Single-Shot Lightweight Model For The Detection of Lesions And The Prediction of COVID-19 From Chest CT Scans

12/06/2020
by   Aram Ter-Sarkisov, et al.
1

We introduce a lightweight model based on Mask R-CNN with ResNet18 and ResNet34 backbone models that segments lesions and predicts COVID-19 from chest CT scans in a single shot. The model requires a small dataset to train: 650 images for the segmentation branch and 3000 for the classification branch, and it is evaluated on 21292 images to achieve a 42.45% average precision (main MS COCO criterion) on the segmentation test split (100 images), 93.00% COVID-19 sensitivity and F1-score of 96.76% on the classification test split (21192 images) across 3 classes: COVID-19, Common Pneumonia and Control/Negative. The full source code, models and pretrained weights are available on https://github.com/AlexTS1980/COVID-Single-Shot-Model.

READ FULL TEXT

page 2

page 3

page 4

research
10/30/2020

Detection and Segmentation of Lesion Areas in Chest CT Scans For The Prediction of COVID-19

In this paper we compare the models for the detection and segmentation o...
research
10/30/2020

COVID-CT-Mask-Net: Prediction of COVID-19 from CT Scans Using Regional Features

We present COVID-CT-Mask-Net model that predicts COVID-19 from CT scans....
research
11/09/2020

Explainable COVID-19 Detection Using Chest CT Scans and Deep Learning

This paper explores how well deep learning models trained on chest CT im...
research
05/17/2023

An Ensemble Deep Learning Approach for COVID-19 Severity Prediction Using Chest CT Scans

Chest X-rays have been widely used for COVID-19 screening; however, 3D c...
research
10/06/2022

COVID-19 Detection Using Segmentation, Region Extraction and Classification Pipeline

Purpose The main purpose in this study is to propose a pipeline for COVI...
research
08/21/2020

Comparative performance analysis of the ResNet backbones of Mask RCNN to segment the signs of COVID-19 in chest CT scans

COVID-19 has been detrimental in terms of the number of fatalities and r...
research
01/29/2020

3D Aggregated Faster R-CNN for General Lesion Detection

Lesions are damages and abnormalities in tissues of the human body. Many...

Please sign up or login with your details

Forgot password? Click here to reset