrual of sample size, and more precise results. We need to educate the neonatal community regarding the use and abuse of diagnostic tests. Diagnostic tests are a critical component of healthcare but also contribute greatly to the cost of medical care worldwide. These costs include the cost of the tests themselves and the costs of misdiagnosis and treatment of individuals who will not benefit from those treatments. Clinicians may have a limited understanding of diagnostic test accuracy, the ability of a diagnostic test to distinguish between patients with and without the disease or target condition [41,42]. Efforts such as Choosing Wisely have tried to identify these deficiencies [40]. As Cochrane has increased the general literacy of both the medical and general population regarding the interpretation of the results of interventions on various diseases, so should Cochrane move forward and improve the understanding of diagnostic testing. We need to become more efficient at creating and maintaining our reviews. The time spent to produce systematic reviews is far too great. In average, it takes between 2½ to 6½ years to produce a systematic review, requiring intense time input for highly trained and expensive experts. Innovations in the ways in which we produce systematic reviews can make the review process more efficient by outsourcing some of the tasks or crowdsourcing to machine learning. We need to let the crowd and machine learning innovations help us sort the massive amounts of information needed to conduct systematic reviews. It can also allow for "live" updating of critical reviews where the research landscape is quickly changing [43]. Lastly, Cochrane Neonatal must focus more on users of the reviews and not necessarily authors of the reviews. Current Cochrane programming speaks of Cochrane training with an eye towards developing the skills of individuals who will conduct systematic reviews. While this is clearly needed and laudable, the fact of the matter is that most of the community will be "users" of the reviews. Individuals who need to understand how to use and interpret the findings of systematic reviews. These review users include clinicians, guideline developers, policy makers and families. Incorporation of GRADE guidelines has been a huge step in adding transparency to the level of uncertainty we have in our findings. From a family's perspective, we need to overcome the environment of mistrust or misunderstanding of scientific evidence and how we convey what we know, and our uncertainty about what we know, to parents and families.

Soll RF ，Ovelman C ，McGuire W 《-》

Tribulations and future opportunities for artificial intelligence in precision medicine.

Upon a diagnosis, the clinical team faces two main questions: what treatment, and at what dose? Clinical trials' results provide the basis for guidance and support for official protocols that clinicians use to base their decisions. However, individuals do not consistently demonstrate the reported response from relevant clinical trials. The decision complexity increases with combination treatments where drugs administered together can interact with each other, which is often the case. Additionally, the individual's response to the treatment varies with the changes in their condition. In practice, the drug and the dose selection depend significantly on the medical protocol and the medical team's experience. As such, the results are inherently varied and often suboptimal. Big data and Artificial Intelligence (AI) approaches have emerged as excellent decision-making tools, but multiple challenges limit their application. AI is a rapidly evolving and dynamic field with the potential to revolutionize various aspects of human life. AI has become increasingly crucial in drug discovery and development. AI enhances decision-making across different disciplines, such as medicinal chemistry, molecular and cell biology, pharmacology, pathology, and clinical practice. In addition to these, AI contributes to patient population selection and stratification. The need for AI in healthcare is evident as it aids in enhancing data accuracy and ensuring the quality care necessary for effective patient treatment. AI is pivotal in improving success rates in clinical practice. The increasing significance of AI in drug discovery, development, and clinical trials is underscored by many scientific publications. Despite the numerous advantages of AI, such as enhancing and advancing Precision Medicine (PM) and remote patient monitoring, unlocking its full potential in healthcare requires addressing fundamental concerns. These concerns include data quality, the lack of well-annotated large datasets, data privacy and safety issues, biases in AI algorithms, legal and ethical challenges, and obstacles related to cost and implementation. Nevertheless, integrating AI in clinical medicine will improve diagnostic accuracy and treatment outcomes, contribute to more efficient healthcare delivery, reduce costs, and facilitate better patient experiences, making healthcare more sustainable. This article reviews AI applications in drug development and clinical practice, making healthcare more sustainable, and highlights concerns and limitations in applying AI.

Carini C ，Seyhan AA 《Journal of Translational Medicine》