[Efficacy and safety of artificial intelligence-based large language models for decision making support in herniology: evaluation by experts and general surgeons]. / Effektivnost' i bezopasnost' bol'shikh yazykovykh modelei na osnove iskusstvennogo intellekta v kachestve instrumenta podderzhki prinyatiya reshenii v gerniologii: otsenka ekspertami i obshchimi khirurgami.

OBJECTIVE: To evaluate the quality of recommendations provided by ChatGPT regarding inguinal hernia repair. MATERIAL AND METHODS: ChatGPT was asked 5 questions about surgical management of inguinal hernias. The chat-bot was assigned the role of expert in herniology and requested to search only specialized medical databases and provide information about references and evidence. Herniology experts and surgeons (non-experts) rated the quality of recommendations generated by ChatGPT using 4-point scale (from 0 to 3 points). Statistical correlations were explored between participants' ratings and their stance regarding artificial intelligence. RESULTS: Experts scored the quality of ChatGPT responses lower than non-experts (2 (1-2) vs. 2 (2-3), p<0.001). The chat-bot failed to provide valid references and actual evidence, as well as falsified half of references. Respondents were optimistic about the future of neural networks for clinical decision-making support. Most of them were against restricting their use in healthcare. CONCLUSION: We would not recommend non-specialized large language models as a single or primary source of information for clinical decision making or virtual searching assistant.

Hyperfibrinolysis: potential guidance for decision-making to avoid futile extracorporeal cardiopulmonary resuscitation.

Out-of-hospital cardiac arrest (OHCA) is associated with very poor outcomes. Extracorporeal cardiopulmonary resuscitation (eCPR) for selected patients is a potential therapeutic option for refractory cardiac arrest. However, randomised controlled studies applying eCPR after refractory OHCA have demonstrated conflicting results regarding survival and good functional neurological outcomes. eCPR is an invasive, labour-intensive, and expensive therapeutic approach with associated side-effects. A rapid monitoring device would be valuable in facilitating selection of appropriate patients for this expensive and complex treatment. To this end, rapid diagnosis of hyperfibrinolysis, or premature clot dissolution, diagnosed by viscoelastic testing might represent a feasible option. Hyperfibrinolysis is an evolutionary response to low or no-flow states. Studies in trauma patients demonstrate a high mortality rate in those with established hyperfibrinolysis upon emergency room admission. Similar findings have now been reported for the first time in OHCA patients. Hyperfibrinolysis upon admission diagnosed by rotational thromboelastometry was strongly associated with mortality and poor neurological outcomes in a small cohort of patients treated with extracorporeal membrane oxygenation.

Effectiveness of Artificial Intelligence (AI) in Clinical Decision Support Systems and Care Delivery.

This review aims to assess the effectiveness of AI-driven CDSSs on patient outcomes and clinical practices. A comprehensive search was conducted across PubMed, MEDLINE, and Scopus. Studies published from January 2018 to November 2023 were eligible for inclusion. Following title and abstract screening, full-text articles were assessed for methodological quality and adherence to inclusion criteria. Data extraction focused on study design, AI technologies employed, reported outcomes, and evidence of AI-CDSS impact on patient and clinical outcomes. Thematic analysis was conducted to synthesise findings and identify key themes regarding the effectiveness of AI-CDSS. The screening of the articles resulted in the selection of 26 articles that satisfied the inclusion criteria. The content analysis revealed four themes: early detection and disease diagnosis, enhanced decision-making, medication errors, and clinicians' perspectives. AI-based CDSSs were found to improve clinical decision-making by providing patient-specific information and evidence-based recommendations. Using AI in CDSSs can potentially improve patient outcomes by enhancing diagnostic accuracy, optimising treatment selection, and reducing medical errors.

Incorporating Patient Values in Large Language Model Recommendations for Surrogate and Proxy Decisions.

