A comparison of the accuracy of mushroom identification applications using digital photographs

蘑菇 鉴定(生物学) 蘑菇中毒 人工智能 计算机科学 计算机视觉 计算机图形学(图像) 生物 食品科学 植物
作者
Sarah E. Hodgson,Christine J. McKenzie,Tom W. May,Shaun L. Greene
出处
期刊:Clinical Toxicology [Informa]
卷期号:61 (3): 166-172 被引量:6
标识
DOI:10.1080/15563650.2022.2162917
摘要

To compare the accuracy of three popular mushroom identification software applications in identifying mushrooms involved in exposures reported to the Victorian Poisons Information Centre and Royal Botanic Gardens Victoria.Over the past 10 years, an increasing number of software applications have been developed for use on smart phones and tablet devices to identify mushrooms. We have observed an increase in poisonings after incorrect identification of poisonous species as edible, using these applications.We compared the accuracy of three iPhone™ and Android™ mushroom identification applications: Picture Mushroom (Next Vision Limited©), Mushroom Identificator (Pierre Semedard©), and iNaturalist (iNaturalist, California Academy of Sciences©). Each app was tested independently by three researchers using digital photographs of 78 specimens sent to the Victorian Poisons Information Centre and Royal Botanic Gardens Victoria over a two-year period, 2020-2021. Mushroom identification was confirmed by an expert mycologist. For each app, individual and combined results were compared.Picture Mushroom was the most accurate of the three apps and correctly identified 49% (95% CI [0-100]) of specimens, compared with Mushroom Identificator (35% [15-56]) and iNaturalist (35% [0-76]). Picture Mushroom correctly identified 44% of poisonous mushrooms [0-95], compared with Mushroom Identificator (30% [1-58]) and iNaturalist (40% [0-84), but Mushroom Identificator identified more specimens of Amanita phalloides correctly (67%), compared to Picture Mushroom (60%) and iNaturalist (27%). Amanita phalloides was falsely identified, twice by Picture Mushroom and once by iNaturalist.Mushroom identification applications may be useful future tools to assist clinical toxicologists and the general public in the accurate identification of mushrooms species but, at present, are not reliable enough to exclude exposure to potentially poisonous mushrooms when used alone.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
蜜意发布了新的文献求助10
刚刚
共享精神应助温茶采纳,获得10
1秒前
2秒前
DCdc555发布了新的文献求助10
2秒前
esder完成签到,获得积分10
2秒前
zhan发布了新的文献求助10
3秒前
w__k完成签到 ,获得积分10
3秒前
朱家晓完成签到,获得积分10
3秒前
4秒前
5秒前
5秒前
Aron完成签到,获得积分10
5秒前
冷艳的璎完成签到,获得积分10
6秒前
6秒前
柚子完成签到,获得积分10
6秒前
6秒前
ycy关闭了ycy文献求助
7秒前
木南应助二狗子采纳,获得10
8秒前
wuhao完成签到,获得积分10
8秒前
Jane发布了新的文献求助10
8秒前
10秒前
527020100发布了新的文献求助10
10秒前
lijingqi发布了新的文献求助10
10秒前
忧郁雅寒完成签到,获得积分10
12秒前
12秒前
13秒前
任天野应助小侯采纳,获得10
14秒前
hhxy完成签到,获得积分10
14秒前
情怀应助蜜意采纳,获得10
14秒前
NexusExplorer应助meiwei采纳,获得10
14秒前
青山发布了新的文献求助10
15秒前
ryy完成签到 ,获得积分10
16秒前
17秒前
18秒前
19秒前
19秒前
20秒前
20秒前
20秒前
21秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
Modern Epidemiology, Fourth Edition 5000
Kinesiophobia : a new view of chronic pain behavior 5000
Molecular Biology of Cancer: Mechanisms, Targets, and Therapeutics 3000
Digital Twins of Advanced Materials Processing 2000
Weaponeering, Fourth Edition – Two Volume SET 2000
Signals, Systems, and Signal Processing 610
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 纳米技术 化学工程 生物化学 物理 计算机科学 内科学 复合材料 催化作用 物理化学 光电子学 电极 冶金 细胞生物学 基因
热门帖子
关注 科研通微信公众号,转发送积分 6018778
求助须知:如何正确求助?哪些是违规求助? 7609483
关于积分的说明 16160244
捐赠科研通 5166562
什么是DOI,文献DOI怎么找? 2765340
邀请新用户注册赠送积分活动 1746976
关于科研通互助平台的介绍 1635419