ArticlePublisher preview availableLiterature Review

Deep learning in disease vector image identification

Wiley
Pest Management Science
Authors:
Shaowen Bai
Shaowen Bai
Shaowen Bai
  • This person is not on ResearchGate, or hasn't claimed this research yet.
Liang Shi
Liang Shi
Liang Shi
  • This person is not on ResearchGate, or hasn't claimed this research yet.
Kun Yang at Chinese Center For Disease Control And Prevention
Kun Yang
Read publisher preview
Download citation
To read the full-text of this research, you can request a copy directly from the authors.

Abstract and Figures

Vector‐borne diseases (VBDs) represent a critical global public health concern, with approximately 80% of the world's population at risk of one or more VBD. Manual disease vector identification is time‐consuming and expert‐dependent, hindering disease control efforts. Deep learning (DL), widely used in image, text, and audio tasks, offers automation potential for disease vector identification. This paper explores the substantial potential of combining DL with disease vector identification. Our aim is to comprehensively summarize the current status of DL in disease vector identification, covering data collection, data preprocessing, model construction, evaluation methods, and applications in identification spanning from species classification to object detection and breeding site identification. We also discuss the challenges and possible prospects for DL in disease vector identification for further research. © 2024 Society of Chemical Industry.
Figures - available from: Pest Management Science
This content is subject to copyright. Terms and conditions apply.
A preview of this full-text is provided by Wiley.
Learn more
Content available from Pest Management Science 
This content is subject to copyright. Terms and conditions apply.
Review
Received: 10 June 2024 Revised: 25 September 2024 Published online in Wiley Online Library: 18 October 2024
(wileyonlinelibrary.com) DOI 10.1002/ps.8473
Deep learning in disease vector
image identification
Shaowen Bai,a,b
†
Liang Shia,c,d,e
†
and Kun Yanga,b
*
Abstract
Vector-borne diseases (VBDs) represent a critical global public health concern, with approximately 80% of the world's popula-
tion at risk of one or more VBD. Manual disease vector identification is time-consuming and expert-dependent, hindering dis-
ease control efforts. Deep learning (DL), widely used in image, text, and audio tasks, offers automation potential for disease
vector identification. This paper explores the substantial potential of combining DL with disease vector identification. Our
aim is to comprehensively summarize the current status of DL in disease vector identification, covering data collection, data
preprocessing, model construction, evaluation methods, and applications in identification spanning from species classification
to object detection and breeding site identification. We also discuss the challenges and possible prospects for DL in disease
vector identification for further research.
© 2024 Society of Chemical Industry.
Supporting information may be found in the online version of this article.
Keywords: disease vector; artificial intelligence; mosquito; snail; convolutional neural network
1 INTRODUCTION
Vector-borne diseases (VBDs) constitute a class of infectious dis-
eases spread by vectors such as mosquitoes, fleas, ticks, and
rodents. These diseases are caused by infectious bacteria, viruses,
or parasites.
1,2
Notable among them are malaria, dengue fever,
chikungunya, yellow fever, Zika virus disease, lymphatic filariasis,
schistosomiasis, onchocerciasis, Chagas disease, leishmaniasis,
and Japanese encephalitis, identified by the World Health Organi-
zation as major VBDs globally.
1,3
VBDs represent a substantial por-
tion of the global burden of infectious diseases, accounting for
17% of total cases and contributing to over 700 000 deaths annu-
ally, thus placing them as a significant public health concern
worldwide.
4–7
Table S1 lists common VBDs and their vectors.
While malaria remains the predominant cause of morbidity and
mortality among VBDs, diseases such as dengue fever, chikungu-
nya, schistosomiasis, and leishmaniasis persistently maintain high
global incidence rates, imposing substantial public health bur-
dens due to insufficient attention and investment.
2,7–10
In recent
years, the management of VBDs has encountered major chal-
lenges stemming from shifts in ecological environments, climate
patterns, population dynamics, globalization, urbanization, and
other factors,
1,11–14
with the added risk of resurgence in previ-
ously controlled VBDs.
