Notice

Revolutionizing Healthcare: Smartphone AI Detects Myocardial Infarction in Minutes

  • admin
  • 2024-03-13
  • 154

The Research team of Professor Jeong-hoon Lee's team has developed the world's fastest diagnostic/predictive technology using deep learning AI on smartphones: Detecting myocardial infarction within 1-2 minutes

 

- Introducing the world’s fastest diagnostic/predictive technology using deep learning technology -

- Enabling rapid response to emergency situations such as myocardial infarction even in the ambulance -

- Presenting a key technology that can dramatically reduce waiting times at hospitals -

- Proposing a paradigm shift in the management of new/variant infectious diseases and chronic diseases using smartphones -

- Published in Nature Communications (IF: 17.69) -

 

The research team of Professor Jeong-hoon Lee (Department of Electrical Engineering) has successfully developed a deep learning algorithm for early diagnosis (TIMESAVER: Time-Efficient Immunoassay with Smart AI-based Verification). The team achieved expert-level accuracy within 1-2 minutes for on-site diagnosis using commercial rapid kits.

 

The research was supported by the Bio-Medical Technology Development Project of the Korea Research Foundation (No. 2023M3E5E3080743) and published in Nature Communications, a top-tier journal by the Nature Portfolio (IF: 17.69).

 

Journal: Nature Communications

DOI: 10.1038/s41467-024-46069-2

Title: Rapid deep learning-assisted predictive diagnostics for point-of-care testing

 

이정훈교수, 이승민, 박정수, 우효원 사진자료

From Left to right: Chief researcher, Professor Jeong-hoon Lee (Kwangwoon University), core researchers, Seung-min Lee (co-first author: Ph.D. candidate), Jung-soo Park (co-first author: Ph.D. candidate), Hyo-won Woo (co-first author: Master's candidate)

<그림설명1> 시계열 예측 딥러닝 (TIMESAVER)를 활용한 진단 시스템의 개요. 기존의 진단 방식이 길게는 수시간의 긴 분석시간을 소요하나, 본 시계열 딥러닝 알고리즘을 적용할 경우 진단/스크리닝 시간을 1~2분 내로 대폭 단축이 가능함.

Figure 1: Overview of the diagnostic system using time series prediction deep learning (TIMESAVER). While conventional diagnostic methods take several hours of lengthy analysis time, applying this time series deep learning algorithm can significantly reduce the diagnosis/screening time to 1-2 minutes.

 

<그림설명2> 본 기술을 통해 크게 3가지 효과 창출이 가능함. 첫째 고민감도/고속진단을 통해 응급진단. 둘째, 대기없는 진단. 셋째, 현장/응급실 뿐만 아니라 홈케어에 있어서 패러다임 혁신. 
(이미지 출저: iStock, Vecteezy)

 

Figure 2: Through this technology, three significant effects can be achieved: First, emergency diagnosis through high sensitivity/high-speed diagnosis. Second, diagnosis without waiting. Third, paradigm shift in home care as well as in field/emergency rooms. (Image source: iStock, Vecteezy)