Embracing Generative Artificial Intelligence as a Support Tool for Clinical Decision-Making.

Background The increasing application of artificial intelligence (AI) in healthcare offers significant opportunities for improving clinical decision-making and medical education. This study assesses the performance of ChatGPT, a large language model (LLM), on the Italian State Exam for Medical Residency (SSM) test, to evaluate its potential as a tool for medical education and clinical decision-making support. Given the high stakes of medical decision-making, understanding the capabilities of AI tools like ChatGPT in this context is crucial. Methods A cross-sectional study design was employed, analyzing 136 questions from the official SSM test. ChatGPT's responses were compared to the performance of medical doctors who took the test in 2022. Questions were classified into clinical cases (CC) and notional questions (NQ). The study assessed ChatGPT's overall accuracy, performance on CC and NQ, and compared its results with participating medical doctors. Results ChatGPT achieved an overall accuracy of 90.44%, with higher performance on clinical cases (92.45%) compared to notional questions (89.15%). ChatGPT's performance was higher than 99.6% of the medical doctor participants. Conclusions Based on the data, ChatGPT shows promise as a useful tool in clinical decision-making, particularly in clinical reasoning. This study recommends further research to explore the potential applications and implementation of LLMs in medical education and medical practice, as well as the initiation of procedures and policies for their safe and effective integration into healthcare systems. Key messages • ChatGPT outperforms 99.6% of medical doctors in the Italian State Exam for Medical Residency, demonstrating potential as a clinical decision-making tool. • Further research is needed to explore LLM implementation in medical education and practice, and to develop policies for safe AI integration.

While very many decision-support (DS) tools (i.e. models and decision support systems (DSS)) have been developed to address forest management problems in Europe, the use of such tools in supporting forest policy processes remains limited. Additionally, while there has been very limited sharing of these tools between European countries, there may be an untapped potential for both users and developers in this area.This paper focuses on improving understanding and capacities in the use of forest DS tools for decision making by identifying major forest policy areas, tools available to support them, compatibility of existing tools with the requirements of forest policy areas, potential areas where tools may be shared between countries and factors limiting the use of DS tools in forest policy.Data collection was based on expert interviews. The questionnaire, which comprised a combination of open- and close-ended questions, was forwarded to experts via email. Expert interviews were completed via Skype with the input of one policy specialist and one modeller/decision support specialist from each country.This study categorised key forest policy areas and the DS tools available to support them. Almost one third of these forest policy areas were not addressed by any DS tool. The analysis also revealed that DS tools are mainly developed to assist scientists and policy decision makers to address smaller spatial scales, that they are more orientated to single decision makers with a predominant focus on market wood products. In addition, through an attribute-matching exercise, the DS tools that could potentially be used in other countries to support similar forest policy areas were also identified.Interviews highlighted some of the reasons why DS tools are seldom used in policy making processes; these include a lack of trust in the actual use of the tools as well as a perception of inadequacy for the specifics of real policy process. This research provides a detailed overview of existing DS tools and the forest policy areas that they address. It further provides information on how to address or reduce the gap between DS tools functionalities and requirements from policy makers.

Clinical Decision Support Tools

Simulation and optimization methods as decision support tools for operation of oil refinery hydrogen networks.

Decision-support tools for detecting physiological deterioration are widely used in clinical medicine but have been criticised for fostering a task-oriented culture and reducing the emphasis on clinical reasoning. Little is understood about what influences clinical decisions aided by decision-support tools, including professional standards, policies, and contextual factors. Therefore, we explored management reasoning employed by anaesthesiologists and PACU nurses in the post-anaesthesia care unit during the discharge of high-risk postoperative patients. A qualitative constructivist study, conducting 18 semi-structured with 6 anaesthesiologists and 12 nurses across three Danish teaching hospitals. We analysed data through thematic analysis, utilising Michael Lipsky's theory of "street-level bureaucracy" in combination with David A. Cook's Management Reasoning Framework as a sensitising concept. Standards are frequently ambiguous, requiring interpretation and prioritisation. This allows for professional discretion by circumventing established policies, reducing task-oriented culture and enhancing the clinical reasoning processes. However, discretion in management reasoning depends on whether the clinician is inclined to uphold or adjust policies to maintain professional standards, influencing discharge decisions. While decision-support tools offer cognitive aid and help standardise patient trajectories, they also limit professional discretion in management reasoning and can potentially compromise care and treatment. This highlights the need for a balanced approach that considers both the benefits and limitations of these tools in clinical decision-making.