15,16
Research indicates that controlling
the respective vectors has been proven to be highly effective in
preventing and controlling VBDs, and in some cases it is consid-
ered to be the sole means of prevention.
2,3,17–19
Deep learning (DL) is a crucial branch of machine learning,
representing an advanced class of machine learning tech-
niques.
20,21
In 2012, Hinton and his colleagues employed convolu-
tional neural networks for the purpose of image identification.
22
Through their participation in the ImageNet evaluation, they suc-
cessfully achieved a significant reduction in the classification error
rate to 15.3%.
22
Since then, DL has rapidly advanced in the field of
image identification. Compared to traditional machine learning
techniques, DL possesses the inherent ability to autonomously
extract features and discern patterns without necessitating
human intervention. With the aid of extensive datasets and
high-performance computing, DL models can attain elevated
levels of accuracy and generalization.
23
In recent years, DL has
ignited a substantial surge in global research endeavors and
has found widespread application across diverse domains.
21,24,25
In the medical field, DL can be involved in the diagnosis, treat-
ment, and prognosis of diseases, prevention, and basic research
into related diseases.
5,26–32
*Correspondence to: Y Kun, Key Laboratory of National Health and Family Plan-
ning Commission on Parasitic Disease Control and Prevention, Jiangsu Provin-
cial Key Laboratory on Parasite and Vector Control Technology, Jiangsu
Institute of Parasitic Diseases, Wuxi 214064, China. E-mail: yangkun@jipd.com
†
Shaowen Bai and Liang Shi should be considered joint first author.
aKey Laboratory of National Health and Family Planning Commission on Para-
sitic Disease Control and Prevention, Jiangsu Provincial Key Laboratory on Par-
asite and Vector Control Technology, Jiangsu Institute of Parasitic Diseases,
Wuxi, China
bSchool of Public Health, Nanjing Medical University, Nanjing, China
cFudan University School of Public Health, Shanghai, China
dKey Laboratory of Public Health Safety, Fudan University, Ministry of Educa-
tion, Shanghai, China
eFudan University Center for Tropical Disease Research, Shanghai, China
Pest Manag Sci 2025; 81: 527–539 www.soci.org © 2024 Society of Chemical Industry.
527
ResearchGate has not been able to resolve any citations for this publication.
A review of convolutional neural networks in computer vision
Article
Full-text available
  • Mar 2024
  • ARTIF INTELL REV
In computer vision, a series of exemplary advances have been made in several areas involving image classification, semantic segmentation, object detection, and image super-resolution reconstruction with the rapid development of deep convolutional neural network (CNN). The CNN has superior features for autonomous learning and expression, and feature extraction from original input data can be realized by means of training CNN models that match practical applications. Due to the rapid progress in deep learning technology, the structure of CNN is becoming more and more complex and diverse. Consequently, it gradually replaces the traditional machine learning methods. This paper presents an elementary understanding of CNN components and their functions, including input layers, convolution layers, pooling layers, activation functions, batch normalization, dropout, fully connected layers, and output layers. On this basis, this paper gives a comprehensive overview of the past and current research status of the applications of CNN models in computer vision fields, e.g., image classification, object detection, and video prediction. In addition, we summarize the challenges and solutions of the deep CNN, and future research directions are also discussed.
View
YOLO-based Object Detection Models: A Review and its Applications
Article
Full-text available
In computer vision, object detection is the classical and most challenging problem to get accurate results in detecting objects. With the significant advancement of deep learning techniques over the past decades, most researchers work on enhancing object detection, segmentation and classification. Object detection performance is measured in both detection accuracy and inference time. The detection accuracy in two stage detectors is better than single stage detectors. In 2015, the real-time object detection system YOLO was published, and it rapidly grew its iterations, with the newest release, YOLOv8 in January 2023. The YOLO achieves a high detection accuracy and inference time with single stage detector. Many applications easily adopt YOLO versions due to their high inference speed. This paper presents a complete survey of YOLO versions up to YOLOv8. This article begins with explained about the performance metrics used in object detection, post-processing methods, dataset availability and object detection techniques that are used mostly; then discusses the architectural design of each YOLO version. Finally, the diverse range of YOLO versions was discussed by highlighting their contributions to various applications.
