@article{3030427,
    title = "COVLIAS 1.0: Lung Segmentation in COVID-19 Computed Tomography Scans
Using Hybrid Deep Learning Artificial Intelligence Models",
    author = "Suri, Jasjit S. and Agarwal, Sushant and Pathak, Rajesh and Ketireddy, and Vedmanvitha and Columbu, Marta and Saba, Luca and Gupta, Suneet K. and and Faa, Gavino and Singh, Inder M. and Turk, Monika and Chadha, Paramjit S. and and Johri, Amer M. and Khanna, Narendra N. and Viskovic, Klaudija and and Mavrogeni, Sophie and Laird, John R. and Pareek, Gyan and Miner, Martin and and Sobel, David W. and Balestrieri, Antonella and Sfikakis, Petros P. and and Tsoulfas, George and Protogerou, Athanasios and Misra, Durga and Prasanna and Agarwal, Vikas and Kitas, George D. and Teji, Jagjit S. and and Al-Maini, Mustafa and Dhanjil, Surinder K. and Nicolaides, Andrew and and Sharma, Aditya and Rathore, Vijay and Fatemi, Mostafa and Alizad, Azra and and Krishnan, Pudukode R. and Frence, Nagy and Ruzsa, Zoltan and Gupta, and Archna and Naidu, Subbaram and Kalra, Mannudeep",
    journal = "DIAGNOSTIC ONCOLOGY",
    year = "2021",
    volume = "11",
    number = "8",
    publisher = "MDPI",
    doi = "10.3390/diagnostics11081405",
    keywords = "COVID-19; computed tomography; lungs; segmentation; hybrid deep learning",
    abstract = "Background: COVID-19 lung segmentation using Computed Tomography (CT)
scans is important for the diagnosis of lung severity. The process of
automated lung segmentation is challenging due to (a) CT radiation
dosage and (b) ground-glass opacities caused by COVID-19. The lung
segmentation methodologies proposed in 2020 were semi- or automated but
not reliable, accurate, and user-friendly. The proposed study presents a
COVID Lung Image Analysis System (COVLIAS 1.0, AtheroPoint (TM),
Roseville, CA, USA) consisting of hybrid deep learning (HDL) models for
lung segmentation. Methodology: The COVLIAS 1.0 consists of three
methods based on solo deep learning (SDL) or hybrid deep learning (HDL).
SegNet is proposed in the SDL category while VGG-SegNet and
ResNet-SegNet are designed under the HDL paradigm. The three proposed AI
approaches were benchmarked against the National Institute of Health
(NIH)-based conventional segmentation model using fuzzy-connectedness. A
cross-validation protocol with a 40:60 ratio between training and
testing was designed, with 10% validation data. The ground truth (GT)
was manually traced by a radiologist trained personnel. For performance
evaluation, nine different criteria were selected to perform the
evaluation of SDL or HDL lung segmentation regions and lungs long axis
against GT. Results: Using the database of 5000 chest CT images (from 72
patients), COVLIAS 1.0 yielded AUC of similar to 0.96, similar to 0.97,
similar to 0.98, and similar to 0.96 (p-value < 0.001), respectively
within 5% range of GT area, for SegNet, VGG-SegNet, ResNet-SegNet, and
NIH. The mean Figure of Merit using four models (left and right lung)
was above 94%. On benchmarking against the National Institute of Health
(NIH) segmentation method, the proposed model demonstrated a 58% and
44% improvement in ResNet-SegNet, 52% and 36% improvement in
VGG-SegNet for lung area, and lung long axis, respectively. The PE
statistics performance was in the following order: ResNet-SegNet >
VGG-SegNet > NIH > SegNet. The HDL runs in <1 s on test data per image.
Conclusions: The COVLIAS 1.0 system can be applied in real-time for
radiology-based clinical settings."
}