Systematic review and longitudinal analysis of implementing Artificial Intelligence to predict clinical deterioration in adult hospitals: what is known and what remains uncertain.

OBJECTIVE: To identify factors influencing implementation of machine learning algorithms (MLAs) that predict clinical deterioration in hospitalized adult patients and relate these to a validated implementation framework. MATERIALS AND METHODS: A systematic review of studies of implemented or trialed real-time clinical deterioration prediction MLAs was undertaken, which identified: how MLA implementation was measured; impact of MLAs on clinical processes and patient outcomes; and barriers, enablers and uncertainties within the implementation process. Review findings were then mapped to the SALIENT end-to-end implementation framework to identify the implementation stages at which these factors applied. RESULTS: Thirty-seven articles relating to 14 groups of MLAs were identified, each trialing or implementing a bespoke algorithm. One hundred and seven distinct implementation evaluation metrics were identified. Four groups reported decreased hospital mortality, 1 significantly. We identified 24 barriers, 40 enablers, and 14 uncertainties and mapped these to the 5 stages of the SALIENT implementation framework. DISCUSSION: Algorithm performance across implementation stages decreased between in silico and trial stages. Silent plus pilot trial inclusion was associated with decreased mortality, as was the use of logistic regression algorithms that used less than 39 variables. Mitigation of alert fatigue via alert suppression and threshold configuration was commonly employed across groups. CONCLUSIONS: : There is evidence that real-world implementation of clinical deterioration prediction MLAs may improve clinical outcomes. Various factors identified as influencing success or failure of implementation can be mapped to different stages of implementation, thereby providing useful and practical guidance for implementers.

Deployment of machine learning algorithms to predict sepsis: systematic review and application of the SALIENT clinical AI implementation framework.

OBJECTIVE: To retrieve and appraise studies of deployed artificial intelligence (AI)-based sepsis prediction algorithms using systematic methods, identify implementation barriers, enablers, and key decisions and then map these to a novel end-to-end clinical AI implementation framework. MATERIALS AND METHODS: Systematically review studies of clinically applied AI-based sepsis prediction algorithms in regard to methodological quality, deployment and evaluation methods, and outcomes. Identify contextual factors that influence implementation and map these factors to the SALIENT implementation framework. RESULTS: The review identified 30 articles of algorithms applied in adult hospital settings, with 5 studies reporting significantly decreased mortality post-implementation. Eight groups of algorithms were identified, each sharing a common algorithm. We identified 14 barriers, 26 enablers, and 22 decision points which were able to be mapped to the 5 stages of the SALIENT implementation framework. DISCUSSION: Empirical studies of deployed sepsis prediction algorithms demonstrate their potential for improving care and reducing mortality but reveal persisting gaps in existing implementation guidance. In the examined publications, key decision points reflecting real-word implementation experience could be mapped to the SALIENT framework and, as these decision points appear to be AI-task agnostic, this framework may also be applicable to non-sepsis algorithms. The mapping clarified where and when barriers, enablers, and key decisions arise within the end-to-end AI implementation process. CONCLUSIONS: A systematic review of real-world implementation studies of sepsis prediction algorithms was used to validate an end-to-end staged implementation framework that has the ability to account for key factors that warrant attention in ensuring successful deployment, and which extends on previous AI implementation frameworks.

Implementation frameworks for end-to-end clinical AI: derivation of the SALIENT framework.

OBJECTIVE: To derive a comprehensive implementation framework for clinical AI models within hospitals informed by existing AI frameworks and integrated with reporting standards for clinical AI research. MATERIALS AND METHODS: (1) Derive a provisional implementation framework based on the taxonomy of Stead et al and integrated with current reporting standards for AI research: TRIPOD, DECIDE-AI, CONSORT-AI. (2) Undertake a scoping review of published clinical AI implementation frameworks and identify key themes and stages. (3) Perform a gap analysis and refine the framework by incorporating missing items. RESULTS: The provisional AI implementation framework, called SALIENT, was mapped to 5 stages common to both the taxonomy and the reporting standards. A scoping review retrieved 20 studies and 247 themes, stages, and subelements were identified. A gap analysis identified 5 new cross-stage themes and 16 new tasks. The final framework comprised 5 stages, 7 elements, and 4 components, including the AI system, data pipeline, human-computer interface, and clinical workflow. DISCUSSION: This pragmatic framework resolves gaps in existing stage- and theme-based clinical AI implementation guidance by comprehensively addressing the what (components), when (stages), and how (tasks) of AI implementation, as well as the who (organization) and why (policy domains). By integrating research reporting standards into SALIENT, the framework is grounded in rigorous evaluation methodologies. The framework requires validation as being applicable to real-world studies of deployed AI models. CONCLUSIONS: A novel end-to-end framework has been developed for implementing AI within hospital clinical practice that builds on previous AI implementation frameworks and research reporting standards.

Late Holocene environmental change and anthropogenic: Ecosystem interaction on the Laikipia Plateau, Kenya.

