[Paper Review] Harmony-Search and Otsu based System for Coronavirus Disease (COVID-19) Detection using Lung CT Scan Images

V. Rajinikanth, Nilanjan Dey+4 more|arXiv (Cornell University)|Apr 6, 2020

COVID-19 diagnosis using AI30 references111 citations

TL;DR

The paper presents an image-assisted system that uses Harmony-Search optimization and Otsu thresholding to detect and assess COVID-19 infection in lung CT scans through ROI extraction and severity features.

ABSTRACT

Pneumonia is one of the foremost lung diseases and untreated pneumonia will lead to serious threats for all age groups. The proposed work aims to extract and evaluate the Coronavirus disease (COVID-19) caused pneumonia infection in lung using CT scans. We propose an image-assisted system to extract COVID-19 infected sections from lung CT scans (coronal view). It includes following steps: (i) Threshold filter to extract the lung region by eliminating possible artifacts; (ii) Image enhancement using Harmony-Search-Optimization and Otsu thresholding; (iii) Image segmentation to extract infected region(s); and (iv) Region-of-interest (ROI) extraction (features) from binary image to compute level of severity. The features that are extracted from ROI are then employed to identify the pixel ratio between the lung and infection sections to identify infection level of severity. The primary objective of the tool is to assist the pulmonologist not only to detect but also to help plan treatment process. As a consequence, for mass screening processing, it will help prevent diagnostic burden.

Motivation & Objective

• Motivate automated assistance for pulmonologists to detect COVID-19 pneumonia from lung CT scans.
• Develop an image-processing pipeline to extract infected regions from coronal lung CT views.
• Enhance CT images using Harmony-Search optimization and Otsu thresholding.
• Segment infected regions and extract ROI-based features to quantify infection severity.
• Provide a tool to aid mass screening and reduce diagnostic burden.

Proposed method

• Apply a threshold filter to isolate the lung region and reduce artifacts.
• Enhance images using Harmony-Search optimization combined with Otsu thresholding.
• Segment infected regions to obtain discrete ROI
• Extract features from the ROI on the binary image to compute lung-to-infection pixel ratios.
• Use ROI features to estimate infection severity level for clinical planning.

Experimental results

Research questions

• RQ1 Can a Harmony-Search and Otsu-based technique reliably enhance CT images for better infection segmentation?
• RQ2 Does ROI-based feature extraction from CT images correlate with infection severity to support clinical decisions?
• RQ3 Can the proposed system aid mass screening by reducing pulmonologist diagnostic burden?
• RQ4 What is the role of the created features in distinguishing infected versus healthy lung regions in CT scans?

Key findings

• Proposes an image-assisted system to extract COVID-19 infected sections from lung CT scans (coronal view).
• Introduces a workflow including thresholding, Harmony-Search-Optimization, and Otsu thresholding for enhancement.
• Performs segmentation to extract infected regions and ROI-based feature extraction for severity assessment.
• Aims to assist pulmonologists in detection and treatment planning, and to support mass screening.