An overview of existing decision-making support software and systems for weakly structured subject domains, using expert information has been performed. Existing tools are analyzed and compared with each other from the standpoint of their mathematical ware and functionality, particularly in the area of strategic planning problem solution. It is outlined the key decision-making support tools development trends of the last few decades. On the one hand, the decision-making support system concept is often loosely used, and different kinds of statistical tools, and whole development environments are marketed as decision support software. On the other, many decision support tools are designed for specific problems and subject domains. In spite of complexity of decision problems, manufacturers of decision support software products should keep universality and flexibility requirements in mind. It has been shown that existing automated decision-making support tools are characterized by a set of functional drawbacks and limitations. None of the decision support systems listed in the article allows users to automate the whole strategic planning cycle (from strategic goal formulation to its decomposition into sub-goals and projects to optimal distribution of limited resources among these projects).Based on conducted analysis, a set of relevant requirements to modern decision-making support tools under present-day realities has been suggested. These include universality (irrespectively of the particular subject domain), expert-friendly scale-agnostic interface, opportunity for group expert session organization in remote mode, and others.Based on these requirements, it is obtained specific recommendations for decision-making support means selection in the process of strategic planning (i.e. which systems work better for specific strategic planning-related problems), and for further improvement of decision support tools. Tabl.: 1. Refs: 33 titles.

Purpose: Use of Artificial Intelligence (AI) as a diagnostic tool or clinical decision support tool has been investigated extensively. The purpose of this study is to review the literature to date, the number and type of systems that are already available.Methods: A search of MEDLINE, Embase, Cochrane Review, and Global Health Library (2000–2022) was performed. Key words used were Ophthalmology AND AI OR Clinical Assist OR Decision Support Tools. Inclusion criteria were articles written in the English language and studies with human participants. Exclusion criteria were editorials, articles using animal models, and pure validation studies.Results: 11 557 papers when ‘Ophthalmology and AI’ filtered applied, and 906 when ‘Decision Support Tool’ is specified. 387 titles filtered for inclusion/exclusion criteria, of which 26 working tools were identified, (Cataract 4, Cornea and refraction 3, Melanoma 2, Oculoplastic 3, Paediatric Ophthalmology 1, Triage 2, Glaucoma and Neuro‐Ophthalmology 4, Retina 7). Of these, 6 are available for clinical use.Conclusions: Out of a large number of studies, we find 6 examples of Diagnostic or Decision support tools ready for implementation. Two are diagnostic, based on input of risk factors alone, and the others use image analysis for disease management and screening. Most AI work is retrospective and observational in nature, with most trials in medical retinal diseases. Further work is needed to translate AI ideas into clinical practice.

Invited review: Helping dairy farmers to improve economic performance utilizing data-driving decision support tools

Decision support tools (DSTs), like models, GIS-based planning tools and assessment tools, play an important role in incorporating scientific information into decision-making and facilitating policy implementation. In an interdisciplinary Baltic research group, we compiled 43 DSTs developed to support ecosystem-based management of the Baltic Sea and conducted a thorough review. Analysed DSTs cover a wide variety of policy issues (e.g. eutrophication, biodiversity, human uses) and address environmental as well as socio-economic aspects. In this study, we aim to identify gaps between existing DSTs and end-user needs for DSTs for supporting coastal and marine policy implementation, and to provide recommendations for future DST development. In two online surveys, we assess the awareness and use of DSTs in general, as well as policy implementation challenges and DST needs of representatives of public authorities from all Baltic countries, in particular. Through a policy review we identify major policy issues, policies, and general implementation steps and requirements and develop the synthesis-matrix, which is used to compare DST demand and supply. Our results show that DSTs are predominantly used by researchers. End-users from public authorities use DSTs mostly as background information. Major obstacles for DST use are lacking awareness and experiences. DST demand is strongest for the policy issue eutrophication. Furthermore, DSTs that support the development of plans or programmes of measures and assess their impacts and effectiveness are needed. DST supply is low for recently emerging topics, such as non-indigenous species, marine litter, and underwater noise. To overcome existing obstacles, a common database for DSTs available in the Baltic Sea Region is needed. Furthermore, end-users need guidance and training, and cooperation between DST developers and end-users needs to be enhanced to ensure the practical relevance of DSTs for supporting coastal and marine policy implementation. To fill existing gaps, DSTs that address impacts on human welfare and link environmental and socio-economic aspects should be developed. The BSR serves as a best practice case for studying DSTs and their practical use. Hence, our results can provide insights for DST development in other marine regions. Furthermore, our methodological approach is transferable to other areas.