View
Green cities and vector-borne diseases: emerging concerns and opportunities
Article
Full-text available
  • Mar 2024
  • Euro Surveill
Aligned with the Sustainable Development Goals, nature-based solutions such as urban greening e.g. public gardens, urban forests, parks and street trees, which aim to protect, sustainably manage or restore an ecosystem, have emerged as a promising tool for improving the health and well-being of an ever-increasing urban population. While urban greening efforts have undeniable benefits for human health and the biological communities inhabiting these green zones, disease vector populations may also be affected, possibly promoting greater pathogen transmission and the emergence of infectious diseases such as dengue, West Nile fever, malaria, leishmaniosis and tick-borne diseases. Evidence for the impact of urban green areas on vector-borne disease (VBD) transmission is scarce. Furthermore, because of vast disparities between cities, variation in green landscapes and differing scales of observation, findings are often contradictory; this calls for careful assessment of how urban greening affects VBD risk. Improved understanding of the effect of urban greening on VBDs would support planning, monitoring and management of green spaces in cities to sustainably mitigate VBD risks for surrounding urban populations.
View
A smartphone-based crowd-sourced real-time surveillance platform (apple snail inspector) for the invasive snails: a design and development study
Article
Full-text available
  • Feb 2024
Background The large amphibious freshwater apple snail is an important invasive species in China, but there is currently no method available for their surveillance. The development and popularization of smartphones provide a new platform for research on surveillance technologies for the early detection and effective control of invasive species. Methods The ASI surveillance system was developed based on the infrastructure of the WeChat platform and Amap. The user can directly enter the game interface through the WeChat port on their mobile phone, and the system automatically obtains their location. The user can then report the location of apple snails. The administrator can audit the reported information, and all information can be exported to Microsoft Excel version 2016 for analysis. The map was generated by ArcGIS 10.2 and was used to characterize the spatial and temporal distribution of apple snails in Jiangsu Province. Results The architecture of ASI consists of three parts: a mobile terminal, a server terminal and a desktop terminal. We published more than 10 tweets on the official WeChat account of the system to announce it to the public, and a total of 207 users in 2020 and 2021 correctly reported sightings of apple snails. We identified 550 apple snails breeding sites in 2020 and 2021, featuring ponds (81%), parks (17%) and farmland (2%). In addition, most of the locations contained snail eggs, and the reporting times mainly occurred between May and September. Conclusions The ASI is an effective surveillance system that can be used to identify the breeding locations of apple snails and provides the basis of prevention and control for its dispersal. Its successful development and operation provide new potential avenues for surveillance of other public health issues. Graphical Abstract
View
A review of small object detection based on deep learning
Article
Full-text available
  • Feb 2024
  • NEURAL COMPUT APPL
Small object detection is widely used in a variety of fields such as automatic driving, UAV-based object detection, and aerial image detection. However, small objects carry limited information, making it difficult for detectors to detect small objects. In recent years, the development of deep learning has significantly improved the performance of small object detection. This paper provides a comprehensive review to help further the development of small target detection. We summarize the challenges related to small object detection and analyze solutions to these challenges in existing works, including integrating the feature at different layers, enriching available information, balancing the number of positive and negative samples for small objects, and increasing sufficient small object instances. We discuss related methods developed in three application areas, including automatic driving, UAV search and rescue, and aerial image detection. In addition, we thoroughly analyze the performance of typical small object detection methods on popular datasets. Finally, based on the comprehensive review of small object detection methods, we point out possible research directions for future studies.
