- Book学术

发布求助

文献互助智能选刊最新文献

IF 1.3 3区农林科学 Q2 VETERINARY SCIENCES

American journal of veterinary research Pub Date : 2025-02-28 Print Date: 2025-03-01 DOI:10.2460/ajvr.24.10.0305

Aliva Bakshi, Jake Stetson, Lihua Wang, Jishu Shi, Doina Caragea, Laura C Miller

{"title":"Toward a rapid, sensitive, user-friendly, field-deployable artificial intelligence tool for enhancing African swine fever diagnosis and reporting.","authors":"Aliva Bakshi, Jake Stetson, Lihua Wang, Jishu Shi, Doina Caragea, Laura C Miller","doi":"10.2460/ajvr.24.10.0305","DOIUrl":null,"url":null,"abstract":"Objective: African swine fever (ASF) is a lethal and highly contagious transboundary animal disease with the potential for rapid international spread. Lateral flow assays (LFAs) are sometimes hard to read by the inexperienced user, mainly due to the LFA sensitivity and reading ambiguities. Our objective was to develop and implement an AI-powered tool to enhance the accuracy of LFA reading, thereby improving rapid and early detection for ASF diagnosis and reporting.Methods: Here, we focus on the development of a deep learning-assisted, smartphone-based AI diagnostic tool to provide accurate decisions with higher sensitivity. The tool employs state-of-the-art You Only Look Once (YOLO) models for image classification. The YOLO models were trained and evaluated using a dataset consisting of images where the lateral flow assays are manually labeled as positives or negatives. A prototype JavaScript website application for ASF reporting and visualization was created in Azure. The application maintains the distribution of the positive predictions on a map as the positive cases are submitted by users.Results: The performance of the models is evaluated using standard evaluation metrics for classification tasks, specifically accuracy, precision, recall, sensitivity, specificity, and F1 measure. We acquired 86.3 ± 7.9% average accuracy, 96.3 ± 2.04% average precision, 79 ± 13.20% average recall, and an average F1 score of 0.87 ± 0.088 across 3 different train/development/test splits of the datasets. Submitting a positive result of the deep learning model updates a map with a location marker for positive results.Conclusions: Combining clinical data learning and 2-step algorithms enables a point-of-need assay with higher accuracy.Clinical relevance: A rapid, sensitive, user-friendly, and deployable deep learning tool was developed for classifying LFA test images to enhance diagnosis and reporting, particularly in settings with limited laboratory resources.","PeriodicalId":7754,"journal":{"name":"American journal of veterinary research","volume":" ","pages":"S27-S37"},"PeriodicalIF":1.3000,"publicationDate":"2025-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11957874/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"American journal of veterinary research","FirstCategoryId":"97","ListUrlMain":"https://doi.org/10.2460/ajvr.24.10.0305","RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/3/1 0:00:00","PubModel":"Print","JCR":"Q2","JCRName":"VETERINARY SCIENCES","Score":null,"Total":0}

引用次数: 0

摘要

目的：非洲猪瘟（ASF）是一种致命的高传染性跨境动物疾病，有可能在国际范围内迅速传播。没有经验的用户有时很难读取侧流检测（LFA），这主要是由于 LFA 的灵敏度和读取模糊性造成的。我们的目标是开发和实施一种人工智能驱动的工具，以提高 LFA 读数的准确性，从而改善 ASF 诊断和报告的快速和早期检测。方法：在此，我们重点开发了一种深度学习辅助的、基于智能手机的人工智能诊断工具，以提供更高灵敏度的准确决策。该工具采用最先进的 "只看一次"（YOLO）模型进行图像分类。YOLO 模型是通过一个数据集进行训练和评估的，该数据集由人工标记为阳性或阴性的侧流检测图像组成。在 Azure 中创建了用于 ASF 报告和可视化的 JavaScript 网站应用程序原型。随着用户提交阳性病例，该应用程序会在地图上维护阳性预测的分布情况：我们使用分类任务的标准评估指标，特别是准确度、精确度、召回率、灵敏度、特异性和 F1 指标，对模型的性能进行了评估。在 3 个不同的训练/开发/测试数据集中，我们获得了 86.3 ± 7.9% 的平均准确率、96.3 ± 2.04% 的平均精确率、79 ± 13.20% 的平均召回率和 0.87 ± 0.088 的平均 F1 分数。提交深度学习模型的阳性结果后，地图上会更新阳性结果的位置标记：结论：结合临床数据学习和两步算法，可实现更高精度的需求点检测：开发了一种快速、灵敏、用户友好且可部署的深度学习工具，用于对 LFA 检测图像进行分类，以提高诊断和报告能力，尤其是在实验室资源有限的情况下。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Toward a rapid, sensitive, user-friendly, field-deployable artificial intelligence tool for enhancing African swine fever diagnosis and reporting.

Objective: African swine fever (ASF) is a lethal and highly contagious transboundary animal disease with the potential for rapid international spread. Lateral flow assays (LFAs) are sometimes hard to read by the inexperienced user, mainly due to the LFA sensitivity and reading ambiguities. Our objective was to develop and implement an AI-powered tool to enhance the accuracy of LFA reading, thereby improving rapid and early detection for ASF diagnosis and reporting.

Methods: Here, we focus on the development of a deep learning-assisted, smartphone-based AI diagnostic tool to provide accurate decisions with higher sensitivity. The tool employs state-of-the-art You Only Look Once (YOLO) models for image classification. The YOLO models were trained and evaluated using a dataset consisting of images where the lateral flow assays are manually labeled as positives or negatives. A prototype JavaScript website application for ASF reporting and visualization was created in Azure. The application maintains the distribution of the positive predictions on a map as the positive cases are submitted by users.

Results: The performance of the models is evaluated using standard evaluation metrics for classification tasks, specifically accuracy, precision, recall, sensitivity, specificity, and F1 measure. We acquired 86.3 ± 7.9% average accuracy, 96.3 ± 2.04% average precision, 79 ± 13.20% average recall, and an average F1 score of 0.87 ± 0.088 across 3 different train/development/test splits of the datasets. Submitting a positive result of the deep learning model updates a map with a location marker for positive results.

Conclusions: Combining clinical data learning and 2-step algorithms enables a point-of-need assay with higher accuracy.

Clinical relevance: A rapid, sensitive, user-friendly, and deployable deep learning tool was developed for classifying LFA test images to enhance diagnosis and reporting, particularly in settings with limited laboratory resources.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

American journal of veterinary research 农林科学-兽医学

CiteScore

1.70

自引率

10.00%

发文量

186

审稿时长

3 months

期刊介绍： The American Journal of Veterinary Research supports the collaborative exchange of information between researchers and clinicians by publishing novel research findings that bridge the gulf between basic research and clinical practice or that help to translate laboratory research and preclinical studies to the development of clinical trials and clinical practice. The journal welcomes submission of high-quality original studies and review articles in a wide range of scientific fields, including anatomy, anesthesiology, animal welfare, behavior, epidemiology, genetics, heredity, infectious disease, molecular biology, oncology, pharmacology, pathogenic mechanisms, physiology, surgery, theriogenology, toxicology, and vaccinology. Species of interest include production animals, companion animals, equids, exotic animals, birds, reptiles, and wild and marine animals. Reports of laboratory animal studies and studies involving the use of animals as experimental models of human diseases are considered only when the study results are of demonstrable benefit to the species used in the research or to another species of veterinary interest. Other fields of interest or animals species are not necessarily excluded from consideration, but such reports must focus on novel research findings. Submitted papers must make an original and substantial contribution to the veterinary medicine knowledge base; preliminary studies are not appropriate.