한빛사논문
Seungmin Lee1,2,9, Sunmok Kim1,9, Dae Sung Yoon2,3,4,9, Jeong Soo Park1, Hyowon Woo1, Dongho Lee5, Sung-Yeon Cho6,7, Chulmin Park6, Yong Kyoung Yoo8 , Ki-Baek Lee1 & Jeong
Hoon Lee1
1Department of Electrical Engineering, Kwangwoon University, 20 Kwangwoon-ro, Nowon, Seoul 01897, Republic of Korea.
2School of Biomedical Engineering, Korea University, 145 Anam-ro, Seongbuk, Seoul 02841, Republic of Korea.
3Interdisciplinary Program in Precision Public Health, Korea University, Seoul 02841, Republic of Korea.
4Astrion Inc, Seoul 02841, Republic of Korea.
5CALTH Inc., Changeop-ro 54, Seongnam, Gyeonggi 13449, Republic of Korea.
6Vaccine Bio Research Institute, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea.
7Division of Infectious Diseases, Department of Internal Medicine, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea.
8Department of Electronic Engineering, Catholic Kwandong University, 24, Beomil-ro 579 beon-gil, Gangneung-si, Gangwon-do 25601, Republic of Korea.
9These authors contributed equally: Seungmin Lee, Sunmok Kim, Dae Sung Yoon.
Corresponding authors : Correspondence to Yong Kyoung Yoo , Ki- Baek Lee or Jeong Hoon Lee.
Abstract
Since many lateral flow assays (LFA) are tested daily, the improvement in accuracy can greatly impact individual patient care and public health. However, current self-testing for COVID-19 detection suffers from low accuracy, mainly due to the LFA sensitivity and reading ambiguities. Here, we present deep learning-assisted smartphone-based LFA (SMARTAI-LFA) diagnostics to provide accurate decisions with higher sensitivity. Combining clinical data learning and two-step algorithms enables a cradle-free on-site assay with higher accuracy than the untrained individuals and human experts via blind tests of clinical data (n = 1500). We acquired 98% accuracy across 135 smartphone application-based clinical tests with different users/smartphones. Furthermore, with more low-titer tests, we observed that the accuracy of SMARTAI-LFA was maintained at over 99% while there was a significant decrease in human accuracy, indicating the reliable performance of SMARTAI-LFA. We envision a smartphone-based SMARTAI-LFA that allows continuously enhanced performance by adding clinical tests and satisfies the new criterion for digitalized real-time diagnostics.
논문정보
관련 링크
연구자 ID
관련분야 연구자보기
관련분야 논문보기