Borna, Sahar

In postoperative care, patient education and follow-up are pivotal for enhancing the quality of care and satisfaction. Artificial intelligence virtual assistants (AIVA) and large language models (LLMs) like Google BARD and ChatGPT-4 offer avenues for addressing patient queries using natural language processing (NLP) techniques. However, the accuracy and appropriateness of the information vary across these platforms, necessitating a comparative study to evaluate their efficacy in this domain. We conducted a study comparing AIVA (using Google Dialogflow) with ChatGPT-4 and Google BARD, assessing the accuracy, knowledge gap, and response appropriateness. AIVA demonstrated superior performance, with significantly higher accuracy (mean: 0.9) and lower knowledge gap (mean: 0.1) compared to BARD and ChatGPT-4. Additionally, AIVA’s responses received higher Likert scores for appropriateness. Our findings suggest that specialized AI tools like AIVA are more effective in delivering precise and contextually relevant information for postoperative care compared to general-purpose LLMs. While ChatGPT-4 shows promise, its performance varies, particularly in verbal interactions. This underscores the importance of tailored AI solutions in healthcare, where accuracy and clarity are paramount. Our study highlights the necessity for further research and the development of customized AI solutions to address specific medical contexts and improve patient outcomes.

Background: OpenAI’s ChatGPT (San Francisco, CA, USA) and Google’s Gemini (Mountain View, CA, USA) are two large language models that show promise in improving and expediting medical decision making in hand surgery. Evaluating the applications of these models within the field of hand surgery is warranted. This study aims to evaluate ChatGPT-4 and Gemini in classifying hand injuries and recommending treatment. Methods: Gemini and ChatGPT were given 68 fictionalized clinical vignettes of hand injuries twice. The models were asked to use a specific classification system and recommend surgical or nonsurgical treatment. Classifications were scored based on correctness. Results were analyzed using descriptive statistics, a paired two-tailed t-test, and sensitivity testing. Results: Gemini, correctly classifying 70.6% hand injuries, demonstrated superior classification ability over ChatGPT (mean score 1.46 vs. 0.87, p-value < 0.001). For management, ChatGPT demonstrated higher sensitivity in recommending surgical intervention compared to Gemini (98.0% vs. 88.8%), but lower specificity (68.4% vs. 94.7%). When compared to ChatGPT, Gemini demonstrated greater response replicability. Conclusions: Large language models like ChatGPT and Gemini show promise in assisting medical decision making, particularly in hand surgery, with Gemini generally outperforming ChatGPT. These findings emphasize the importance of considering the strengths and limitations of different models when integrating them into clinical practice.

With abundant information and interconnectedness among people, identifying knowledgeable individuals in specific domains has become crucial for organizations. Artificial intelligence (AI) algorithms have been employed to evaluate the knowledge and locate experts in specific areas, alleviating the manual burden of expert profiling and identification. However, there is a limited body of research exploring the application of AI algorithms for expert finding in the medical and biomedical fields. This study aims to conduct a scoping review of existing literature on utilizing AI algorithms for expert identification in medical domains. We systematically searched five platforms using a customized search string, and 21 studies were identified through other sources. The search spanned studies up to 2023, and study eligibility and selection adhered to the PRISMA 2020 statement. A total of 571 studies were assessed from the search. Out of these, we included six studies conducted between 2014 and 2020 that met our review criteria. Four studies used a machine learning algorithm as their model, while two utilized natural language processing. One study combined both approaches. All six studies demonstrated significant success in expert retrieval compared to baseline algorithms, as measured by various scoring metrics. AI enhances expert finding accuracy and effectiveness. However, more work is needed in intelligent medical expert retrieval.

Primary Care Physicians (PCPs) are the first point of contact in healthcare. Because PCPs face the challenge of managing diverse patient populations while maintaining up-to-date medical knowledge and updated health records, this study explores the current outcomes and effectiveness of implementing Artificial Intelligence-based Clinical Decision Support Systems (AI-CDSSs) in Primary Healthcare (PHC). Following the PRISMA-ScR guidelines, we systematically searched five databases, PubMed, Scopus, CINAHL, IEEE, and Google Scholar, and manually searched related articles. Only CDSSs powered by AI targeted to physicians and tested in real clinical PHC settings were included. From a total of 421 articles, 6 met our criteria. We found AI-CDSSs from the US, Netherlands, Spain, and China whose primary tasks included diagnosis support, management and treatment recommendations, and complication prediction. Secondary objectives included lessening physician work burden and reducing healthcare costs. While promising, the outcomes were hindered by physicians’ perceptions and cultural settings. This study underscores the potential of AI-CDSSs in improving clinical management, patient satisfaction, and safety while reducing physician workload. However, further work is needed to explore the broad spectrum of applications that the new AI-CDSSs have in several PHC real clinical settings and measure their clinical outcomes.