BackgroundDecisional tools have demonstrated their importance in informing manufacturing and commercial decisions in the monoclonal antibody domain. Recent approved therapies in regenerative medicine have shown great clinical benefits to patients.ObjectiveThe objective of this review was to investigate what decisional tools are available and what issues and gaps have been raised for their use in regenerative medicine.MethodsWe systematically searched MEDLINE to identify articles on decision support tools relevant to tissue engineering, and cell and gene therapy, with the aim of identifying gaps for future decisional tool development. We included published studies in English including a description of decisional tools in regenerative medicines. We extracted data using a predesigned Excel table and assessed the data both quantitatively and qualitatively.ResultsWe identified 9 articles addressing key decisions in manufacturing and product development challenges in cell therapies. The decision objectives, parameters, assumptions, and solution methods were analyzed in detail. We found that all decisional tools focused on cell therapies, and 6 of the 9 reviews focused on allogeneic cell therapy products. We identified no available tools on tissue-engineering and gene therapy products. These studies addressed key decisions in manufacturing and product development challenges in cell therapies, such as choice of technology, through modeling.ConclusionsOur review identified a limited number of decisional tools. While the monoclonal antibodies and biologics decisional tool domain has been well developed and has shown great importance in driving more cost-effective manufacturing processes and better investment decisions, there is a lot to be learned in the regenerative medicine domain. There is ample space for expansion, especially with regard to autologous cell therapies, tissue engineering, and gene therapies. To consider the problem more comprehensively, the full needle-to-needle process should be modeled and evaluated.

The system of stakeholders impacting water heater selection decisions in the U.S. is complex, with numerous actors and key players engaging with varying degrees of information asymmetry. The limited availability of decision-making tools could lead to unintended consequences of water and energy saving decisions on public health, such as the growth of opportunistic pathogens in water systems. We use a qualitative meta-synthesis to identify key stakeholders, map interactions among these stakeholders, identify decisions, roles, and influences, and inventory potential interventions. This study identifies and characterizes the important attributes of the residential water heater stakeholder system (leverage points) that influence the selection of water heating technologies. The ultimate desired outcome of the work is to facilitate the selection of water heating components and design configurations that meet occupant objectives and constraints while limiting the potential for health and safety risks due to scalding or opportunistic pathogens. This effort identifies a clear need for decision-making support tools for selecting residential water heaters, one that takes a whole-systems perspective in the water heater stakeholder system and takes advantage of key leverage points for effective intervention.

The HENVINET Health and Environment Network aimed to enhance the use of scientific knowledge in environmental health for policy making. One of the goals was to identify and evaluate Decision Support Tools (DST) in current use. Special attention was paid to four “priority” health issues: asthma and allergies, cancer, neurodevelopment disorders, and endocrine disruptors.We identified a variety of tools that are used for decision making at various levels and by various stakeholders. We developed a common framework for information acquisition about DSTs, translated this to a database structure and collected the information in an online Metadata Base (MDB).The primary product is an open access web-based MDB currently filled with 67 DSTs, accessible through the HENVINET networking portal http://www.henvinet.eu and http://henvinet.nilu.no. Quality assurance and control of the entries and evaluation of requirements to use the DSTs were also a focus of the work.The HENVINET DST MDB is an open product that enables the public to get basic information about the DSTs, and to search the DSTs using pre-designed attributes or free text. Registered users are able to 1) review and comment on existing DSTs; 2) evaluate each DST’s functionalities, and 3) add new DSTs, or change the entry for their own DSTs.Assessment of the available 67 DSTs showed: 1) more than 25% of the DSTs address only one pollution source; 2) 25% of the DSTs address only one environmental stressor; 3) almost 50% of the DSTs are only applied to one disease; 4) 41% of the DSTs can only be applied to one decision making area; 5) 60% of the DSTs’ results are used only by national authority and/or municipality/urban level administration; 6) almost half of the DSTs are used only by environmental professionals and researchers. This indicates that there is a need to develop DSTs covering an increasing number of pollution sources, environmental stressors and health end points, and considering links to other ‘Driving forces-Pressures-State-Exposure-Effects-Actions’ (DPSEEA) elements. Of interest to both researchers and decision makers should be the standardization of the way DSTs are described for easier access to the knowledge, and the identification of coverage gaps.

