ChatGPT: A Conceptual Review of Applications and Utility in the Field of Medicine.

Artificial Intelligence, specifically advanced language models such as ChatGPT, have the potential to revolutionize various aspects of healthcare, medical education, and research. In this narrative review, we evaluate the myriad applications of ChatGPT in diverse healthcare domains. We discuss its potential role in clinical decision-making, exploring how it can assist physicians by providing rapid, data-driven insights for diagnosis and treatment. We review the benefits of ChatGPT in personalized patient care, particularly in geriatric care, medication management, weight loss and nutrition, and physical activity guidance. We further delve into its potential to enhance medical research, through the analysis of large datasets, and the development of novel methodologies. In the realm of medical education, we investigate the utility of ChatGPT as an information retrieval tool and personalized learning resource for medical students and professionals. There are numerous promising applications of ChatGPT that will likely induce paradigm shifts in healthcare practice, education, and research. The use of ChatGPT may come with several benefits in areas such as clinical decision making, geriatric care, medication management, weight loss and nutrition, physical fitness, scientific research, and medical education. Nevertheless, it is important to note that issues surrounding ethics, data privacy, transparency, inaccuracy, and inadequacy persist. Prior to widespread use in medicine, it is imperative to objectively evaluate the impact of ChatGPT in a real-world setting using a risk-based approach.

[Interpreting a randomized trial in oncology: Key points for the clinician]. / Interpréter un essai randomisé en oncologie : éléments clés en clinique.

The analysis of randomized clinical trials presents a challenge for clinicians. A set of critical elements can facilitate their interpretation. One must question whether the inclusion and exclusion criteria accurately mirror clinical practice. Does the control arm align with what is currently recognized as best practice? Do patients in the control group have access to the best options when the cancer progresses or recurs? The degree of confidence with which phase II trial results can be interpreted also warrants consideration. Finally, informative censoring can be searched for by comparing early censoring rates between treatment arms. Faced with the challenges of interpreting scientific literature, these keys can help the clinician and guide the eventual integration of new results into shared medical decision-making.

L'analyse d'essais cliniques randomisés est un défi pour le clinicien. Une série d'éléments clés peuvent toutefois aider à l'interprétation. Tout d'abord, les critères d'inclusion et d'exclusion reflètent-ils la pratique quotidienne ? Ensuite, le bras contrôle correspond-il aux meilleures pratiques reconnues ? Est-ce que les patients du groupe contrôle ont un accès aux meilleures options lorsque le cancer progresse ou récidive ? Avec quelle confiance interpréter des résultats de phase II ? Enfin, la censure informative peut être recherchée en comparant les taux de censure précoce entre les bras de traitements. Face aux défis de l'interprétation de la littérature scientifique, ces clés peuvent être une aide pour le clinicien et guider l'intégration éventuelle de nouveaux résultats dans la décision médicale partagée.

Co-creation of a toolkit to assist risk communication and clinical decision-making in severe preeclampsia: SPOT-Impact study design.

Globally, the incidence of hypertensive disorders of pregnancy, especially preeclampsia, remains high, particularly in low- and middle-income countries. The burden of adverse maternal and perinatal outcomes is particularly high for women who develop a hypertensive disorder remote from term (<34 weeks). In parallel, many women have a suboptimal experience of care. To improve the quality of care in terms of provision and experience, there is a need to support the communication of risks and making of treatment decision in ways that promote respectful maternity care. Our study objective is to co-create a tool(kit) to support clinical decision-making, communication of risks and shared decision-making in preeclampsia with relevant stakeholders, incorporating respectful maternity care, justice, and equity principles. This qualitative study detailing the exploratory phase of co-creation takes place over 17 months (Nov 2021-March 2024) in the Greater Accra and Eastern Regions of Ghana. Informed by ethnographic observations of care interactions, in-depth interviews and focus group and group discussions, the tool(kit) will be developed with survivors and women with hypertensive disorders of pregnancy and their families, health professionals, policy makers, and researchers. The tool(kit) will consist of three components: quantitative predicted risk (based on external validated risk models or absolute risk of adverse outcomes), risk communication, and shared decision-making support. We expect to co-create a user-friendly tool(kit) to improve the quality of care for women with preeclampsia remote from term which will contribute to better maternal and perinatal health outcomes as well as better maternity care experience for women in Ghana.