BACKGROUND: Surrogates, proxies, and clinicians making shared treatment decisions for patients who have lost decision-making capacity often fail to honor patients' wishes, due to stress, time pressures, misunderstanding patient values, and projecting personal biases. Advance directives intend to align care with patient values but are limited by low completion rates and application to only a subset of medical decisions. Here, we investigate the potential of large language models (LLMs) to incorporate patient values in supporting critical care clinical decision-making for incapacitated patients in a proof-of-concept study. METHODS: We simulated text-based scenarios for 50 decisionally incapacitated patients for whom a medical condition required imminent clinical decisions regarding specific interventions. For each patient, we also simulated five unique value profiles captured using alternative formats: numeric ranking questionnaires, text-based questionnaires, and free-text narratives. We used pre-trained generative LLMs for two tasks: 1) text extraction of the treatments under consideration and 2) prompt-based question-answering to generate a recommendation in response to the scenario information, extracted treatment, and patient value profiles. Model outputs were compared with adjudications by three domain experts who independently evaluated each scenario and decision. RESULTS AND CONCLUSIONS: Automated extractions of the treatment in question were accurate for 88% (n = 44/50) of scenarios. LLM treatment recommendations received an average Likert score by the adjudicators of 3.92 of 5.00 (five being best) across all patients for being medically plausible and reasonable treatment recommendations, and 3.58 of 5.00 for reflecting the documented values of the patient. Scores were highest when patient values were captured as short, unstructured, and free-text narratives based on simulated patient profiles. This proof-of-concept study demonstrates the potential for LLMs to function as support tools for surrogates, proxies, and clinicians aiming to honor the wishes and values of decisionally incapacitated patients.

Evaluation of preoperative difficult airway prediction methods for adult patients without obvious airway abnormalities: a systematic review and meta-analysis.

BACKGROUND: This systematic review aims to assist clinical decision-making in selecting appropriate preoperative prediction methods for difficult tracheal intubation by identifying and synthesizing literature on these methods in adult patients undergoing all types of surgery. METHODS: A systematic review and meta-analysis were conducted following PRISMA guidelines. Comprehensive electronic searches across multiple databases were completed on March 28, 2023. Two researchers independently screened, selected studies, and extracted data. A total of 227 articles representing 526 studies were included and evaluated for bias using the QUADAS-2 tool. Meta-Disc software computed pooled sensitivity (SEN), specificity (SPC), positive likelihood ratio (PLR), negative likelihood ratio (NLR), and diagnostic odds ratio (DOR). Heterogeneity was assessed using the Spearman correlation coefficient, Cochran's-Q, and I2 index, with meta-regression exploring sources of heterogeneity. Publication bias was evaluated using Deeks' funnel plot. RESULTS: Out of 2906 articles retrieved, 227 met the inclusion criteria, encompassing a total of 686,089 patients. The review examined 11 methods for predicting difficult tracheal intubation, categorized into physical examination, multivariate scoring system, and imaging test. The modified Mallampati test (MMT) showed a SEN of 0.39 and SPC of 0.86, while the thyromental distance (TMD) had a SEN of 0.38 and SPC of 0.83. The upper lip bite test (ULBT) presented a SEN of 0.52 and SPC of 0.84. Multivariate scoring systems like LEMON and Wilson's risk score demonstrated moderate sensitivity and specificity. Imaging tests, particularly ultrasound-based methods such as the distance from the skin to the epiglottis (US-DSE), exhibited higher sensitivity (0.80) and specificity (0.77). Significant heterogeneity was identified across studies, influenced by factors such as sample size and study design. CONCLUSION: No single preoperative prediction method shows clear superiority for predicting difficult tracheal intubation. The evidence supports a combined approach using multiple methods tailored to specific patient demographics and clinical contexts. Future research should focus on integrating advanced technologies like artificial intelligence and deep learning to improve predictive models. Standardizing testing procedures and establishing clear cut-off values are essential for enhancing prediction reliability and accuracy. Implementing a multi-modal predictive approach may reduce unanticipated difficult intubations, improving patient safety and outcomes.

Assessing Decision Fatigue in General Practitioners' Prescribing Decisions Using the Australian BEACH Data Set.