Background An interactive decision support tool (DST) was adapted to support patients diagnosed with DCIS who are making treatment decisions. The DST provides the following risk estimates over 10 years: 1) future DCIS or invasive breast cancer in the same breast, 2) the risk of dying from causes other than breast cancer, and 3) the risk of dying from invasive breast cancer. Estimates are personalized based on patient age and DCIS grade (low/intermediate versus high grade or “don’t know”) and were based on a model incorporating data from clinical trials and life tables. Methods The DST was implemented in collaboration with the COMET study on the website DCISoptions.org (www.dcisoptions.org). On the DST, personalized results are displayed separately using a 100-woman icon array and percentages with each outcome (future DCIS/invasive breast cancer, death due to breast cancer, death due to other causes) in a different color for each treatment selected by the patient (lumpectomy only, lumpectomy + radiation therapy, lumpectomy + endocrine therapy, lumpectomy + radiation and endocrine therapy, mastectomy with or without reconstruction, bilateral mastectomy with or without reconstruction). In addition, information regarding active monitoring was provided in descriptive terms without icon array display of personalized outcomes. DST users were defined as those who navigated to the website and entered age and DCIS grade allowing them to access the information about expected outcomes. Users were asked to complete an optional survey both prior to use of the DST and after to assess the impact of the DST on: 1) their awareness of options for DCIS treatment, 2) their willingness to consider these options, 3) their knowledge of mortality risks associated with DCIS, and 4) how helpful the DST was to them (after use only). Results As of June 1, 2020, there were 420 users of the DST (total) with 362 completing the pre-tool survey and 58 of whom completed the post-tool survey. Among all DST users, mean age was 54.0 (9.6 years SD) and DCIS was low/intermediate for 72.0%, high for 18.5% and unknown for the remaining 9.5%. Among users who submitted both the pre- and post-tool survey, median time spent on the tool was 10.4 minutes. Awareness of each treatment option was high and did not change with the tool: 90% among both pre-survey and post-survey users except for bilateral mastectomy which remained at 82.9% among pre-survey and post-survey responders. Among those users who completed the pre- and post-tool surveys, the DST increased the percentage of patients who believed the chance of dying from DCIS is very low from 60.3% at baseline to 74.1% (p<0.0001) and reduced the median estimated numerical risk of dying from DCIS in 10 years from 9.0% at baseline to 3.0% (p<0.0001). A large majority of DST users found the tool very helpful or helpful (79.3% of those who responded) in making a treatment decision for DCIS. Discussion DCIS patients have been shown to greatly overestimate the risks of dying from breast cancer and this has been associated with increased anxiety and potential overtreatment. Our personalized online DST significantly improved knowledge about DCIS risks. Future studies of the DST should assess patient characteristics associated with knowledge gains and whether improved knowledge translates to improved patient outcomes including more patient preference and values-based treatment decisions. Citation Format: Rinaa Sujata Punglia, Ann Partridge, Shelley Hwang, Alastair Thompson, Elizabeth Frank, Donna Pinto, Deborah Collyar, Desiree Basila, Thomas Lynch, Terry Hyslop, Marc Ryser, Elissa Ozanne. Impact of an online ductal carcinoma in situ (DCIS) decision support tool on awareness of treatment options and knowledge of breast cancer risks [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PD5-06.

Understanding how stakeholders manage risks associated with nanomaterials is a key input to the design of strategies and tools to achieve safe and sustainable nanomanufacturing. The paper presents some results of a study aiming firstly to inform the development of a software decision support tool. Further, we seek also to understand existing tools used by stakeholders as a source of capabilities and potential adaptation into decision support framework and tools. Central research questions of this study are: How is collective decision-making on risk management and sustainable nanomaterials organised? Which aspects are taken into account in this collective decision-making? And what role can a decision support tool play in such decision-making? The paper analyses 13 responses to a questionnaire survey held among participants in a meeting in October 2013 and a series of 27 semi-structured telephone interviews conducted from January until April 2014 with decision-makers from mainly European industry and regulators involved in risk management and sustainable manufacturing of nanomaterials. Findings from the study on the social organisation of collective decision-making, aspects taken into account in decisions and potential role of decision support tools are presented.

Learn It–Memorize It! Better Yet–Open Your Smartphone and Use the Information!