Adverse maternal and perinatal outcomes is high for women who develop preeclampsia remote from term (<34 weeks). To improve the quality of provision and experience of care, there is a need to support communication of risks and treatment decisions that promotes respectful maternity care.This article describes the methodology deployed to cocreate a user-friendly tool(kit) to support risk communication and shared decision-making in the context of severe preeclampsia in a low resource setting.

Enhanced clinical decisions for management of benign prostatic hyperplasia using patient-reported outcomes: protocol for a prospective observational study.

BACKGROUND: Lower urinary tract symptoms (LUTS) due to benign prostatic hyperplasia (BPH) significantly impact quality of life among older men. Despite the prevalent use of the American Urological Association Symptom Index (AUA-SI) for BPH, this measure overlooks key symptoms such as pain and incontinence, underscoring the need for more comprehensive patient-reported outcome (PRO) tools. This study aims to integrate enhanced PROs into routine clinical practice to better capture the spectrum of LUTS, thereby improving clinical outcomes and patient care. METHODS: This prospective observational study will recruit men with LUTS secondary to BPH aged ≥ 50 years from urology clinics. Participants will be stratified into medical and surgical management groups, with PRO assessments scheduled at regular intervals to monitor LUTS and other health outcomes. The study will employ the LURN Symptom Index (SI)-29 alongside the traditional AUA-SI and other non-urologic PROs to evaluate a broad range of symptoms. Data on comorbidities, symptom severity, and treatment efficacy will be collected through a combination of electronic health records and PROs. Analyses will focus on the predictive power of these tools in relation to symptom trajectories and treatment responses. Aims are to: (1) integrate routine clinical tests with PRO assessment to enhance screening, diagnosis, and management of patients with BPH; (2) examine psychometric properties of the LURN SIs, including test-retest reliability and establishment of clinically meaningful differences; and (3) create care-coordination recommendations to facilitate management of persistent symptoms and common comorbidities measured by PROs. DISCUSSION: By employing comprehensive PRO measures, this study expects to refine symptom assessment and enhance treatment monitoring, potentially leading to improved personalized care strategies. The integration of these tools into clinical settings could revolutionize the management of LUTS/BPH by providing more nuanced insights into patient experiences and outcomes. The findings could have significant implications for clinical practices, potentially leading to updates in clinical guidelines and better health management strategies for men with LUTS/BPH. TRIAL REGISTRATION: This study is registered in ClinicalTrials.gov (NCT05898932).

Leveraging graph neural networks for supporting automatic triage of patients.

Patient triage is crucial in emergency departments, ensuring timely and appropriate care based on correctly evaluating the emergency grade of patient conditions. Triage methods are generally performed by human operator based on her own experience and information that are gathered from the patient management process. Thus, it is a process that can generate errors in emergency-level associations. Recently, Traditional triage methods heavily rely on human decisions, which can be subjective and prone to errors. A growing interest has recently been focused on leveraging artificial intelligence (AI) to develop algorithms to maximize information gathering and minimize errors in patient triage processing. We define and implement an AI-based module to manage patients' emergency code assignments in emergency departments. It uses historical data from the emergency department to train the medical decision-making process. Data containing relevant patient information, such as vital signs, symptoms, and medical history, accurately classify patients into triage categories. Experimental results demonstrate that the proposed algorithm achieved high accuracy outperforming traditional triage methods. By using the proposed method, we claim that healthcare professionals can predict severity index to guide patient management processing and resource allocation.

The new paradigm in machine learning - foundation models, large language models and beyond: a primer for physicians.

Foundation machine learning models are deep learning models capable of performing many different tasks using different data modalities such as text, audio, images and video. They represent a major shift from traditional task-specific machine learning prediction models. Large language models (LLM), brought to wide public prominence in the form of ChatGPT, are text-based foundational models that have the potential to transform medicine by enabling automation of a range of tasks, including writing discharge summaries, answering patients questions and assisting in clinical decision-making. However, such models are not without risk and can potentially cause harm if their development, evaluation and use are devoid of proper scrutiny. This narrative review describes the different types of LLM, their emerging applications and potential limitations and bias and likely future translation into clinical practice.