Background: Despite numerous measures used to prevent pressure ulcers, their growing prevalence in recent years is expected to continue as the population ages. This review aims to report the outcomes of the regenerative potential of MSCs in treating pressure ulcers, assessing the effectiveness of MSCs in treating pressure ulcers. Methods: A computerized search for articles on animal models that use MSCs as primary therapy to treat pressure ulcers, published from conception to present, was conducted using PubMed, MEDLINE, Embase, and CINAHL. Our search yielded 52 articles, narrowed to 44 after excluding duplicates. Results: Out of 52 articles collected from four databases, 11 met the inclusion criteria. A total of 11 articles published between 2008 and 2020 met the inclusion criteria. Eight studies were observational descriptive papers in animal models, and three were prospective. Six studies used autologous MSCs, while five used allogenic MSCs. Three studies were conducted in humans, and the remaining eight were conducted in animals. The most common method of cell delivery was an intradermal injection in the margins of the ulcer. All studies reported positive results, including improved wound healing, reduced inflammation, and improved tissue regeneration. Conclusions: MSCs have shown promising results in treating pressure ulcers in animal and clinical trials. The combination of MSCs and scaffold materials has also been studied and found to be effective in wound healing. A standardized human wound model has been proposed further to investigate the efficacy of cell-based therapies for chronic wounds. However, more research is needed to determine the best quantity of cells to apply for pressure ulcers and to ensure the safety and efficacy of these treatments in clinical settings.

Background: The use of bone morphogenic protein and mesenchymal stem cells has shown promise in promoting bone regeneration in calvarial defects. However, a systematic review of the available literature is needed to evaluate the efficacy of this approach. Methods: We comprehensively searched electronic databases using MeSH terms related to skull defects, bone marrow mesenchymal stem cells, and bone morphogenic proteins. Eligible studies included animal studies that used BMP therapy and mesenchymal stem cells to promote bone regeneration in calvarial defects. Reviews, conference articles, book chapters, and non-English language studies were excluded. Two independent investigators conducted the search and data extraction. Results: Twenty-three studies published between 2010 and 2022 met our inclusion criteria after a full-text review of the forty-five records found in the search. Eight of the 23 studies used mice as models, while 15 used rats. The most common mesenchymal stem cell was bone marrow-derived, followed by adipose-derived. BMP-2 was the most popular. Stem cells were embedded in Scaffold (13), Transduction (7), and Transfection (3), and they were delivered BMP to cells. Each treatment used 2 × 104–1 × 107 mesenchymal stem cells, averaging 2.26 × 106. Most BMP-transduced MSC studies used lentivirus. Conclusions: This systematic review examined BMP and MSC synergy in biomaterial scaffolds or alone. BMP therapy and mesenchymal stem cells in calvarial defects, alone, or with a scaffold regenerated bone. This method treats skull defects in clinical trials. The best scaffold material, therapeutic dosage, administration method, and long-term side effects need further study.

Pain is a complex and subjective experience, and traditional methods of pain assessment can be limited by factors such as self-report bias and observer variability. Voice is frequently used to evaluate pain, occasionally in conjunction with other behaviors such as facial gestures. Compared to facial emotions, there is less available evidence linking pain with voice. This literature review synthesizes the current state of research on the use of voice recognition and voice analysis for pain detection in adults, with a specific focus on the role of artificial intelligence (AI) and machine learning (ML) techniques. We describe the previous works on pain recognition using voice and highlight the different approaches to voice as a tool for pain detection, such as a human effect or biosignal. Overall, studies have shown that AI-based voice analysis can be an effective tool for pain detection in adult patients with various types of pain, including chronic and acute pain. We highlight the high accuracy of the ML-based approaches used in studies and their limitations in terms of generalizability due to factors such as the nature of the pain and patient population characteristics. However, there are still potential challenges, such as the need for large datasets and the risk of bias in training models, which warrant further research.