Jenish Maharjan

👉🏼 Prompt engineering on leveraging large language models in generating response to InBasket messages 🤓 Sherry Yan 👇🏻 https://lnkd.in/eZpReUeH 🔍 Focus on data insights: - 📈 The study demonstrated a significant increase in physician acceptance of LLM-generated draft responses, rising from 62% to 84% after three iterations of prompt engineering. - 🗣️ Patients reported a notable improvement in the tone (78%) and overall quality (80%) of messages generated by the optimized prompts compared to the original ones. - 🤖 A majority of patients (76%) could not distinguish between human and LLM-generated responses, indicating high-quality output from the model. 💡 Main outcomes and implications: - 🔄 Iterative prompt engineering proved effective in enhancing the quality of responses to Patient Medical Advice Requests (PMARs), suggesting a viable method for improving clinician-patient communication. - 🧠 Clinicians expressed that the use of LLMs could significantly reduce cognitive load when managing InBasket messages, potentially leading to improved efficiency in clinical settings. - 💬 The findings highlight the potential for LLMs to bridge gaps in patient care by providing timely and relevant information, thus improving patient satisfaction. 📚 Field significance: - 🌐 This research underscores the transformative role of AI and LLMs in healthcare, particularly in managing communication and administrative tasks. - 🏥 The study contributes to the growing body of evidence supporting the integration of AI tools in clinical workflows, paving the way for future innovations in patient care. - 📊 It emphasizes the importance of user feedback in refining AI applications, ensuring that they meet the needs of both clinicians and patients effectively. 🗄️: [#AI #LargeLanguageModels #Healthcare #PatientCare #PromptEngineering #ClinicalCommunication #CognitiveLoad #PatientSatisfaction #InBasketMessages]

👉🏼 Unveiling the e-Servicescape of ChatGPT: Exploring User Psychology and Engagement in AI-Powered Chatbot Experiences 🤓 Minseong Kim 👇🏻 https://lnkd.in/eZZeTE-Y 🔍 Focus on data insights: - 📊 The study utilized structural equation modeling to analyze user responses, revealing significant correlations between e-servicescape dimensions and emotional states. - 🧠 Negative emotions were notably influenced by factors such as security and visual appeal, while positive emotions were linked to customization and interactivity. - 📈 User satisfaction was found to be a critical mediator between emotional responses and the intention to engage with ChatGPT. 💡 Main outcomes and implications: - 🌐 The findings highlight the importance of e-servicescapes in shaping user experiences with AI technologies, suggesting that businesses should prioritize these elements to enhance user engagement. - 🔄 Understanding the emotional drivers can help marketers tailor their strategies to improve user satisfaction and retention in digital environments. - 💬 The research extends the concept of environmental psychology into the realm of AI, providing a framework for future studies on user interaction with digital services. 📚 Field significance: - 📖 This study contributes to the growing body of literature on digital environmental psychology, emphasizing the role of emotional states in technology adoption. - 🔍 It offers valuable insights for both scholars and practitioners in the fields of marketing and user experience design, particularly in the context of AI-driven services. - 🌟 The implications of this research can inform the development of more effective AI interfaces that cater to user emotional needs. 🗄️: [#eServicescape #UserExperience #EmotionalDesign #AI #Chatbots #DigitalPsychology #MarketingInsights #UserEngagement]

👉🏼 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

👉🏼 How deep learning is transforming protein structural biology through advanced neural networks 🤓 Matthias Bochtler 👇🏻 https://lnkd.in/g9HaXxTN 🔍 Focus on data insights: - 📈 Deep neural networks like CNNs, LLMs, and GNNs are significantly enhancing protein structure prediction. - 🧬 Applications of AI are expanding in areas such as inverse folding and protein design, leading to more effective and efficient workflows. - 🤖 Machine learning models are being leveraged to translate complex biological data into understandable and actionable insights. 💡 Main outcomes and implications: - 🔄 The integration of advanced deep learning techniques is revolutionizing the methodology used in structural biology. - 🌐 This transformation highlights the necessity for structural biologists to adapt to new technologies for improved research outcomes. - 📊 Insights derived from AI applications are likely to accelerate discoveries in protein-related biological processes and therapeutics. 📚 Field significance: - 🔍 The convergence of artificial intelligence and structural biology is setting a new paradigm in the field, facilitating unprecedented levels of discovery. - 💡 This review serves as a comprehensive guide, not only for structural biologists but also for those with backgrounds in machine learning, physics, or statistics. - 📖 As new methodologies emerge, continuous education and adaptation will be essential for scientists to remain at the forefront of research. 🗄️: [#deepLearning] [#neuralNetworks] [#structuralBiology] [#proteinDesign] [#machineLearning] [#AI] [#CNN] [#LLM] [#GNN]

