Generative Artificial Intellegence (AI) in Pathology and Medicine: A Deeper Dive DOI Creative Commons
Hooman H. Rashidi, Joshua Pantanowitz, Alireza Chamanzar

et al.

Modern Pathology, Journal Year: 2024, Volume and Issue: unknown, P. 100687 - 100687

Published: Dec. 1, 2024

This review article builds upon the introductory piece in our seven-part series, delving deeper into transformative potential of generative artificial intelligence (Gen AI) pathology and medicine. The explores applications Gen AI models medicine, including use custom chatbots for diagnostic report generation, synthetic image synthesis training new models, dataset augmentation, hypothetical scenario generation educational purposes, multimodal along with multi-agent models. also provides an overview common categories within discussing open-source closed-source as well specific examples popular such GPT-4, Llama, Mistral, DALL-E, Stable Diffusion their associated frameworks (e.g. transformers, GANs, diffusion-based neural networks), limitations challenges, especially medical domain. We libraries, tools that are currently deemed necessary to build integrate Finally, we look future, impact on healthcare, benefits, concerns related privacy, bias, ethics, API costs, security measures.

Language: Английский

How to optimize the systematic review process using AI tools DOI Creative Commons
Nicholas Fabiano, Arnav Gupta,

Nishaant Bhambra

et al.

JCPP Advances, Journal Year: 2024, Volume and Issue: 4(2)

Published: April 23, 2024

Systematic reviews are a cornerstone for synthesizing the available evidence on given topic. They simultaneously allow gaps in literature to be identified and provide direction future research. However, due ever-increasing volume complexity of literature, traditional methods conducting systematic less efficient more time-consuming. Numerous artificial intelligence (AI) tools being released with potential optimize efficiency academic writing assist various stages review process including developing refining search strategies, screening titles abstracts inclusion or exclusion criteria, extracting essential data from studies summarizing findings. Therefore, this article we an overview currently how they can incorporated into improve quality research synthesis. We emphasize that authors must report all AI have been used at each stage ensure replicability as part reporting methods.

Language: Английский

Citations

23
The Role of Large Language Models (LLMs) in Providing Triage for Maxillofacial Trauma Cases: A Preliminary Study DOI Creative Commons
Andrea Frosolini, Lisa Catarzi, Simone Benedetti

et al.

Diagnostics, Journal Year: 2024, Volume and Issue: 14(8), P. 839 - 839

Published: April 18, 2024

In the evolving field of maxillofacial surgery, integrating advanced technologies like Large Language Models (LLMs) into medical practices, especially for trauma triage, presents a promising yet largely unexplored potential. This study aimed to evaluate feasibility using LLMs triaging complex cases by comparing their performance against expertise tertiary referral center.

Language: Английский

Citations

16
Disparities in medical recommendations from AI-based chatbots across different countries/regions DOI Creative Commons

Khanisyah E. Gumilar,

Birama Robby Indraprasta,

Yu-Cheng Hsu

et al.

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: July 24, 2024

This study explores disparities and opportunities in healthcare information provided by AI chatbots. We focused on recommendations for adjuvant therapy endometrial cancer, analyzing responses across four regions (Indonesia, Nigeria, Taiwan, USA) three platforms (Bard, Bing, ChatGPT-3.5). Utilizing previously published cases, we asked identical questions to chatbots from each location within a 24-h window. Responses were evaluated double-blinded manner relevance, clarity, depth, focus, coherence ten experts cancer. Our analysis revealed significant variations different countries/regions (p < 0.001). Interestingly, Bing's Nigeria consistently outperformed others 0.05), excelling all evaluation criteria Bard also performed better compared other surpassing them categories 0.001, with relevance reaching p 0.01). Notably, Bard's overall scores significantly higher than those of ChatGPT-3.5 Bing locations These findings highlight the quality AI-powered based user platform. emphasizes necessity more research development guarantee equal access trustworthy medical through technologies.

Language: Английский

Citations

10
Understanding natural language: Potential application of large language models to ophthalmology DOI Creative Commons
Zefeng Yang, Biao Wang, Fengqi Zhou

et al.

Asia-Pacific Journal of Ophthalmology, Journal Year: 2024, Volume and Issue: 13(4), P. 100085 - 100085

Published: July 1, 2024

Large language models (LLMs), a natural processing technology based on deep learning, are currently in the spotlight. These closely mimic comprehension and generation. Their evolution has undergone several waves of innovation similar to convolutional neural networks. The transformer architecture advancement generative artificial intelligence marks monumental leap beyond early-stage pattern recognition via supervised learning. With expansion parameters training data (terabytes), LLMs unveil remarkable human interactivity, encompassing capabilities such as memory retention comprehension. advances make particularly well-suited for roles healthcare communication between medical practitioners patients. In this comprehensive review, we discuss trajectory their potential implications clinicians For clinicians, can be used automated documentation, given better inputs extensive validation, may able autonomously diagnose treat future. patient care, triage suggestions, summarization documents, explanation patient's condition, customizing education materials tailored level. limitations possible solutions real-world use also presented. Given rapid advancements area, review attempts briefly cover many that play ophthalmic space, with focus improving quality delivery.

Language: Английский

Citations

7
LLMs for science: Usage for code generation and data analysis DOI Creative Commons

Mohamed Nejjar,

Luca Zacharias,

Fabian Stiehle

et al.