High throughput screening aids clinical decision-making in refractory acute myeloid leukaemia.

BACKGROUND: Despite advances in therapeutics for adverse-risk acute myeloid leukaemia (AML), overall survival remains poor, especially in refractory disease. Comprehensive tumour profiling and pre-clinical drug testing can identify effective personalised therapies. CASE: We describe a case of ETV6-MECOM fusion-positive refractory AML, where molecular analysis and in vitro high throughput drug screening identified a tolerable, novel targeted therapy and provided rationale for avoiding what could have been a toxic treatment regimen. Ruxolitinib combined with hydroxyurea led to disease control and enhanced quality-of-life in a patient unsuitable for intensified chemotherapy or allogeneic stem cell transplantation. CONCLUSION: This case report demonstrates the feasibility and role of combination pre-clinical high throughput screening to aid decision making in high-risk leukaemia. It also demonstrates the role a JAK1/2 inhibitor can have in the palliative setting in select patients with AML.

Clinical decision making for VA ECMO weaning in patients with cardiogenic shock A formative qualitative study.

Weaning and liberation from VA ECMO in cardiogenic shock patients comprises a complex process requiring a continuous trade off between multiple clinical parameters. In the absence of dedicated international guidelines, we hypothesized a great heterogeneity in weaning practices among ECMO centers due to a variety in local preferences, logistics, case load and individual professional experience. This qualitative study focused on the appraisal of clinicians' preferences in decision processes towards liberation from VA ECMO after cardiogenic shock while using focus group interviews in 4 large hospitals. The goal was to provide novel and unique insights in daily clinical weaning practices. As expected, we found we a great heterogeneity of weaning strategies among centers and professionals, although participants appeared to find common ground in a clinically straightforward approach to assess the feasibility of ECMO liberation at the bedside. This was shown in a preference for robust, easily accessible parameters such as arterial pulse pressure, stable cardiac index ≥2.1 L/min, VTI LVOT and 'eyeballing' LVEF.

Collective Intelligence Increases Diagnostic Accuracy in a General Practice Setting.

BACKGROUND: General practitioners (GPs) work in an ill-defined environment where diagnostic errors are prevalent. Previous research indicates that aggregating independent diagnoses can improve diagnostic accuracy in a range of settings. We examined whether aggregating independent diagnoses can also improve diagnostic accuracy for GP decision making. In addition, we investigated the potential benefit of such an approach in combination with a decision support system (DSS). METHODS: We simulated virtual groups using data sets from 2 previously published studies. In study 1, 260 GPs independently diagnosed 9 patient cases in a vignette-based study. In study 2, 30 GPs independently diagnosed 12 patient actors in a patient-facing study. In both data sets, GPs provided diagnoses in a control condition and/or DSS condition(s). Each GP's diagnosis, confidence rating, and years of experience were entered into a computer simulation. Virtual groups of varying sizes (range: 3-9) were created, and different collective intelligence rules (plurality, confidence, and seniority) were applied to determine each group's final diagnosis. Diagnostic accuracy was used as the performance measure. RESULTS: Aggregating independent diagnoses by weighing them equally (i.e., the plurality rule) substantially outperformed average individual accuracy, and this effect increased with increasing group size. Selecting diagnoses based on confidence only led to marginal improvements, while selecting based on seniority reduced accuracy. Combining the plurality rule with a DSS further boosted performance. DISCUSSION: Combining independent diagnoses may substantially improve a GP's diagnostic accuracy and subsequent patient outcomes. This approach did, however, not improve accuracy in all patient cases. Therefore, future work should focus on uncovering the conditions under which collective intelligence is most beneficial in general practice. HIGHLIGHTS: We examined whether aggregating independent diagnoses of GPs can improve diagnostic accuracy.Using data sets of 2 previously published studies, we composed virtual groups of GPs and combined their independent diagnoses using 3 collective intelligence rules (plurality, confidence, and seniority).Aggregating independent diagnoses by weighing them equally substantially outperformed average individual GP accuracy, and this effect increased with increasing group size.Combining independent diagnoses may substantially improve GP's diagnostic accuracy and subsequent patient outcomes.