View
Control, elimination, and eradication efforts for neglected tropical diseases in the World Health Organization African region over the last 30 years – A scoping review
Article
Full-text available
  • Jan 2024
  • INT J INFECT DIS
Objectives NTDs historically receive less attention than other diseases in the same regions. Recent gap analyses revealed notable shortcomings despite NTD elimination progress. This systematic scoping review was conducted to understand NTD control, elimination, and eradication efforts in the WHO African region over the last 30 years. Methods Peer-reviewed publications from PubMed, Web of Science, and Cochrane databases related to NTD control, elimination, and eradication in the WHO African Region from 1990 to 2022 were reviewed. Included articles were categorized based on NTD; study location, type, and period; and topic areas. Technical and guidance documents from WHO, UN, partner, and academic/research institutions were reviewed. Country-specific multi-year NTD master plans were documented. Results Four hundred eighty peer-reviewed articles, six Cochrane reviews, and 134 technical reports were included. MDA and non-interventional/survey-related studies were common topics. Lymphatic filariasis, trachoma, schistosomiasis, and onchocerciasis were the most frequently studied NTDs. Tanzania, Ethiopia, and Nigeria were the most represented countries; multi-country studies were limited. Conclusion The review highlights progress made in NTD control, elimination, and eradication efforts in the WHO African Region and can inform national/regional strategies. Disease and geographical disparities were evident, warranting focus and research in certain countries. A standardized approach to NTD control programs is needed for sustained progress. Funding There was no funding source for this study.
View
GeoVin - a project with technological innovation and community participation for the comprehensive approach to Chagas disease and triatomines (Hemiptera: Reduviidae)
Article
Full-text available
  • Dec 2023
Chagas disease is one of the most important public health problems in America, involving triatomines (Hemiptera: Reduviidae: Triatominae) in the vectorial transmission of Trypanosoma cruzi (Protista: Kinetoplastea: Trypanosomatidae). In Argentina, 15 species of triatomines are distributed, of which seven are the most common in the domicile and peridomicile, while the rest are mainly associated with wild environments. Due to the lack of updated data of geographic distribution, the GeoVin Project arises to provide not only a triatomine digital tool of triatomine notification, but also resources with information about Chagas and triatomine vectors. The aim of this paper is to present, describe and highlight the progress of the project, emphasizing the principles of participatory science and entomological surveillance. So far, more than 2,300 web visits have been recorded and the application has been downloaded over 3,400 times. The contribution of the project in articulating with the corresponding government health agencies and the immediacy of the interaction with the community are key elements for the project success. At the same time, the information collected is a constant input for updating the geographical distributions of Argentinian triatomines.
View
From Beasts to Bytes: Revolutionizing Zoological Research with Artificial Intelligence
Article
Full-text available
  • Nov 2023
Since the late 2010s, Artificial Intelligence (AI) including machine learning, boosted through deep learning, has boomed as a vital tool to leverage computer vision, natural language processing and speech recognition in revolutionizing zoological research. This review provides an overview of the primary tasks, core models, datasets, and applications of AI in zoological research, including animal classification, resource conservation, behavior, development, genetics and evolution, breeding and health, disease models, and paleontology. Additionally, we explore the challenges and future directions of integrating AI into this field. Based on numerous case studies, this review outlines various avenues for incorporating AI into zoological research and underscores its potential to enhance our understanding of the intricate relationships that exist within the animal kingdom. As we build a bridge between beast and byte realms, this review serves as a resource for envisioning novel AI applications in zoological research that have not yet been explored.
View
View
Effects of climate change and human activities on vector-borne diseases
Article
  • Mar 2024
  • NAT REV MICROBIOL
Vector-borne diseases are transmitted by haematophagous arthropods (for example, mosquitoes, ticks and sandflies) to humans and wild and domestic animals, with the largest burden on global public health disproportionately affecting people in tropical and subtropical areas. Because vectors are ectothermic, climate and weather alterations (for example, temperature, rainfall and humidity) can affect their reproduction, survival, geographic distribution and, consequently, ability to transmit pathogens. However, the effects of climate change on vector-borne diseases can be multifaceted and complex, sometimes with ambiguous consequences. In this Review, we discuss the potential effects of climate change, weather and other anthropogenic factors, including land use, human mobility and behaviour, as possible contributors to the redistribution of vectors and spread of vector-borne diseases worldwide.
View