Опубликована: Янв. 1, 2024
Язык: Английский
Cogent Social Sciences, Год журнала: 2025, Номер 11(1)
Опубликована: Март 8, 2025
Язык: Английский
Процитировано
0
Transplantation Proceedings, Год журнала: 2025, Номер unknown
Опубликована: Март 1, 2025
Язык: Английский
Процитировано
0
Wiley Interdisciplinary Reviews Data Mining and Knowledge Discovery, Год журнала: 2025, Номер 15(2)
Опубликована: Апрель 9, 2025
ABSTRACT This paper reviews benchmarking methods for evaluating large language models (LLMs) in healthcare settings. It highlights the importance of rigorous to ensure LLMs' safety, accuracy, and effectiveness clinical applications. The review also discusses challenges developing standardized benchmarks metrics tailored healthcare‐specific tasks such as medical text generation, disease diagnosis, patient management. Ethical considerations, including privacy, data security, bias, are addressed, underscoring need multidisciplinary collaboration establish robust frameworks that facilitate reliable ethical use healthcare. Evaluation LLMs remains challenging due lack comprehensive datasets. Key concerns include model better explainability, all which impact overall trustworthiness
Язык: Английский
Процитировано
0
Communications in computer and information science, Год журнала: 2025, Номер unknown, С. 260 - 270
Опубликована: Янв. 1, 2025
Язык: Английский
Процитировано
0
Biomedical Journal, Год журнала: 2025, Номер unknown, С. 100868 - 100868
Опубликована: Апрель 1, 2025
Large Language Models (LLMs) are capable of transforming healthcare by demonstrating remarkable capabilities in language understanding and generation. They have matched or surpassed human performance standardized medical examinations assisted diagnostics across specialties like dermatology, radiology, ophthalmology. LLMs can enhance patient education providing accurate, readable, empathetic responses, they streamline clinical workflows through efficient information extraction from unstructured data such as notes. Integrating LLM into practice involves user interface design, clinician training, effective collaboration between Artificial Intelligence (AI) systems professionals. Users must possess a solid generative AI domain knowledge to assess the generated content critically. Ethical considerations ensure privacy, security, mitigating biases, maintaining transparency critical for responsible deployment. Future directions include interdisciplinary collaboration, developing new benchmarks that incorporate safety ethical measures, advancing multimodal integrate text imaging data, creating LLM-based agents complex decision-making, addressing underrepresented rare diseases, integrating with robotic precision procedures. Emphasizing safety, integrity, human-centered implementation is essential maximizing benefits LLMs, while potential risks, thereby helping these tools rather than replace expertise compassion healthcare.
Язык: Английский
Процитировано
0
International Journal of Human-Computer Interaction, Год журнала: 2024, Номер unknown, С. 1 - 32
Опубликована: Ноя. 25, 2024
Язык: Английский
Процитировано
3
Journal of Multidisciplinary Healthcare, Год журнала: 2024, Номер Volume 17, С. 5091 - 5104
Опубликована: Ноя. 1, 2024
In healthcare applications, AI-driven innovations are set to revolutionise patient interactions and care, with the aim of improving satisfaction. Recent advancements in Artificial Intelligence have significantly affected nursing, assistive management, medical diagnoses, other critical procedures.
Язык: Английский
Процитировано
2
Symmetry, Год журнала: 2024, Номер 16(4), С. 478 - 478
Опубликована: Апрель 15, 2024
The processing and comprehension of numerical information in natural language represent pivotal focal points scholarly inquiry. Across diverse applications spanning text analysis to retrieval, the adept management understanding content within are indispensable achieving task success. Specialized encoding embedding techniques tailored data offer an avenue toward improved performance tasks involving masked prediction reasoning, inherently characterized by values. Consequently, treating numbers merely as words is inadequate; their semantics must be underscored. Recent years have witnessed emergence a range specific methodologies designed explicitly for content, demonstrating promising outcomes. We observe similarities between Transformer architecture CPU architecture, with symmetry playing crucial role. In light this observation drawing inspiration from computer system theory, we introduce floating-point representation devise corresponding module. representations correspond one-to-one semantic vector values, rendering both symmetric regarding intermediate transformation methods. Our proposed methodology facilitates more comprehensive predefined precision range, thereby ensuring distinctive each entity. Rigorous testing on multiple encoder-only models datasets yielded results that stand out terms competitiveness. comparison default methods employed models, our approach achieved improvement approximately 3.8% Top-1 accuracy reduction perplexity 0.43. These outcomes affirm efficacy method. Furthermore, enrichment through contributes augmentation model’s capacity understanding.
Язык: Английский
Процитировано
1
Опубликована: Авг. 16, 2024
Background: Pharmacists need up-to-date knowledge and decision-making support in HIV care. We aim to develop MARVIN-Pharma, an adapted artificial intelligence-based chatbot initially for people with HIV, assist pharmacists, considering evidence-based needs. Methods: From Dec 2022 2023, online needs-assessment survey evaluated Québec pharmacists' knowledge, attitudes, involvement, barriers relative care, alongside perceptions relevant the usability of MARVIN-Pharma. Recruitment involved convenience snowball sampling, targeting National Hepatitis Mentoring Program affiliates. Results: Forty-one pharmacists (28 community, 13 hospital-based) across 15 municipalities participated. Participants perceived their as moderate (M=3.74/6). They held largely favorable attitudes towards providing care (M=4.02/6). reported a “little” involvement delivery services (M=2.08/5), most often ART adherence counseling, refilling, monitoring. The common were lack time, staff resources, clinical tools, information/training, at least somewhat agreeing that they experienced each (M≥4.00/6). On average, MARVIN-Pharma's acceptability compatibility ‘undecided’ range (M=4.34, M=4.13/7, respectively), while agreed self-efficacy use health (M=5.6/7). Conclusion: MARVIN-Pharma might help address gaps treatment but pharmacist engagement chatbot’s development seems vital its future uptake usability.
Язык: Английский
Процитировано
1
Healthcare, Год журнала: 2024, Номер 12(16), С. 1661 - 1661
Опубликована: Авг. 20, 2024
Background: Pharmacists need up-to-date knowledge and decision-making support in HIV care. We aim to develop MARVIN-Pharma, an adapted artificial intelligence-based chatbot initially for people with HIV, assist pharmacists considering evidence-based needs. Methods: From December 2022 2023, online needs-assessment survey evaluated Québec pharmacists’ knowledge, attitudes, involvement, barriers relative care, alongside perceptions relevant the usability of MARVIN-Pharma. Recruitment involved convenience snowball sampling, targeting National Hepatitis Mentoring Program affiliates. Results: Forty-one (28 community, 13 hospital-based) across 15 municipalities participated. Participants perceived their as moderate (M = 3.74/6). They held largely favorable attitudes towards providing care 4.02/6). reported a “little” involvement delivery services 2.08/5), most often ART adherence counseling, refilling, monitoring. The common were lack time, staff resources, clinical tools, information/training, at least somewhat agreeing that they experienced each ≥ 4.00/6). On average, MARVIN-Pharma’s acceptability compatibility ‘undecided’ range 4.34, M 4.13/7, respectively), while agreed self-efficacy use health 5.6/7). Conclusion: MARVIN-Pharma might help address gaps treatment but pharmacist engagement chatbot’s development seems vital its future uptake usability.
Язык: Английский
Процитировано
1