Scaffolding learning: From specific to generic with large language models This article discusses the limitations of large language models (LLMs) like ChatGPT in solving basic arithmetic problems despite their proficiency in complex math tasks. It proposes a method called Scaffolding Learning, which mimics the step-by-step learning process of students. The approach involves training LLMs on specific operations first, such as multiplication and division, before applying these skills to more generic tasks like word problems. This method is akin to Curriculum Training, where models are trained on progressively difficult tasks. Empirical studies indicate that once an LLM masters a specific skill, minimal additional training is needed to generalize that skill to broader applications. - Publication Date: September 20, 2024 - DOI: 10.1371/journal.pone.0310409 Link: https://lnkd.in/eHZr73ut

👉🏼 Fine-Tuned Bidirectional Encoder Representations From Transformers Versus ChatGPT for Text-Based Outpatient Department Recommendation: Comparative Study 🤓 Eunbeen Jo 👇🏻 https://lnkd.in/gx9UH89R 🔍 Focus on data insights: - 📊 ChatGPT demonstrated a 6.2% answer avoidance rate, indicating areas for improvement in responsiveness. - ⚖️ The fine-tuned KM-BERT model achieved an accuracy of 97.7%, highlighting its superior performance in medical specialty recommendations compared to ChatGPT's 93.9%. - 🚦 Precision and recall metrics indicate KM-BERT's effectiveness in correctly identifying relevant specialties, with precision at 57.0% and recall at 65.2%, whereas ChatGPT struggled with lower figures

👉🏼 A Systematic Review of ChatGPT and Other Conversational Large Language Models in Healthcare 🤓 Leyao Wang 👇🏻 https://lnkd.in/e--t65tY 🔍 Focus on data insights: - ChatGPT from OpenAI was the most popular conversational LLM in healthcare applications. - Concerns about reliability and bias were prevalent in research papers utilizing LLMs. - LLMs showed promising results in summarization and providing medical knowledge with high accuracy. 💡 Main outcomes and implications: - Future studies should prioritize improving the reliability of LLM applications in complex health-related tasks. - Investigation into how LLM applications introduce bias and privacy issues is crucial for further development. - Legal, social, and technical efforts are necessary to address concerns and enhance the application of LLMs in healthcare. 📚 Field significance: - Healthcare applications - Conversational Large Language Models - Data insights 🗄️: [#healthcare #LLMs #datainsights]

👉🏼 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

👉🏼 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]

👉🏼 An Autonomous AI Agent for Streamlined Multi-Omic Bioinformatics 🤓 Juexiao Zhou 👇🏻 https://lnkd.in/egsngA94 🔍 Focus on data insights: - 📊 AutoBA leverages large language models

👉🏼 Large language models in neurosurgery: a systematic review and meta-analysis 🤓 Advait Patil 👇🏻 https://lnkd.in/eE7Tt8jd 🔍 Focus on data insights: - 📊 The review analyzed 51 articles showcasing various applications of Large Language Models

👉🏼 Evaluating the efficacy of few-shot learning for GPT-4Vision in neurodegenerative disease histopathology: A comparative analysis with convolutional neural network model 🤓 Daisuke Ono 👇🏻 https://lnkd.in/e2RK6-xr 🔍 Focus on data insights: - GPT-4Vision accurately recognized staining techniques and tissue origin but struggled with specific lesion identification. - The interpretation of images was notably influenced by the provided textual context, sometimes leading to diagnostic inaccuracies. - Few-shot learning significantly improved GPT-4Vision's diagnostic capabilities, enhancing accuracy from 40% in zero-shot learning to 90% with 20-shot learning. 💡 Main outcomes and implications: - GPT-4Vision faces challenges in independently interpreting histopathological images. - Few-shot learning is a promising approach to enhance GPT-4Vision's performance, especially in neuropathology where acquiring extensive labeled datasets is challenging. 📚 Field significance: - Artificial intelligence advancements like GPT-4Vision have expanded the potential for medical image interpretation. - Comparative analysis with traditional convolutional neural network models provides insights into improving diagnostic accuracy in neurodegenerative disease histopathology. 🗄️: #neurodegenerative #histopathology #GPT-4Vision #fewshotlearning