How artificial intelligence could transform emergency care.

Artificial intelligence (AI) in healthcare is the ability of a computer to perform tasks typically associated with clinical care (e.g. medical decision-making and documentation). AI will soon be integrated into an increasing number of healthcare applications, including elements of emergency department (ED) care. Here, we describe the basics of AI, various categories of its functions (including machine learning and natural language processing) and review emerging and potential future use-cases for emergency care. For example, AI-assisted symptom checkers could help direct patients to the appropriate setting, models could assist in assigning triage levels, and ambient AI systems could document clinical encounters. AI could also help provide focused summaries of charts, summarize encounters for hand-offs, and create discharge instructions with an appropriate language and reading level. Additional use cases include medical decision making for decision rules, real-time models that predict clinical deterioration or sepsis, and efficient extraction of unstructured data for coding, billing, research, and quality initiatives. We discuss the potential transformative benefits of AI, as well as the concerns regarding its use (e.g. privacy, data accuracy, and the potential for changing the doctor-patient relationship).

Effects of explainable artificial intelligence in neurology decision support.

OBJECTIVE: Artificial intelligence (AI)-based decision support systems (DSS) are utilized in medicine but underlying decision-making processes are usually unknown. Explainable AI (xAI) techniques provide insight into DSS, but little is known on how to design xAI for clinicians. Here we investigate the impact of various xAI techniques on a clinician's interaction with an AI-based DSS in decision-making tasks as compared to a general population. METHODS: We conducted a randomized, blinded study in which members of the Child Neurology Society and American Academy of Neurology were compared to a general population. Participants received recommendations from a DSS via a random assignment of an xAI intervention (decision tree, crowd sourced agreement, case-based reasoning, probability scores, counterfactual reasoning, feature importance, templated language, and no explanations). Primary outcomes included test performance and perceived explainability, trust, and social competence of the DSS. Secondary outcomes included compliance, understandability, and agreement per question. RESULTS: We had 81 neurology participants with 284 in the general population. Decision trees were perceived as the more explainable by the medical versus general population (P < 0.01) and as more explainable than probability scores within the medical population (P < 0.001). Increasing neurology experience and perceived explainability degraded performance (P = 0.0214). Performance was not predicted by xAI method but by perceived explainability. INTERPRETATION: xAI methods have different impacts on a medical versus general population; thus, xAI is not uniformly beneficial, and there is no one-size-fits-all approach. Further user-centered xAI research targeting clinicians and to develop personalized DSS for clinicians is needed.

Evaluating the accuracy of ChatGPT-4 in predicting ASA scores: A prospective multicentric study ChatGPT-4 in ASA score prediction.

BACKGROUND: This study investigates the potential of ChatGPT-4, developed by OpenAI, in enhancing medical decision-making processes, particularly in preoperative assessments using the American Society of Anesthesiologists (ASA) scoring system. The ASA score, a critical tool in evaluating patients' health status and anesthesia risks before surgery, categorizes patients from I to VI based on their overall health and risk factors. Despite its widespread use, determining accurate ASA scores remains a subjective process that may benefit from AI-supported assessments. This research aims to evaluate ChatGPT-4's capability to predict ASA scores accurately compared to expert anesthesiologists' assessments. METHODS: In this prospective multicentric study, ethical board approval was obtained, and the study was registered with clinicaltrials.gov (NCT06321445). We included 2851 patients from anesthesiology outpatient clinics, spanning neonates to all age groups and genders, with ASA scores between I-IV. Exclusion criteria were set for ASA V and VI scores, emergency operations, and insufficient information for ASA score determination. Data on patients' demographics, health conditions, and ASA scores by anesthesiologists were collected and anonymized. ChatGPT-4 was then tasked with assigning ASA scores based on the standardized patient data. RESULTS: Our results indicate a high level of concordance between ChatGPT-4 predictions and anesthesiologists' evaluations, with Cohen's kappa analysis showing a kappa value of 0.858 (p = 0.000). While the model demonstrated over 90% accuracy in predicting ASA scores I to III, it showed a notable variance in ASA IV scores, suggesting a potential limitation in assessing patients with more complex health conditions. DISCUSSION: The findings suggest that ChatGPT-4 can significantly contribute to the medical field by supporting anesthesiologists in preoperative assessments. This study not only demonstrates ChatGPT-4's efficacy in medical data analysis and decision-making but also opens new avenues for AI applications in healthcare, particularly in enhancing patient safety and optimizing surgical outcomes. Further research is needed to refine AI models for complex case assessments and integrate them seamlessly into clinical workflows.