BACKGROUND: General practitioners (GPs) make numerous care decisions throughout their workdays. Extended periods of decision making can result in decision fatigue, a gradual shift toward decisions that are less cognitively effortful. This study examines whether observed patterns in GPs' prescribing decisions are consistent with the decision fatigue phenomenon. We hypothesized that the likelihood of prescribing frequently overprescribed medications (antibiotics, benzodiazepines, opioids; less effortful to prescribe) will increase and the likelihood of prescribing frequently underprescribed medications (statins, osteoporosis medications; more effortful to prescribe) will decrease over the workday. METHODS: This study used nationally representative primary care data on GP-patient encounters from the Bettering the Evaluation and Care of Health program from Australia. The association between prescribing decisions and order of patient encounters over a GP's workday was assessed with generalized linear mixed models accounting for clustering and adjusting for patient, provider, and encounter characteristics. RESULTS: Among 262,456 encounters recorded by 2,909 GPs, the odds of prescribing antibiotics significantly increased by 8.7% with 15 additional patient encounters (odds ratio [OR] = 1.087; confidence interval [CI] = 1.059-1.116). The odds of prescribing decreased significantly with 15 additional patient encounters by 6.3% for benzodiazepines (OR = 0.937; CI = 0.893-0.983), 21.9% for statins (OR = 0.791; CI = 0.753-0.831), and 25.0% for osteoporosis medications (OR = 0.750; CI = 0.690-0.814). No significant effects were observed for opioids. All findings were replicated in confirmatory analyses except the effect of benzodiazepines. CONCLUSIONS: GPs were increasingly likely to prescribe antibiotics and were less likely to prescribe statins and osteoporosis medications as the workday wore on, which was consistent with decision fatigue. There was no convincing evidence of decision fatigue effects in the prescribing of opioids or benzodiazepines. These findings establish decision fatigue as a promising target for optimizing prescribing behavior. HIGHLIGHTS: We found that as general practitioners progress through their workday, they become more likely to prescribe antibiotics that are reportedly overprescribed and less likely to prescribe statins and osteoporosis medications that are reportedly underprescribed.This change in decision making over time is consistent with the decision fatigue phenomenon. Decision fatigue occurs when we make many decisions without taking a rest break. As we make those decisions, we become gradually more likely to make decisions that are less difficult.The findings of this study show that decision fatigue is a possible target for improving guideline-compliant prescribing of pharmacologic medications.

Examining the role of ChatGPT in the management of distal radius fractures: insights into its accuracy and consistency.

BACKGROUND: The optimal management of distal radius fractures remains a challenge for orthopaedic surgeons. The emergence of Artificial Intelligence (AI) and Large Language Models (LLMs), especially ChatGPT, affords significant potential in improving healthcare and research. This study aims to assess the accuracy and consistency of ChatGPT's knowledge in managing distal radius fractures, with a focus on its capability to provide information for patients and assist in the decision-making processes of orthopaedic clinicians. METHODS: We presented ChatGPT with seven questions on distal radius fracture management over two sessions, resulting in 14 responses. These questions covered a range of topics, including patient inquiries and orthopaedic clinical decision-making. We requested references for each response and involved two orthopaedic registrars and two senior orthopaedic surgeons to evaluate response accuracy and consistency. RESULTS: All 14 responses contained a mix of both correct and incorrect information. Among the 47 cited references, 13% were accurate, 28% appeared to be fabricated, 57% were incorrect, and 2% were correct but deemed inappropriate. Consistency was observed in 71% of the responses. CONCLUSION: ChatGPT demonstrates significant limitations in accuracy and consistency when providing information on distal radius fractures. In its current format, it offers limited utility for patient education and clinical decision-making.

Medical artificial intelligence for clinicians: the lost cognitive perspective.

The development and commercialisation of medical decision systems based on artificial intelligence (AI) far outpaces our understanding of their value for clinicians. Although applicable across many forms of medicine, we focus on characterising the diagnostic decisions of radiologists through the concept of ecologically bounded reasoning, review the differences between clinician decision making and medical AI model decision making, and reveal how these differences pose fundamental challenges for integrating AI into radiology. We argue that clinicians are contextually motivated, mentally resourceful decision makers, whereas AI models are contextually stripped, correlational decision makers, and discuss misconceptions about clinician-AI interaction stemming from this misalignment of capabilities. We outline how future research on clinician-AI interaction could better address the cognitive considerations of decision making and be used to enhance the safety and usability of AI models in high-risk medical decision-making contexts.

The hidden value of MRI: modifying treatment decisions in C-spine injuries.

BACKGROUND DATA: Computed Tomography (CT) is the gold standard for cervical spine (c-spine) evaluation. Magnetic resonance imaging (MRI) emerges due to its increasing availability and the lack of radiation exposure. However, MRI is costly and time-consuming, questioning its role in the emergency department (ED). This study investigates the added the value of an additional MRI for patients presenting with a c-spine injury in the ED. METHODS: We conducted a retrospective monocenter cohort study that included all patients with neck trauma presenting in the ED, who received imaging based on the NEXUS criteria. Spine surgeons performed a full-case review to classify each case into "c-spine injured" and "c-spine uninjured". Injuries were classified according to the AO Spine classification. We assessed patients with a c-spine injury detected by CT, who received a subsequent MRI. In this subset, injuries were classified separately in both imaging modalities. We monitored the treatment changes after the additional MRI to evaluate characteristics of this cohort and the impact of the AO Spine Neurology/Modifier modifiers. RESULTS: We identified 4496 subjects, 2321 were eligible for inclusion and 186 were diagnosed with c-spine injuries in the retrospective case review. Fifty-six patients with a c-spine injury initially identified through CT received an additional MRI. The additional MRI significantly extended (geometric mean ratio 1.32, p < 0.001) the duration of the patients' stay in the ED. Of this cohort, 25% had a change in treatment strategy and among the patients with neurological symptoms (AON ≥ 1), 45.8% experienced a change in treatment. Patients that were N-positive, had a 12.4 (95% CI 2.7-90.7, p < 0.01) times higher odds of a treatment change after an additional MRI than neurologically intact patients. CONCLUSION AND RELEVANCE: Our study suggests that patients with a c-spine injury and neurological symptoms benefit from an additional MRI. In neurologically intact patients, an additional MRI retains value only when carefully evaluated on a case-by-case basis.

