2022
DOI: 10.1007/978-3-031-13324-4_42
|View full text |Cite
|
Sign up to set email alerts
|

Deep Regression by Feature Regularization for COVID-19 Severity Prediction

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1

Citation Types

0
2
0

Year Published

2023
2023
2024
2024

Publication Types

Select...
3

Relationship

0
3

Authors

Journals

citations
Cited by 3 publications
(2 citation statements)
references
References 19 publications
0
2
0
Order By: Relevance
“…The EIDOSlab_Unito team from the University of Turin, Italy [ 39 ] proposed a contrastive learning-based approach for the CIPE. Their approach consists of four main steps: (i) the input slices were preprocessed using image resizing and pixel intensity scaling to reduce the brightness and pixel intensity contrast between the training and test splits; (ii) DenseNet-121 [ 24 ] architecture was used as a feature projector to the 1024-dimensional target space; (iii) a distance loss function was proposed to correlate the distances between the samples in the deep feature space and their GT values in the target space (i.e., the COVID-19 level of infection); and (iv) the COVID-19 percentage estimation of a new image, the query, is computed by averaging the values from the nearest neighbours by means of Euclidean distance in the projection feature space from images in the training set (the reference set).…”
Section: Participating Teamsmentioning
confidence: 99%
“…The EIDOSlab_Unito team from the University of Turin, Italy [ 39 ] proposed a contrastive learning-based approach for the CIPE. Their approach consists of four main steps: (i) the input slices were preprocessed using image resizing and pixel intensity scaling to reduce the brightness and pixel intensity contrast between the training and test splits; (ii) DenseNet-121 [ 24 ] architecture was used as a feature projector to the 1024-dimensional target space; (iii) a distance loss function was proposed to correlate the distances between the samples in the deep feature space and their GT values in the target space (i.e., the COVID-19 level of infection); and (iv) the COVID-19 percentage estimation of a new image, the query, is computed by averaging the values from the nearest neighbours by means of Euclidean distance in the projection feature space from images in the training set (the reference set).…”
Section: Participating Teamsmentioning
confidence: 99%
“… Percentage estimation from CT images The augmentation techniques helped achieving the desired prediction performance. Tricarico, Chaudhry, Fiandrotti, and Grangetto (2022) Proposed feature regularization based deep regression approach for the severity prediction task. Percentage estimation from CT scan.…”
Section: Related Workmentioning
confidence: 99%