East African ecosystems have been shaped by long-term socio-ecological-environmental interactions. Although much previous work on human-environment interrelationships have emphasised the negative impacts of human interventions, a growing body of work shows that there have also often been strong beneficial connections between people and ecosystems, especially in savanna environments. However, limited information and understanding of past interactions between humans and ecosystems of periods longer than a century hampers effective management of contemporary environments. Here, we present a late Holocene study of pollen, fern spore, fungal spore, and charcoal analyses from radiocarbon-dated sediment sequences and assess this record against archaeological and historical data to describe socio-ecological changes on the Laikipia Plateau in Rift Valley Province, Kenya. The results suggest a landscape characterised by closed forests between 2268 years before present (cal year BP) and 1615 cal year BP when there was a significant change to a more open woodland/grassland mosaic that continues to prevail across the study area. Increased amounts of charcoal in the sediment are observed for this same period, becoming particularly common from around 900 cal year BP associated with fungal spores commonly linked to the presence of herbivores. It is likely these trends reflect changes in land use management as pastoral populations improved and extended pasture, using fire to eradicate disease-prone habitats. Implications for contemporary land use management are discussed in the light of these findings.

Integrating stakeholders' perspectives and spatial modelling to develop scenarios of future land use and land cover change in northern Tanzania.

Rapid rates of land use and land cover change (LULCC) in eastern Africa and limited instances of genuinely equal partnerships involving scientists, communities and decision makers challenge the development of robust pathways toward future environmental and socioeconomic sustainability. We use a participatory modelling tool, Kesho, to assess the biophysical, socioeconomic, cultural and governance factors that influenced past (1959-1999) and present (2000-2018) LULCC in northern Tanzania and to simulate four scenarios of land cover change to the year 2030. Simulations of the scenarios used spatial modelling to integrate stakeholders' perceptions of future environmental change with social and environmental data on recent trends in LULCC. From stakeholders' perspectives, between 1959 and 2018, LULCC was influenced by climate variability, availability of natural resources, agriculture expansion, urbanization, tourism growth and legislation governing land access and natural resource management. Among other socio-environmental-political LULCC drivers, the stakeholders envisioned that from 2018 to 2030 LULCC will largely be influenced by land health, natural and economic capital, and political will in implementing land use plans and policies. The projected scenarios suggest that by 2030 agricultural land will have expanded by 8-20% under different scenarios and herbaceous vegetation and forest land cover will be reduced by 2.5-5% and 10-19% respectively. Stakeholder discussions further identified desirable futures in 2030 as those with improved infrastructure, restored degraded landscapes, effective wildlife conservation, and better farming techniques. The undesirable futures in 2030 were those characterized by land degradation, poverty, and cultural loss. Insights from our work identify the implications of future LULCC scenarios on wildlife and cultural conservation and in meeting the Sustainable Development Goals (SDGs) and targets by 2030. The Kesho approach capitalizes on knowledge exchanges among diverse stakeholders, and in the process promotes social learning, provides a sense of ownership of outputs generated, democratizes scientific understanding, and improves the quality and relevance of the outputs.

Anthropological contributions to historical ecology: 50 questions, infinite prospects.

This paper presents the results of a consensus-driven process identifying 50 priority research questions for historical ecology obtained through crowdsourcing, literature reviews, and in-person workshopping. A deliberative approach was designed to maximize discussion and debate with defined outcomes. Two in-person workshops (in Sweden and Canada) over the course of two years and online discussions were peer facilitated to define specific key questions for historical ecology from anthropological and archaeological perspectives. The aim of this research is to showcase the variety of questions that reflect the broad scope for historical-ecological research trajectories across scientific disciplines. Historical ecology encompasses research concerned with decadal, centennial, and millennial human-environmental interactions, and the consequences that those relationships have in the formation of contemporary landscapes. Six interrelated themes arose from our consensus-building workshop model: (1) climate and environmental change and variability; (2) multi-scalar, multi-disciplinary; (3) biodiversity and community ecology; (4) resource and environmental management and governance; (5) methods and applications; and (6) communication and policy. The 50 questions represented by these themes highlight meaningful trends in historical ecology that distill the field down to three explicit findings. First, historical ecology is fundamentally an applied research program. Second, this program seeks to understand long-term human-environment interactions with a focus on avoiding, mitigating, and reversing adverse ecological effects. Third, historical ecology is part of convergent trends toward transdisciplinary research science, which erodes scientific boundaries between the cultural and natural.

Mapping the Elephants of the 19th Century East African Ivory Trade with a Multi-Isotope Approach.

East African elephants have been hunted for their ivory for millennia but the nineteenth century witnessed strongly escalating demand from Europe and North America. It has been suggested that one consequence was that by the 1880s elephant herds along the coast had become scarce, and to meet demand, trade caravans trekked farther into interior regions of East Africa, extending the extraction frontier. The steady decimation of elephant populations coupled with the extension of trade networks have also been claimed to have triggered significant ecological and socio-economic changes that left lasting legacies across the region. To explore the feasibility of using an isotopic approach to uncover a 'moving frontier' of elephant extraction, we constructed a baseline isotope data set (δ13C, δ15N, δ18O and 87Sr/86Sr) for historic East African elephants known to have come from three distinct regions (coastal, Rift Valley, and inland Lakes). Using the isotope results with other climate data and geographical mapping tools, it was possible to characterise elephants from different habitats across the region. This baseline data set was then used to provenance elephant ivory of unknown geographical provenance that was exported from East Africa during the late nineteenth and early twentieth centuries to determine its likely origin. This produced a better understanding of historic elephant geography in the region, and the data have the potential to be used to provenance older archaeological ivories, and to inform contemporary elephant conservation strategies.