Journal of Software Evolution and Process, Journal Year: 2024, Volume and Issue: unknown

Published: Sept. 12, 2024

Abstract Large language models (LLMs) have been touted to enable increased productivity in many areas of today's work life. Scientific research as an area is no exception: The potential LLM‐based tools assist the daily scientists has become a highly discussed topic across disciplines. However, we are only at very onset this subject study. It still unclear how LLMs will materialize practice. With study, give first empirical evidence on use process. We investigated set cases for scientific and conducted study assess which degree current helpful. In position paper, report specifically related software engineering, specifically, generating application code developing scripts data analytics visualization. While studied seemingly simple cases, results differ significantly. Our highlight promise general, yet also observe various issues, particularly regarding integrity output these provide.

Language: Английский

Citations

4
A guide to prompt design: foundations and applications for healthcare simulationists DOI Creative Commons

Sara Maaz,

Janice C. Palaganas,

Gerry Palaganas

et al.

Frontiers in Medicine, Journal Year: 2025, Volume and Issue: 11

Published: Jan. 30, 2025

Large Language Models (LLMs) like ChatGPT, Gemini, and Claude gain traction in healthcare simulation; this paper offers simulationists a practical guide to effective prompt design. Grounded structured literature review iterative testing, proposes best practices for developing calibrated prompts, explores various types techniques with use cases, addresses the challenges, including ethical considerations using LLMs simulation. This helps bridge knowledge gap on LLM simulation-based education, offering tailored guidance Examples were created through testing ensure alignment simulation objectives, covering cases such as clinical scenario development, OSCE station creation, simulated person scripting, debriefing facilitation. These provide easy-to-apply methods enhance realism, engagement, educational simulations. Key challenges associated integration, bias, privacy concerns, hallucinations, lack of transparency, need robust oversight evaluation, are discussed alongside unique education. Recommendations provided help craft prompts that align objectives while mitigating these challenges. By insights, contributes valuable, timely seeking leverage generative AI’s capabilities education responsibly.

Language: Английский

Citations

0
Large language model‐supported interactive case‐based learning: a pilot study DOI Open Access

Haelynn Gim,

Benjamin K. Cook,

Jasmin Le

et al.

Internal Medicine Journal, Journal Year: 2025, Volume and Issue: unknown

Published: March 24, 2025

Abstract Large language models (LLMs) have been proposed as a means to augment case‐based learning but are prone generating factually incorrect content. In this study, an LLM‐based tool was developed, and its performance evaluated. response student‐generated questions, the LLM adhered provided screenplay in 832/857 (97.1%) instances, remaining it medically appropriate 24/25 (96.0%) cases. Use of appears be feasible for purpose, further studies required examine their educational impact.

Language: Английский

Citations

0
Cut the crap: a critical response to “ChatGPT is bullshit” DOI Creative Commons
David J. Gunkel, Simon Coghlan

Ethics and Information Technology, Journal Year: 2025, Volume and Issue: 27(2)

Published: April 18, 2025

Language: Английский

Citations

0
Ethical AI Integration in Academia DOI
Zander Janse van Rensburg, Sonja van der Westhuizen

Advances in educational technologies and instructional design book series, Journal Year: 2024, Volume and Issue: unknown, P. 23 - 48

Published: March 22, 2024

The chapter explores the transformative potential and challenges of integrating large language models (LLMs) into higher education. It highlights opportunities AI presents for enhancing academic literacy, writing, pedagogy, while also acknowledging risks to traditional educational values practices. proposes a framework, developed with guidance information digital integrity, aimed at leveraging AI's capabilities support success without undermining foundational skills. discussion extends implications in South African context, addressing divide advocating equitable access technology. This encapsulates essence proposed proactive framework navigating impact on academia, focusing adaptation, critical engagement, cultivation an advanced form literacy that integrates technologies responsibly.

Language: Английский

Citations

2
OptiComm-GPT: a GPT-based versatile research assistant for optical fiber communication systems DOI Creative Commons
Xiaotian Jiang, Min Zhang, Yuchen Song

et al.

Optics Express, Journal Year: 2024, Volume and Issue: 32(12), P. 20776 - 20776

Published: May 14, 2024

With the increasing capacity and complexity of optical fiber communication systems, both academic industrial requirements for essential tasks transmission systems simulation, digital signal processing (DSP) algorithms verification, system performance evaluation, quality (QoT) optimization are becoming significantly important. However, due to intricate nonlinear nature these generally implemented in a divide-and-conquer manner, which necessitates profound level expertise proficiency software programming from researchers or engineers. To lower this threshold facilitate professional research easy-to-start, GPT-based versatile assistant named OptiComm-GPT is proposed flexibly automatically performs DSP QoT with only natural language. enhance OptiComm-GPT's abilities complex communications improve accuracy generated results, domain information base containing rich knowledge, tools, data as well comprehensive prompt engineering well-crafted elements, techniques, examples established under LangChain-based framework. The evaluated multiple tasks, results show that can effectively comprehend user's intent, accurately extract parameters request, intelligently invoke resources solve simultaneously. Moreover, statistical typical errors, running time also investigated illustrate its practical reliability, potential limitations, further improvements.

Language: Английский

Citations

2