Clinical practice, decision-making, and use of clinical decision support systems in invasive mechanical ventilation: a narrative review.

Invasive mechanical ventilation is a key supportive therapy for patients on intensive care. There is increasing emphasis on personalised ventilation strategies. Clinical decision support systems (CDSS) have been developed to support this. We conducted a narrative review to assess evidence that could inform device implementation. A search was conducted in MEDLINE (Ovid) and EMBASE. Twenty-nine studies met the inclusion criteria. Role allocation is well described, with interprofessional collaboration dependent on culture, nurse:patient ratio, the use of protocols, and perception of responsibility. There were no descriptions of process measures, quality metrics, or clinical workflow. Nurse-led weaning is well-described, with factors grouped by patient, nurse, and system. Physician-led weaning is heterogenous, guided by subjective and objective information, and 'gestalt'. No studies explored decision-making with CDSS. Several explored facilitators and barriers to implementation, grouped by clinician (facilitators: confidence using CDSS, retaining decision-making ownership; barriers: undermining clinician's role, ambiguity moving off protocol), intervention (facilitators: user-friendly interface, ease of workflow integration, minimal training requirement; barriers: increased documentation time), and organisation (facilitators: system-level mandate; barriers: poor communication, inconsistent training, lack of technical support). One study described factors that support CDSS implementation. There are gaps in our understanding of ventilation practice. A coordinated approach grounded in implementation science is required to support CDSS implementation. Future research should describe factors that guide clinical decision-making throughout mechanical ventilation, with and without CDSS, map clinical workflow, and devise implementation toolkits. Novel research design analogous to a learning organisation, that considers the commercial aspects of device design, is required.

ChatGPT: Evaluating answers on contrast media related questions and finetuning by providing the model with the ESUR guideline on contrast agents.

OBJECTIVE: This study aimed to assess the feasibility of GPT-4 for answering questions related to contrast media with and without the context of the European Society of Urogenital Radiology (ESUR) guideline on contrast agents. The overarching goal was to determine whether contextual enrichment by providing guideline information improves answers of GPT-4 for clinical decision-making in radiology. METHODS: A set of 64 questions, based on the ESUR guideline on contrast agents mirroring pertinent sections, was developed and posed to GPT-4 both directly and after providing the guideline using a plugin. Responses were graded by experienced radiologists for quality of information and accuracy in pinpointing information from the guideline as well as by radiology residents for utility, using Likert-scales. RESULTS: GPT-4's performance improved significantly with the guideline. Without the guideline, average quality rating was 3.98, which increased to 4.33 with the guideline (p = 0036). In terms of accuracy, 82.3% of answers matched the information from the guideline. Utility scores also reflected a significant improvement with the guideline, with average scores of 4.1 (without) and 4.4 (with) (p = 0.008) with a Fleiss´ Kappa of 0.44. CONCLUSION: GPT-4, when contextually enriched with a guideline, demonstrates enhanced capability in providing guideline-backed recommendations. This approach holds promise for real-time clinical decision-support, making guidelines more actionable. However, further refinements are necessary to maximize the potential of large language models (LLMs). Inherent limitations need to be addressed.

Clinical Guideline for Preimplantation Genetic Testing in Inherited Cardiac Diseases.

