Machine learning-based CT radiomics model for predicting hospital stay in patients with pneumonia associated with SARS-CoV-2 infection: A multicenter study

Abstract

medRxiv preprint 6 datasets. The predictive performance was further evaluated in test dataset on lung lobe-and patients-level. Main outcomesShort-term hospital stay (≤10 days) and long-term hospital stay (>10 days). ResultsThe CT radiomics models based on 6 second-order features were effective in discriminating short-and long-term hospital stay in patients with pneumonia associated with SARS-CoV-2 infection, with areas under the curves of 0.97 (95%CI 0.83-1.0) and 0.92 (95%CI 0.67-1.0) by LR and RF, respectively, in the test dataset. The LR model showed a sensitivity and specificity of 1.0 and 0.89, and the RF model showed similar performance with sensitivity and specificity of 0.75 and 1.0 in test dataset. ConclusionsThe machine learning-based CT radiomics models showed feasibility and accuracy for predicting hospital stay in patients with pneumonia associated with SARS-CoV-2 infection.All rights reserved. No reuse allowed without permission.author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Results Patient characteristicsA total of 52 patients with laboratory-confirmed SARS-CoV-2 infection and initial CT images were enrolled from 5 designated hospitals in Ankang, Lishui, Zhenjiang, Lanzhou, and Linxia, China. As of February 20, 14 patients were still hospitalized, and 7 patients had non-findings in CT images. Therefore, 31 patients with 72 lesion segments were included in the final analysis. The training and inter-validation cohort comprised 26 patients (12 from Ankang, 8 from Lishui, 4 from Lanzhou, and 2 from Linxia) with 59 lesion segments, and test cohort comprised 5 patients from Zhenjiang with 13 lesion segments. The median age was 38.00 (interquartile range, 26.00-47.00) years and 17 (57%) were male. Comorbidities, symptoms and laboratory findings at admission were summarized in Table 1. Performance of CT radiomics modelThe CT radiomics model, based on 6 features (supplementary Table1), showed the highest AUC on the training and inter-validation dataset. The performance of modeling using LR and RF methods was shown in Figure 2. On lung lobe-level, models using LR method significantly distinguished short-and long-term hospital stay (In training and inter-validation datasets, cut-off value 0.31, AUC 0.94 (95%CI 0.92-0.97), sensitivity 1.0, specificity 0.87, NPV 1.0, and PPV 0.88; In test dataset, AUC 0.97 (95%CI 0.83-1.0), sensitivity 1.0, specificity 0.89, NPV 1.0, and PPV 0.8). Besides, models using RF method obtained satisfied results (In training and inter-validation datasets, cut-off value 0.68, AUC 1.0 (95%CI 1.0-1.0), All rights reserved. No reuse allowed without permission.author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

Review of Artificial Intelligence Techniques in Imaging Data Acquisition, Segmentation, and Diagnosis for COVID-19

“…In the literature, there have been numerous techniques for lung segmentation with different purposes [63][64][65][66][67]. The U-Net is a commonly used technique for segmenting both lung regions and lung lesions in COVID applications [50][51][52][53]. The U-Net, a type of fully convolutional network proposed by Ronneberger [68], has a U-shape architecture with symmetric encoding and decoding signal paths.…”

“…For example, Shan et al [58] integrate human-in-the-loop strategy into the training of a VB-net based segmentation network, which involves interactivity with radiologists into the training of the network. Qi et al [53] delineate the lesions in the lung using U-Net with the initial seeds given by a radiologist. Several other works used diagnostic knowledge and identified the infection regions by the attention mechanism [57].…”

“…To segment ROIs in CT, deep learning methods are widely used. The popular segmentation networks for COVID-19 include classic U-Net [50][51][52][53][54][55], UNet++ [56,57], VB-Net [58]. Compared with CT, X-ray is more easily accessible around the world.…”

Systematic review and critical appraisal of prediction models for diagnosis and prognosis of COVID-19 infection

“…Bai, Fang, et al 9 low unclear unclear high 38 Caramelo, Ferreira, et al 18 high high high high Gong, Ou, et al 32 low unclear unclear high Lu, Hu, et al 19 low low low high Qi, Jiang, et al 20 unclear low low high Shi, Yu, et al 37 high high high high Xie, Hungerford, et al 7 low low low high Yan, Zhang, et al 21 low high low high Yuan, Yin, et al 22 low high low high *1 Risk of bias high due to not evaluating calibration. If this criterion is not taken into account, analysis risk of bias would have been unclear.…”

Development and External Validation of a Prognostic Tool for COVID-19 Critical Disease

“…[7][8][9][10] Several attempts have been made to develop prognostic models for COVID-19 disease, largely based on early data from patient cohorts in China. [11][12][13][14][15][16][17][18] These models have used demographic features, including age, sex, and comorbidities, and a limited set of laboratory values including lymphocyte count, lactate dehydrogenase (LDH), C-reactive protein (CRP), which have been reported to be associated with more severe disease. 19,20 These initial models are of variable quality, with a high likelihood of biases and limited numbers of variables, and performance evaluation is limited by suboptimal reporting and limited validation.…”