본문 바로가기

Development and Validation of a Multimodal-Based Prognosis and Intervention Prediction Model for COVID-19 Patients in a Multicenter Cohort

문서 유형
연구출판물 · 학술지 논문
등재
SCIEScopus
저널 순위
JCR Q2Scopus Q1
저널
SENSORS22(13) · 2022
피인용
Web of Science 8Scopus 8
FWCI (Scopus)
0.36
JCR IF
JCR 2021 3.847
ISSN
1424-3210

저자

  1. Lee, Jeong Hoon제1저자Lunit Inc, 27 Teheran Ro 2 Gil, Seoul 06241, South Korea
  2. Ahn, Jong SeokLunit Inc, 27 Teheran Ro 2 Gil, Seoul 06241, South Korea
  3. Chung, Myung JinSungkyunkwan Univ, Sch Med, Samsung Med Ctr, Dept Radiol, Seoul 06351, South Korea
  4. 정연주Pusan Natl Univ, Sch Med, Pusan Natl Univ Hosp, Dept Radiol, Busan 49241, South Korea
  5. Kim, Jin HwanChungnam Natl Univ, Coll Med, Chungnam Natl Univ Hosp, Dept Radiol, Daejeon 35015, South Korea
  6. Lim, Jae KwangKyungpook Natl Univ, Kyungpook Natl Univ Hosp, Sch Med, Dept Radiol, Daegu 41944, South Korea
  7. Kim, Jin YoungKeimyung Univ, Sch Med, Dongsan Hosp, Dept Radiol, Daegu 42601, South Korea
  8. Kim, Young JaeGachon Univ, Coll Med, Dept Biomed Engn, Incheon 21565, South Korea
  9. Lee, Jong Eun교신저자Chonnam Natl Univ Hosp, Dept Radiol, 42 Jebong Ro, Gwangju 61469, South Korea
  10. Kim, Eun Young교신저자Gachon Univ, Coll Med, Gil Med Ctr, Dept Radiol, Namdong Daero 774 Beon Gil, Incheon 21565, South Korea

초록

© 2022 by the authors. Licensee MDPI, Basel, Switzerland.The ability to accurately predict the prognosis and intervention requirements for treating highly infectious diseases, such as COVID-19, can greatly support the effective management of pa-tients, especially in resource-limited settings. The aim of the study is to develop and validate a mul-timodal artificial intelligence (AI) system using clinical findings, laboratory data and AI-interpreted features of chest X-rays (CXRs), and to predict the prognosis and the required interventions for patients diagnosed with COVID-19, using multi-center data. In total, 2282 real-time reverse tran-scriptase polymerase chain reaction-confirmed COVID-19 patients’ initial clinical findings, laboratory data and CXRs were retrospectively collected from 13 medical centers in South Korea, between January 2020 and June 2021. The prognostic outcomes collected included intensive care unit (ICU) admission and in-hospital mortality. Intervention outcomes included the use of oxygen (O2) sup-plementation, mechanical ventilation and extracorporeal membrane oxygenation (ECMO). A deep learning algorithm detecting 10 common CXR abnormalities (DLAD-10) was used to infer the initial CXR taken. A random forest model with a quantile classifier was used to predict the prognostic and intervention outcomes, using multimodal data. The area under the receiver operating curve (AU-ROC) values for the single-modal model, using clinical findings, laboratory data and the outputs from DLAD-10, were 0.742 (95% confidence interval [CI], 0.696–0.788), 0.794 (0.745–0.843) and 0.770 (0.724–0.815), respectively. The AUROC of the combined model, using clinical findings, laboratory data and DLAD-10 outputs, was significantly higher at 0.854 (0.820–0.889) than that of all other models (p < 0.001, using DeLong’s test). In the order of importance, age, dyspnea, consolidation and fever were significant clinical variables for prediction. The most predictive DLAD-10 output was consolidation. We have shown that a multimodal AI model can improve the performance of predicting both the prognosis and intervention in COVID-19 patients, and this could assist in effective treatment and subsequent resource management. Further, image feature extraction using an estab-lished AI engine with well-defined clinical outputs, and combining them with different modes of clinical data, could be a useful way of creating an understandable multimodal prediction model.

키워드

artificial intelligencechest radiographCOVID-19prognosisCHEST RADIOGRAPHSSYSTEM

식별자

Scopus EID
2-s2.0-85133143544
WoS UID
000822123000001