Multi-View Pareto Optimization for Minimal-Diagnostic-Set Identification of Disease Vectors
by
,
Nuofei Lin
1,†,
Jingjing Wang
2,†,
Yixiang Qian
1,
Li Wei
1,
Hongxia Liu
2,*,
Bo Dai
1,*,
Songlin Zhuang
1 and
Dawei Zhang
1 1
Engineering Research Center of Optical Instrument and System, The Ministry of Education, Shanghai Key Laboratory of Modern Optical System, University of Shanghai for Science and Technology, Shanghai 200093, China
2
Department of Infectious Disease Control, Shanghai Municipal Center for Disease Control and Prevention, Shanghai 201107, China
*
Authors to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Insects 2026, 17(4), 381; https://doi.org/10.3390/insects17040381
Submission received: 28 January 2026
/
Revised: 21 March 2026
/
Accepted: 27 March 2026
/
Published: 1 April 2026
(This article belongs to the Section Insect Pest and Vector Management)
Simple Summary
Accurate identification of disease vectors such as mosquitoes and flies is essential for preventing vector-borne disease outbreaks. However, distinguishing morphologically similar species is challenging and typically requires expert examination of fine diagnostic traits. Existing AI approaches often depend on large, multi-perspective image datasets, which are resource-intensive to acquire. To address this, we developed MVP-Net, an intelligent system that maintains high accuracy with limited data by learning a set of key anatomical views. Using regionally collected fly and mosquito datasets from routine surveillance in Shanghai, the model retained comparable classification performance using only 5 and 2 views for flies and mosquitoes, respectively. This approach reduces image acquisition effort and computational cost and may support regional auxiliary identification workflows.
Abstract
Accurate identification of disease vectors is crucial for public health, yet distinguishing morphologically similar species demands significant taxonomic expertise and data resources. This study proposes MVP-Net, an AI-driven framework designed to extract a minimal sufficient set of diagnostic anatomical views from multi-view imagery for efficient identification. The framework was evaluated on regionally collected datasets of Calyptratae (8 views) and Culicidae (11 views) from routine surveillance in Shanghai. Under all-view fusion, MVP-Net achieved Top-1 accuracies of 87.04% for Calyptratae and 100% for Culicidae. After Pareto-based view optimization, the required input was reduced to 5 views for Calyptratae and 2 views for Culicidae, lowering computational cost by 37.49% and 81.82%, respectively, while retaining comparable classification performance (86.11% for the recommended Calyptratae configuration and 100% for the recommended Culicidae configuration). These results show that MVP-Net can reduce view redundancy while preserving comparable identification performance within the current Shanghai surveillance setting, providing a practical approach for optimizing regional multi-view auxiliary identification workflows.
