Your browser doesn't support javascript.
COVID-19 mortality prediction in the intensive care unit with deep learning based on longitudinal chest X-rays and clinical data.
Cheng, Jianhong; Sollee, John; Hsieh, Celina; Yue, Hailin; Vandal, Nicholas; Shanahan, Justin; Choi, Ji Whae; Tran, Thi My Linh; Halsey, Kasey; Iheanacho, Franklin; Warren, James; Ahmed, Abdullah; Eickhoff, Carsten; Feldman, Michael; Mortani Barbosa, Eduardo; Kamel, Ihab; Lin, Cheng Ting; Yi, Thomas; Healey, Terrance; Zhang, Paul; Wu, Jing; Atalay, Michael; Bai, Harrison X; Jiao, Zhicheng; Wang, Jianxin.
  • Cheng J; Hunan Provincial Key Lab on Bioinformatics, School of Computer Science and Engineering, Central South University, 932 Lushan S Rd, Yuelu District, Changsha, Hunan, China.
  • Sollee J; Department of Diagnostic Radiology, Rhode Island Hospital, 593 Eddy St., Providence, RI, 02903, USA.
  • Hsieh C; Warren Alpert Medical School of Brown University, Providence, RI, 02903, USA.
  • Yue H; Department of Diagnostic Radiology, Rhode Island Hospital, 593 Eddy St., Providence, RI, 02903, USA.
  • Vandal N; Warren Alpert Medical School of Brown University, Providence, RI, 02903, USA.
  • Shanahan J; Hunan Provincial Key Lab on Bioinformatics, School of Computer Science and Engineering, Central South University, 932 Lushan S Rd, Yuelu District, Changsha, Hunan, China.
  • Choi JW; Department of Diagnostic Radiology, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, 19104, USA.
  • Tran TML; Department of Diagnostic Radiology, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, 19104, USA.
  • Halsey K; Department of Diagnostic Radiology, Rhode Island Hospital, 593 Eddy St., Providence, RI, 02903, USA.
  • Iheanacho F; Warren Alpert Medical School of Brown University, Providence, RI, 02903, USA.
  • Warren J; Department of Diagnostic Radiology, Rhode Island Hospital, 593 Eddy St., Providence, RI, 02903, USA.
  • Ahmed A; Warren Alpert Medical School of Brown University, Providence, RI, 02903, USA.
  • Eickhoff C; Department of Diagnostic Radiology, Rhode Island Hospital, 593 Eddy St., Providence, RI, 02903, USA.
  • Feldman M; Warren Alpert Medical School of Brown University, Providence, RI, 02903, USA.
  • Mortani Barbosa E; Department of Diagnostic Radiology, Rhode Island Hospital, 593 Eddy St., Providence, RI, 02903, USA.
  • Kamel I; Warren Alpert Medical School of Brown University, Providence, RI, 02903, USA.
  • Lin CT; Department of Data Science, University of London, London, UK.
  • Yi T; Department of Diagnostic Radiology, Rhode Island Hospital, 593 Eddy St., Providence, RI, 02903, USA.
  • Healey T; Warren Alpert Medical School of Brown University, Providence, RI, 02903, USA.
  • Zhang P; Center for Biomedical Informatics, Brown University, Providence, RI, 02912, USA.
  • Wu J; Department of Pathology and Laboratory Medicine, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, 19104, USA.
  • Atalay M; Department of Diagnostic Radiology, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, 19104, USA.
  • Bai HX; Department of Radiology and Radiological Sciences, Johns Hopkins University School of Medicine, 601 N Caroline St, Baltimore, MD, 21205, USA.
  • Jiao Z; Department of Radiology and Radiological Sciences, Johns Hopkins University School of Medicine, 601 N Caroline St, Baltimore, MD, 21205, USA.
  • Wang J; Department of Diagnostic Radiology, Rhode Island Hospital, 593 Eddy St., Providence, RI, 02903, USA.
Eur Radiol ; 32(7): 4446-4456, 2022 Jul.
Article in English | MEDLINE | ID: covidwho-1707890
ABSTRACT

OBJECTIVES:

We aimed to develop deep learning models using longitudinal chest X-rays (CXRs) and clinical data to predict in-hospital mortality of COVID-19 patients in the intensive care unit (ICU).

METHODS:

Six hundred fifty-four patients (212 deceased, 442 alive, 5645 total CXRs) were identified across two institutions. Imaging and clinical data from one institution were used to train five longitudinal transformer-based networks applying five-fold cross-validation. The models were tested on data from the other institution, and pairwise comparisons were used to determine the best-performing models.

RESULTS:

A higher proportion of deceased patients had elevated white blood cell count, decreased absolute lymphocyte count, elevated creatine concentration, and incidence of cardiovascular and chronic kidney disease. A model based on pre-ICU CXRs achieved an AUC of 0.632 and an accuracy of 0.593, and a model based on ICU CXRs achieved an AUC of 0.697 and an accuracy of 0.657. A model based on all longitudinal CXRs (both pre-ICU and ICU) achieved an AUC of 0.702 and an accuracy of 0.694. A model based on clinical data alone achieved an AUC of 0.653 and an accuracy of 0.657. The addition of longitudinal imaging to clinical data in a combined model significantly improved performance, reaching an AUC of 0.727 (p = 0.039) and an accuracy of 0.732.

CONCLUSIONS:

The addition of longitudinal CXRs to clinical data significantly improves mortality prediction with deep learning for COVID-19 patients in the ICU. KEY POINTS • Deep learning was used to predict mortality in COVID-19 ICU patients. • Serial radiographs and clinical data were used. • The models could inform clinical decision-making and resource allocation.
Subject(s)
Keywords

Full text: Available Collection: International databases Database: MEDLINE Main subject: Deep Learning / COVID-19 Type of study: Diagnostic study / Observational study / Prognostic study / Randomized controlled trials / Risk factors Limits: Humans Language: English Journal: Eur Radiol Journal subject: Radiology Year: 2022 Document Type: Article Affiliation country: S00330-022-08588-8

Similar

MEDLINE

...
LILACS

LIS


Full text: Available Collection: International databases Database: MEDLINE Main subject: Deep Learning / COVID-19 Type of study: Diagnostic study / Observational study / Prognostic study / Randomized controlled trials / Risk factors Limits: Humans Language: English Journal: Eur Radiol Journal subject: Radiology Year: 2022 Document Type: Article Affiliation country: S00330-022-08588-8