FreshRSS

🔒
❌ Acerca de FreshRSS
Hay nuevos artículos disponibles. Pincha para refrescar la página.
AnteayerTus fuentes RSS

AE-GPT: Using Large Language Models to extract adverse events from surveillance reports-A use case with influenza vaccine adverse events

by Yiming Li, Jianfu Li, Jianping He, Cui Tao

Though Vaccines are instrumental in global health, mitigating infectious diseases and pandemic outbreaks, they can occasionally lead to adverse events (AEs). Recently, Large Language Models (LLMs) have shown promise in effectively identifying and cataloging AEs within clinical reports. Utilizing data from the Vaccine Adverse Event Reporting System (VAERS) from 1990 to 2016, this study particularly focuses on AEs to evaluate LLMs’ capability for AE extraction. A variety of prevalent LLMs, including GPT-2, GPT-3 variants, GPT-4, and Llama2, were evaluated using Influenza vaccine as a use case. The fine-tuned GPT 3.5 model (AE-GPT) stood out with a 0.704 averaged micro F1 score for strict match and 0.816 for relaxed match. The encouraging performance of the AE-GPT underscores LLMs’ potential in processing medical data, indicating a significant stride towards advanced AE detection, thus presumably generalizable to other AE extraction tasks.

Development of a supportive care framework for breast cancer survivor's unmet needs: A modified Delphi study

Abstract

Aim

To establish a supportive care framework for addressing unmet needs among breast cancer survivors, providing practical guidance for healthcare providers to assess and manage these needs, ultimately enhancing the health outcomes and quality of life of breast cancer survivors.

Design

We conducted a two-round Delphi survey to gather expert opinions regarding the unmet needs supportive care framework for breast cancer survivors.

Methods

Initial framework identification and inquiry questionnaire creation was achieved via literature search and expert group discussions, which included 15 experts from nursing practice, clinical medicine, nursing management and nursing education was conducted using a Delphi survey. To establish consensus, a two-round Delphi poll was done, using criteria based on the mean (≥4.0), coefficient of variation (CV < 0.25) and percentage for entire score (≥20%).

Results

Experts reached a consensus, leading to six care modules, and 28 care entries: Tumour Detection Support (three care entries), Management of Complications of Antitumor Therapy (seven care entries), Healthy Lifestyle Management (five care entries), Sexual and Fertility Support (four care entries), Psychosocial Support (four care entries) and Resource and Linkage Support (five care entries).

Conclusion

To address breast cancer survivors' unmet needs, a supportive framework was developed to actively enhance their health outcomes. However, further refinement and feasibility testing using mobile devices or artificial intelligence are required.

Implications for the Profession and Patient Care

This pioneering framework prioritises addressing unmet needs and equips healthcare providers to assess and manage these needs effectively, facilitating the implementation of programs aimed at improving the well-being of breast cancer survivors.

Reporting Method

This study was guided by a modified guideline for the Conducting and Reporting of Delphi Studies (CREDES) (Palliative Medicine, 31(8), 684, 2017).

Patient or Public Contribution

No Patient or Public Contribution.

Trial and protocol registration

The Delphi study methodology does not require registration.

Exploration of machine learning models for surgical incision healing assessment based on thermal imaging: A feasibility study

Abstract

In this study, we explored the use of thermal imaging technology combined with computer vision techniques for assessing surgical incision healing. We processed 1189 thermal images, annotated by experts to define incision boundaries and healing statuses. Using these images, we developed a machine learning model based on YOLOV8, which automates the recognition of incision areas, lesion segmentation and healing classification. The dataset was divided into training, testing and validation sets in a 7:2:1 ratio. Our results show high accuracy rates in incision location recognition, lesion segmentation and healing classification, indicating the model's effectiveness as a precise and automated diagnostic tool for surgical incision healing assessment. Conclusively, our thermal image-based machine learning model demonstrates excellent performance in wound assessment, paving the way for its clinical application in intelligent and standardized wound management.

