Your browser doesn't support javascript.
Deep learning with robustness to missing data: A novel approach to the detection of COVID-19.
Çalli, Erdi; Murphy, Keelin; Kurstjens, Steef; Samson, Tijs; Herpers, Robert; Smits, Henk; Rutten, Matthieu; van Ginneken, Bram.
  • Çalli E; Diagnostic Image Analysis Group, Radboudumc, Nijmegen, The Netherlands.
  • Murphy K; Diagnostic Image Analysis Group, Radboudumc, Nijmegen, The Netherlands.
  • Kurstjens S; Laboratory for Clinical Chemistry and Hematology, Jeroen Bosch Hospital, 's-Hertogenbosch, The Netherlands.
  • Samson T; Laboratory for Clinical Chemistry and Hematology, Jeroen Bosch Hospital, 's-Hertogenbosch, The Netherlands.
  • Herpers R; Laboratory for Clinical Chemistry and Hematology, Bernhoven Hospital, Uden, The Netherlands.
  • Smits H; Laboratory for Clinical Chemistry and Hematology, Bernhoven Hospital, Uden, The Netherlands.
  • Rutten M; Diagnostic Image Analysis Group, Radboudumc, Nijmegen, The Netherlands.
  • van Ginneken B; Department of Radiology, Jeroen Bosch Hospital, 's-Hertogenbosch, The Netherlands.
PLoS One ; 16(7): e0255301, 2021.
Article in English | MEDLINE | ID: covidwho-1334776
ABSTRACT
In the context of the current global pandemic and the limitations of the RT-PCR test, we propose a novel deep learning architecture, DFCN (Denoising Fully Connected Network). Since medical facilities around the world differ enormously in what laboratory tests or chest imaging may be available, DFCN is designed to be robust to missing input data. An ablation study extensively evaluates the performance benefits of the DFCN as well as its robustness to missing inputs. Data from 1088 patients with confirmed RT-PCR results are obtained from two independent medical facilities. The data includes results from 27 laboratory tests and a chest x-ray scored by a deep learning model. Training and test datasets are taken from different medical facilities. Data is made publicly available. The performance of DFCN in predicting the RT-PCR result is compared with 3 related architectures as well as a Random Forest baseline. All models are trained with varying levels of masked input data to encourage robustness to missing inputs. Missing data is simulated at test time by masking inputs randomly. DFCN outperforms all other models with statistical significance using random subsets of input data with 2-27 available inputs. When all 28 inputs are available DFCN obtains an AUC of 0.924, higher than any other model. Furthermore, with clinically meaningful subsets of parameters consisting of just 6 and 7 inputs respectively, DFCN achieves higher AUCs than any other model, with values of 0.909 and 0.919.
Subject(s)

Full text: Available Collection: International databases Database: MEDLINE Main subject: Databases, Factual / Deep Learning / COVID-19 Nucleic Acid Testing / SARS-CoV-2 / COVID-19 / Models, Theoretical Type of study: Diagnostic study / Experimental Studies / Prognostic study / Randomized controlled trials Limits: Humans Language: English Journal: PLoS One Journal subject: Science / Medicine Year: 2021 Document Type: Article Affiliation country: Journal.pone.0255301

Similar

MEDLINE

...
LILACS

LIS


Full text: Available Collection: International databases Database: MEDLINE Main subject: Databases, Factual / Deep Learning / COVID-19 Nucleic Acid Testing / SARS-CoV-2 / COVID-19 / Models, Theoretical Type of study: Diagnostic study / Experimental Studies / Prognostic study / Randomized controlled trials Limits: Humans Language: English Journal: PLoS One Journal subject: Science / Medicine Year: 2021 Document Type: Article Affiliation country: Journal.pone.0255301