Darko Medin

👉🏼 Comparative Analysis of Artificial Intelligence (AI) Languages in Predicting Sequential Organ Failure Assessment (SOFA) Scores 🤓 Fuat H Saner 👇🏻 https://lnkd.in/eTihTeEv 🔍 Focus on data insights: - ChatGPT 4.0 Plus showed the strongest correlation with the actual SOFA score (r=0.92). - Bard exhibited a moderate correlation with the actual SOFA score (r=0.59). - Perplexity AI demonstrated a substantial correlation with the actual SOFA score (r=0.89). 💡 Main outcomes and implications: - AI tools, especially ChatGPT 4.0 Plus, hold promise for automating SOFA score calculations in ICU settings. - These tools can reduce manual workload and enhance continuous patient monitoring and assessment efficiency. - Further development and validation are essential to ensure accuracy and reliability in critical care environments. 📚 Field significance: - AI data mining offers a transformative approach to improving patient care in intensive care units. - Automation of SOFA score calculations can streamline clinical workflows and enhance patient outcomes. 🗄️: [#ArtificialIntelligence #SOFA #ICU #DataInsights #PatientMonitoring]

👉🏼 Physics-informed genetic programming for discovery of partial differential equations from scarce and noisy data 🤓 Benjamin G Cohen 👇🏻 https://lnkd.in/erTv89Nx 🔍 Focus on data insights: - 📈 The proposed framework identifies free-form partial differential equations

👉🏼 AI-generated text in otolaryngology publications: a comparative analysis before and after the release of ChatGPT 🤓 Jonathan M Carnino 👇🏻 https://lnkd.in/eUspYs5s 🔍 Focus on data insights: - A significant increase in the mean percentage of AI-generated text post-ChatGPT release - Higher percentage of AI-generated text in publications from authors in non-English speaking countries 💡 Main outcomes and implications: - Impact on writing practices among researchers in JAMA Otolaryngology - Head and Neck Surgery - Concerns over the accuracy of AI-created content and potential misinformation risks 📚 Field significance: - Evolving dynamics between AI technologies, scientific communication, and publication integrity - Urgent need for continued research in the field of AI-driven scientific writing 🗄️: [#AI #otolaryngology #ChatGPT #scientificcommunication #AItools]

👉🏼 ChatGPT as an effective tool for quality evaluation of radiomics research 🤓 Ismail Mese 👇🏻 https://lnkd.in/eGks8btb 🔍 Focus on data insights: - 📊 ChatGPT-4o achieved a median radiomics quality score

👉🏼 Potential application of ChatGPT in Helicobacter pylori disease relevant queries 🤓 Zejun Gao 👇🏻 https://lnkd.in/emVwfvch 🔍 Focus on data insights: - 📊 The use of ChatGPT-4 showed a high level of repeatability and adequate response length across various prompts related to Helicobacter pylori. - 🧠 Expert evaluations indicated strong scores in accuracy, completeness, usefulness, comprehension, and satisfaction regarding ChatGPT's responses. - 🗣️ There is a notable difference in the perceived usefulness and comprehension of responses between experts and ordinary people, with medical students having a more favorable view. 💡 Main outcomes and implications: - 🔬 ChatGPT-4 has demonstrated the potential to be an effective tool for addressing specific medical queries, particularly concerning Helicobacter pylori. - 🏫 The platform may serve as a valuable resource for medical education, especially for medical students seeking supplemental information. - 🔄 Further research is essential to validate these findings and to explore the broader applicability of AI-driven tools in medical settings. 📚 Field significance: - 🌐 The integration of AI language models into healthcare could enhance patient education and inform clinical practice. - 💡 This study suggests that large language models might bridge knowledge gaps in healthcare communication, signaling a shift towards technology-assisted learning and support. - 📈 The findings advocate for continued exploration of AI capabilities in delivering accurate medical information, which may lead to improved patient outcomes. 🗄️: [#ChatGPT #HelicobacterPylori #ArtificialIntelligence #MedicalEducation #PatientCare #DataInsights #AIinHealthcare]

👉🏼 Towards automated phenotype definition extraction using large language models 🤓 Ramya Tekumalla 👇🏻 https://lnkd.in/ezhYqxVd 🔍 Focus on data insights: 🌟 Electronic phenotyping leverages both structured and unstructured data to enhance accuracy. 📊 The study highlights the inefficiency of current methods, which rely heavily on manual literature reviews. 🤖 Large language models present a potential solution, moving towards automation in phenotype extraction. 💡 Main outcomes and implications: ✅ A standard evaluation set was developed to ensure the reliability of outputs from large language models. 🧪 Evaluation of various prompting approaches showed promise in extracting more nuanced phenotype definitions. ⚠️ Despite progress, human validation remains crucial to mitigate risks of non-factual data and misleading results. 📚 Field significance: 🔬 The proposed advancements could significantly reduce the time spent in literature review and evaluation. 🚀 Automation in electronic phenotyping may lead to broader application in clinical settings, improving efficiency and scalability. 🌐 Heightened interest in the intersection of machine learning and clinical informatics could foster further innovations. 🗄️: [#genomics] [#phenotyping] [#machinelearning] [#naturallanguagemodels] [#dataextraction] [#clinicalinformatics]

👉🏼 Developing large language models to detect adverse drug events in posts on X 🤓 Yu Deng 👇🏻 https://lnkd.in/eVVYKA8b 🔍 Focus on data insights: - 📊 The study developed several large language models