Knowledge and practice of nurses with respect to perioperative hypothermia prevention in the Northwest Amhara Regional State Referral Hospitals, Ethiopia: a cross-sectional study

Por: Woretaw · A. W. · Yimer Mekonnen · B. · Tsegaye · N. · Dellie · E.
Objectives

It has been reported that maintaining a normal body temperature among surgical patients can reduce the length of hospitalisation by up to 40%, decrease the risk of surgical site infection by 64% and reduce mortality by fourfold. Nurses are primarily responsible for preventing the occurrence of hypothermia among surgical patients. This study assessed nurses’ knowledge and practices with respect to perioperative hypothermia prevention in Northwest Ethiopia, and investigated the factors associated with good knowledge and practice.

Design

Cross-sectional study.

Setting

Northwest Amhara Regional State Referral Hospitals, Northwest Ethiopia, 25 March–20 May 2021.

Participants

413 nurses working in the perioperative units of five referral hospitals.

Outcome measures

Perioperative hypothermia prevention knowledge and practice among nurses.

Results

Nearly three-fifths (59.1%; 95% CI: 54.7% to 63.7%) of respondents had good knowledge and about half (50.4%; 95% CI: 45.5% to 55.0%) had good practice with respect to perioperative hypothermia prevention. Factors associated with nurses’ knowledge of prevention of perioperative hypothermia included male sex (adjusted OR (AOR): 1.61, 95% CI: 1.02 to 2.53), having a bachelor’s degree (AOR: 2.50, 95% CI: 1.25 to 5.00), having a master’s degree (AOR: 4.39, 95% CI: 1.45 to 13.20) and training participation (AOR: 3.68, 95% CI: 2.14 to 6.33). Factors associated with nurses’ practice of prevention of perioperative hypothermia included working in recovery (AOR: 2.87, 95% CI: 1.08 to 7.58) and intensive care units (AOR: 2.39, 95% CI: 1.09 to 5.22), training participation (AOR: 2.64, 95% CI: 1.53 to 4.57), being satisfied with their job (AOR: 2.15, 95% CI: 1.34 to 3.43) and having good knowledge (AOR: 2.64, 95% CI: 1.63 to 4.27).

Conclusion

Nurses’ knowledge and practice of the prevention of perioperative hypothermia were inadequate. Hospital managers need to design and strengthen training programmes and work to enhance job satisfaction.

Perspectives of HPV vaccination among young adults: a qualitative systematic review and evidence synthesis protocol

Por: Mantina · N. M. · Nakayima Miiro · F. · Smith · J. · McClelland · D. J. · Magrath · P. A. · Madhivanan · P.
Introduction

Human papillomavirus (HPV) is the causative agent of nearly all cervical cancers. Despite the proven safety and efficacy of HPV vaccines in preventing HPV-related cancers, the global vaccine coverage rate is estimated to only be 15%. HPV vaccine coverage rates are more actively tracked and reported for adolescents 17 years and younger but there is still a critical window of opportunity to intervene and promote HPV vaccination among young adults aged 18–26 years who are still eligible to be vaccinated. This protocol for a qualitative evidence synthesis aims to review perspectives of HPV vaccination among young adults (18–26 years) and identify facilitators and barriers that influence HPV vaccination uptake and decision-making.

Methods and analysis

Seven databases will be searched from 1 January 2006 to the date of final search. For inclusion, studies must report HPV vaccination perspectives of young adults aged 18–26 years and use qualitative study methods or analysis techniques. Studies will be screened in a two-stage process guided by the eligibility criteria. Final included studies will be evaluated for methodological strengths and limitations using the Critical Appraisal Skills Programme quality assessment tool for qualitative studies. After data extraction, framework analysis will be used to analyse the data applying the socioecological model. Finally, the Grading of Recommendations Assessment, Development and Evaluation - Confidence in the Evidence from Reviews of Qualitative research will be applied to evaluate the confidence in synthesised qualitative findings. The methodology of this review follows the Cochrane Handbook guidelines on qualitative evidence syntheses.