Artificial intelligence in total and unicompartmental knee arthroplasty.

The application of Artificial intelligence (AI) and machine learning (ML) tools in total (TKA) and unicompartmental knee arthroplasty (UKA) emerges with the potential to improve patient-centered decision-making and outcome prediction in orthopedics, as ML algorithms can generate patient-specific risk models. This review aims to evaluate the potential of the application of AI/ML models in the prediction of TKA outcomes and the identification of populations at risk.An extensive search in the following databases: MEDLINE, Scopus, Cinahl, Google Scholar, and EMBASE was conducted using the PIOS approach to formulate the research question. The PRISMA guideline was used for reporting the evidence of the data extracted. A modified eight-item MINORS checklist was employed for the quality assessment. The databases were screened from the inception to June 2022.Forty-four out of the 542 initially selected articles were eligible for the data analysis; 5 further articles were identified and added to the review from the PUBMED database, for a total of 49 articles included. A total of 2,595,780 patients were identified, with an overall average age of the patients of 70.2 years ± 7.9 years old. The five most common AI/ML models identified in the selected articles were: RF, in 38.77% of studies; GBM, in 36.73% of studies; ANN in 34.7% of articles; LR, in 32.65%; SVM in 26.53% of articles.This systematic review evaluated the possible uses of AI/ML models in TKA, highlighting their potential to lead to more accurate predictions, less time-consuming data processing, and improved decision-making, all while minimizing user input bias to provide risk-based patient-specific care.

Evaluation of the prediagnosis and management of ChatGPT-4.0 in clinical cases in cardiology.

Aim: Evaluation of the performance of ChatGPT-4.0 in providing prediagnosis and treatment plans for cardiac clinical cases by expert cardiologists. Methods: 20 cardiology clinical cases developed by experienced cardiologists were divided into two groups according to preparation methods. Cases were reviewed and analyzed by the ChatGPT-4.0 program, and analyses of ChatGPT were then sent to cardiologists. Eighteen expert cardiologists evaluated the quality of ChatGPT-4.0 responses using Likert and Global quality scales. Results: Physicians rated case difficulty (median 2.00), revealing high ChatGPT-4.0 agreement to differential diagnoses (median 5.00). Management plans received a median score of 4, indicating good quality. Regardless of the difficulty of the cases, ChatGPT-4.0 showed similar performance in differential diagnosis (p: 0.256) and treatment plans (p: 0.951). Conclusion: ChatGPT-4.0 excels at delivering accurate management and demonstrates its potential as a valuable clinical decision support tool in cardiology.

Have you ever wondered if an artificial intelligence (AI) program could help doctors figure out what the problem is when someone has heart complaints? Our research examined this by testing an AI program called ChatGPT-4.0 on clinical cases. We wanted to see if it could help doctors by giving good advice on what might be wrong with patients who have heart issues and what should be done to help them. To test this, we used ChatGPT-4.0 to look at 20 different stories about patients with heart problems. These stories were made to cover a variety of common heart conditions faced by heart doctors. Then, we asked 18 heart doctors to check if the advice from ChatGPT-4.0 was good and made sense. What we found was quite interesting! Most of the time, the doctors agreed that the computer gave good advice on what might be wrong with the patients and how to help them. This means that this smart computer program could be a helpful tool for doctors, especially when they are trying to figure out tricky heart problems. But, it's important to say that computers like ChatGPT-4.0 are not ready to replace doctors. They are tools that can offer suggestions. Doctors still need to use their knowledge and experience to make the final call on what's best for their patients. In simple terms, our study shows that with more development and testing, AI like ChatGPT-4.0 could be a helpful assistant to doctors in treating heart disease, making sure patients get the best care possible.