👉🏼 Application of Artificial Intelligence in the Headache Field 🤓 Keiko Ihara 👇🏻 https://lnkd.in/ehrnf-_C 🔍 Focus on data insights: - Machine learning models and evaluation metrics are briefly introduced. - AI is utilized to enhance diagnostic accuracy, classification, treatment response prediction, data insights, and migraine attack forecasting. - The potential of large language models (LLMs) like ChatGPT in advancing the field is discussed. 💡 Main outcomes and implications: - Incorporation of machine learning, generative AI, and LLMs in recent headache medicine studies. - Awareness of potential pitfalls and biases is crucial for the ethical use of AI in headache medicine. - AI has the potential to revolutionize headache medicine when employed appropriately. 📚 Field significance: - AI offers promising solutions to unmet needs in headache research. - Advancements in AI can lead to improved diagnosis, treatment, and understanding of headache disorders. 🗄️: #ArtificialIntelligence #HeadacheField #MachineLearning #DataInsights #MedicalResearch

👉🏼 Pre-trained Language Models in Medicine: A Survey 🤓 Xudong Luo https://lnkd.in/eubyPuA5 🔍 Focus on data insights: - Pre-trained Language Models (PLMs) like BERT, BioBERT, and ChatGPT have shown potential in medical NLP tasks. - Survey covers achievements and methodologies of applying PLMs in medical NLP tasks. 💡 Main outcomes and implications: - Overview of basic concepts, methodologies, advantages, and steps of applying PLMs in medical NLP tasks. - Summary of recent research findings, strengths vs weaknesses, and potential limitations. 📚 Field significance: - Enhancing models' reliability, explainability, and fairness for promoting PLMs in clinical practice. 🗄️: #NaturalLanguageProcessing #PretrainedModels #MedicalNLP #DataInsights

👉🏼 Evaluating Large Language Models on their Accuracy and Completeness: Immune Checkpoint Inhibitors and their Ocular Toxicities 🤓 Camellia Edalat 👇🏻 https://lnkd.in/eXu_uEfz 🔍 Focus on data insights: - 📊 ChatGPT, Bard, and LLaMA demonstrated comparable performance in accuracy

👉🏼 Perceptions and detection of AI use in manuscript preparation for academic journals 🤓 Nir Chemaya 👇🏻 https://lnkd.in/eM4HxwiP 🔍 Focus on data insights: - The survey of 271 academics revealed differing opinions on reporting ChatGPT use in manuscript preparation. - AI detection software flagged abstracts using GPT for grammar fixes as potentially AI-written, despite differing academic perceptions. - Disagreements among academics on reporting more substantial ChatGPT text rewriting were influenced by ethics, academic role, and English language background. 💡 Main outcomes and implications: - Authors may face challenges in reporting AI use due to varying perceptions on what should be disclosed. - Journals may struggle to enforce nuanced reporting requirements using AI detection tools. 📚 Field significance: - Emerging norms around AI use in manuscript preparation require clarity and consensus to ensure transparent reporting practices. 🗄️: #AI #manuscriptpreparation #academicwriting #datainsights

👉🏼 Performance of Novel GPT-4 in Otolaryngology Knowledge Assessment 🤓 Lucy Revercomb 👇🏻 [Link to Article](https://lnkd.in/eKD5MTse) 🔍 Focus on data insights: - 📊 GPT-4 achieved a 72.0% accuracy rate, significantly outperforming GPT-3.5, which scored only 51.3%. - 🔬 A specialized custom GPT-4 model had an improved accuracy of 81.3%, highlighting the effectiveness of tailoring AI for specific fields. - ⚖️ Statistical analysis revealed that the customized model had 2.19 times higher odds of answering questions correctly compared to the standard GPT-4. 💡 Main outcomes and implications: - ✅ The study illustrates that enhanced AI models like GPT-4 can substantially impact medical education by providing more accurate assessments of knowledge. - 📚 Customization is key; AI models designed with specific educational goals in mind can yield better outcomes in student learning and assessment. - 🧑‍⚕️ This advancement suggests a shift toward integrating AI technology in professional training programs, potentially revolutionizing how medical knowledge is taught and evaluated. 📚 Field significance: - 🏥 The findings support the integration of advanced AI in clinical education, improving readiness for board examinations in specialties like otolaryngology. - 🌐 Results may encourage further research into specialized AI applications across various medical fields, leading to broader educational improvements. - 🚀 As AI continues to evolve, its role in enhancing medical competencies and education will likely become increasingly critical. 🗄️: [#GPT4] [#Otolaryngology] [#MedicalEducation] [#AIinHealthcare] [#KnowledgeAssessment]

👉🏼 Me-LLaMA: Foundation Large Language Models for Medical Applications 🤓 Qianqian Xie 👇🏻 https://lnkd.in/eQJREuAf 🔍 Focus on data insights: - Me-LLaMA models achieve better performance than existing open-source medical LLMs in zero-shot, few-shot, and supervised learning abilities. - Task-specific instruction tuning improves performance compared to other models. - Me-LLaMA models outperform ChatGPT and GPT-4 on various datasets. 💡 Main outcomes and implications: - Me-LLaMA is a significant advancement in medical language models, showcasing superior performance across general and medical tasks. - The study addresses limitations in applying LLMs to clinical settings due to a lack of specialized training on medical-specific data. - Me-LLaMA models mitigate the catastrophic forgetting problem better than other medical LLMs. 📚 Field significance: - Me-LLaMA introduces a novel approach to medical language models, leveraging large datasets and continual pre-training for improved performance. - The study highlights the potential of Me-LLaMA in revolutionizing medical applications and advancing AI in healthcare. 🗄️: [#medicalapplications #LLMs #datainsights #MeLLaMA #healthcareAI]