Ethics and dissemination

Formal ethical approval is not required for this study. Findings will be disseminated through peer-reviewed publications, conference presentations and professional networks.

PROSPERO registration number

CRD42023417052.

Associations of serum DNA methylation levels of chemokine signaling pathway genes with mild cognitive impairment (MCI) and Alzheimer’s disease (AD)

by Ting Zou, Xiaohui Zhou, Qinwen Wang, Yongjie Zhao, Meisheng Zhu, Lei Zhang, Wei Chen, Pari Abuliz, Haijun Miao, Keyimu Kabinur, Kader Alimu

Objective

To investigate the associations of serum DNA methylation levels of chemokine signaling pathway genes with Alzheimer’s disease (AD) and mild cognitive impairment (MCI) in elderly people in Xinjiang, China, and to screen out genes whose DNA methylation could distinguish AD and MCI.

Materials and methods

37 AD, 40 MCI and 80 controls were included in the present study. DNA methylation assay was done using quantitative methylation-specific polymerase chain reaction (qMSP). Genotyping was done using Sanger sequencing.

Results

DNA methylation levels of ADCY2, MAP2K1 and AKT1 were significantly different among AD, MCI and controls. In the comparisons of each two groups, AKT1 and MAP2K1’s methylation was both significantly different between AD and MCI (p MAP2K1’s methylation was also significantly different between MCI and controls. Therefore, AKT1’s methylation was considered as the candidate serum marker to distinguish AD from MCI, and its association with AD was independent of APOE ε4 allele (p AKT1 hypermethylation was an independent risk factor for AD and MAP2K1 hypomethylation was an independent risk factor for MCI in logistic regression analysis (p Conclusion

This study found that the serum of AKT1 hypermethylation is related to AD independently of APOE ε4, which was differentially expressed in the Entorhinal Cortex of the brain and was an independent risk factor for AD. It could be used as one of the candidate serum markers to distinguish AD and MCI. Serum of MAP2K1 hypomethylation is an independent risk factor for MCI.

Efficacy of mesenchymal stromal cells in the treatment of unexplained recurrent spontaneous abortion in mice: An analytical and systematic review of meta-analyses

by Xiaoxuan Zhao, Yijie Hu, Wenjun Xiao, Yiming Ma, Dan Shen, Yuepeng Jiang, Yi Shen, Suxia Wang, Jing Ma

Objectives

Unexplained recurrent spontaneous abortion (URSA) remains an intractable reproductive dilemma due to the lack of understanding of the pathogenesis. This study aimed to evaluate the preclinical evidence for the mesenchymal stromal cell (MSC) treatment for URSA.

Methods

A meticulous literature search was independently performed by two authors across the Cochrane Library, EMBASE, and PubMed databases from inception to April 9, 2023. Each study incorporated was assessed using the Systematic Review Centre for Laboratory Animal Experimentation (SYRCLE) risk of bias tool. The amalgamated standardized mean difference (SMD) accompanied by 95% confidence interval (CI) were deduced through a fixed-effects or random-effects model analysis.

Results

A total of ten studies incorporating 140 mice were subjected to data analysis. The MSC treatment yielded a significant reduction in the abortion rate within the URSA model (OR = 0.23, 95%CI [0.17, 0.3], PP = 0.01), IL10 (SMD 1.60, 95% CI [0.58, 2.61], P = 0.002), IFN-γ (SMD -1.66, 95%CI [-2.79, -0.52], P = 0.004), and TNF-α (SMD -1.98, 95% CI [-2.93, -1.04], PPP>0.05).

Conclusions

The findings underscore the considerable potential of MSCs in URSA therapy. Nonetheless, the demand for enhanced transparency in research design and direct comparisons between various MSC sources and administration routes in URSA is paramount to engendering robust evidence that could pave the way for successful clinical translation.

❌