BACKGROUND: Preimplantation genetic testing (PGT) is a reproductive technology that selects embryos without (familial) genetic variants. PGT has been applied in inherited cardiac disease and is included in the latest American Heart Association/American College of Cardiology guidelines. However, guidelines selecting eligible couples who will have the strongest risk reduction most from PGT are lacking. We developed an objective decision model to select eligibility for PGT and compared its results with those from a multidisciplinary team. METHODS: All couples with an inherited cardiac disease referred to the national PGT center were included. A multidisciplinary team approved or rejected the indication based on clinical and genetic information. We developed a decision model based on published risk prediction models and literature, to evaluate the severity of the cardiac phenotype and the penetrance of the familial variant in referred patients. The outcomes of the model and the multidisciplinary team were compared in a blinded fashion. RESULTS: Eighty-three couples were referred for PGT (1997-2022), comprising 19 different genes for 8 different inherited cardiac diseases (cardiomyopathies and arrhythmias). Using our model and proposed cutoff values, a definitive decision was reached for 76 (92%) couples, aligning with 95% of the multidisciplinary team decisions. In a prospective cohort of 11 couples, we showed the clinical applicability of the model to select couples most eligible for PGT. CONCLUSIONS: The number of PGT requests for inherited cardiac diseases increases rapidly, without the availability of specific guidelines. We propose a 2-step decision model that helps select couples with the highest risk reduction for cardiac disease in their offspring after PGT.

The process of clinical decision-making in chronic wound care: A scenario-based think-aloud study.

AIMS: To undertake a comprehensive investigation into both the process of information acquisition and the clinical decision-making process utilized by primary care nurses in the course of treating chronic wounds. DESIGN: Scenario-based think-aloud method, enriched by the integration of information processing theory. The study was conducted within the framework of home care nursing organizations situated in Flanders, the Flemish speaking part of Belgium. A cohort of primary care nurses (n = 10), each possessing a minimum of one year of nursing experience, was recruited through the collaboration of three home care nursing organizations. METHODS: Two real-life clinical practice scenarios were employed for the interviews, with the researcher adopting the roles of either the patient or another clinician to enhance the realism of the think-aloud process. Each think-aloud session was promptly succeeded by a subsequent follow-up interview. The Consolidated criteria for Reporting Qualitative research checklist was followed to guarantee a consistent and complete report of the study. RESULTS: Amidst noticeable variations, a discernible pattern surfaced, delineating three sequential concepts: 1. gathering overarching information, 2. collecting and documenting wound-specific data, and 3. interpreting information to formulate wound treatment strategies. These concepts encompassed collaborative discussions with stakeholders, while the refinement of wound treatment strategies was interwoven within both concepts 2 and 3. CONCLUSIONS: Evident variations were identified in chronic wound care clinical decision-making, regardless of educational background or experience. These insights hold the potential to inform the development of clinical decision support systems for chronic wound management and provide guidance to clinicians in their decision-making endeavours.

Impact of intraoperative imaging on decision-making during spine surgery: a survey among spine surgeons using simulated intraoperative images.

PURPOSE: To assess whether the intention to intraoperatively reposition pedicle screws differs when spine surgeons evaluate the same screws with 2D imaging or 3D imaging. METHODS: In this online survey study, 21 spine surgeons evaluated eight pedicle screws from patients who had undergone posterior spinal fixation. In a simulated intraoperative setting, surgeons had to decide if they would reposition a marked pedicle screw based on its position in the provided radiologic imaging. The eight assessed pedicle screws varied in radiologic position, including two screws positioned within the pedicle, two breaching the pedicle cortex < 2 mm, two breaching the pedicle cortex 2-4 mm, and two positioned completely outside the pedicle. Surgeons assessed each pedicle screw twice without knowing and in random order: once with a scrollable three-dimensional (3D) image and once with two oblique fluoroscopic two-dimensional (2D) images. RESULTS: Almost all surgeons (19/21) intended to reposition more pedicle screws based on 3D imaging than on 2D imaging, with a mean number of pedicle screws to be repositioned of, respectively, 4.1 (± 1.3) and 2.0 (± 1.3; p < 0.001). Surgeons intended to reposition two screws placed completely outside the pedicle, one breaching 2-4mm, and one breaching < 2 mm more often based on 3D imaging. CONCLUSION: When provided with 3D imaging, spine surgeons not only intend to intraoperatively reposition pedicle screws at risk of causing postoperative complications more often but also screws with acceptable positions. This study highlights the potential of intraoperative 3D imaging as well as the need for consensus on how to act on intraoperative 3D information.