High accurate and explainable multi-pill detection framework with graph neural network-assisted multimodal data fusion
PLOS ONE · 28.09.2023 · 18 (9) · e0291865
DOI: 10.1371/journal.pone.0291865 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS ONE, online erschienen am 28.09.2023. Verfasst von Anh Duy Nguyen und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Information and Communication Technology, Hanoi University of Science and Technology, Hanoi, Vietnam. Hauptthemen laut MEDLINE: Benchmarking, Environment, Controlled. Die Arbeit wurde bisher 23-mal zitiert (Crossref) und nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Vingroup Innovation Foundation, Vingroup Joint Stock Company.
Abstract (Originalsprache)
Due to the significant resemblance in visual appearance, pill misuse is prevalent and has become a critical issue, responsible for one-third of all deaths worldwide. Pill identification, thus, is a crucial concern that needs to be investigated thoroughly. Recently, several attempts have been made to exploit deep learning to tackle the pill identification problem. However, most published works consider only single-pill identification and fail to distinguish hard samples with identical appearances. Also, most existing pill image datasets only feature single pill images captured in carefully controlled environments under ideal lighting conditions and clean backgrounds. In this work, we are the first to tackle the multi-pill detection problem in real-world settings, aiming at localizing and identifying pills captured by users during pill intake. Moreover, we also introduce a multi-pill image dataset taken in unconstrained conditions. To handle hard samples, we propose a novel method for constructing heterogeneous a priori graphs incorporating three forms of inter-pill relationships, including co-occurrence likelihood, relative size, and visual semantic correlation. We then offer a framework for integrating a priori with pills’ visual features to enhance detection accuracy. Our experimental results have proved the robustness, reliability, and explainability of the proposed framework. Experimentally, it outperforms all detection benchmarks in terms of all evaluation metrics. Specifically, our proposed framework improves COCO mAP metrics by 9.4% over Faster R-CNN and 12.0% compared to vanilla YOLOv5. Our study opens up new opportunities for protecting patients from medication errors using an AI-based pill identification solution.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- School of Information and Communication Technology, Hanoi University of Science and Technology, Hanoi, Vietnam.
- VinUni-Illinois Smart Health Center, VinUniversity, Hanoi, Vietnam.
- College of Engineering & Computer Science, VinUniversity, Hanoi, Vietnam.
- École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland.
- Griffith University, Southport, Queensland, Australia.
- National Institute of Advanced Industrial Science and Technology, Tokyo, Japan.
ORCID-Profile: Anh Duy Nguyen, Huynh Thanh Trung, Quoc Viet Hung Nguyen, Phi Le Nguyen
Förderung
- Vingroup Innovation Foundation
- Vingroup Joint Stock Company
Zitieren
APA
Nguyen, A. D., Pham, H. H., Trung, H. T., Nguyen, Q. V. H., Truong, T. N., & Nguyen, P. L. (2023). High accurate and explainable multi-pill detection framework with graph neural network-assisted multimodal data fusion. PLOS ONE, 18(9), e0291865. https://doi.org/10.1371/journal.pone.0291865
Vancouver
Nguyen AD, Pham HH, Trung HT, Nguyen QVH, Truong TN, Nguyen PL. High accurate and explainable multi-pill detection framework with graph neural network-assisted multimodal data fusion. PLOS ONE. 2023;18(9):e0291865. doi:10.1371/journal.pone.0291865
BibTeX
@article{Nguyen2023,
author = {Nguyen, Anh Duy and Pham, Huy Hieu and Trung, Huynh Thanh and Nguyen, Quoc Viet Hung and Truong, Thao Nguyen and Nguyen, Phi Le},
title = {High accurate and explainable multi-pill detection framework with graph neural network-assisted multimodal data fusion},
journal = {PLOS ONE},
year = {2023},
volume = {18},
number = {9},
pages = {e0291865},
doi = {10.1371/journal.pone.0291865}
}
EUVIMED