COVLIAS 2.0-cXAI: Cloud-Based Explainable Deep Learning System for COVID-19 Lesion Localization in Computed Tomography Scans

Jasjit S Suri; Sushant Agarwal; Gian Luca Chabert; Alessandro Carriero; Alessio Paschè; Pietro S C Danna; Luca Saba; Armin Mehmedović; Gavino Faa; Inder M Singh; Monika Turk; Paramjit S Chadha; Amer M Johri; Narendra N Khanna; Sophie Mavrogeni; John R Laird; Gyan Pareek; Martin Miner; David W Sobel; Antonella Balestrieri; Petros P Sfikakis; George Tsoulfas; Athanasios D Protogerou; Durga Prasanna Misra; Vikas Agarwal; George D Kitas; Jagjit S Teji; Mustafa Al-Maini; Surinder K Dhanjil; Andrew Nicolaides; Aditya Sharma; Vijay Rathore; Mostafa Fatemi; Azra Alizad; Pudukode R Krishnan; Ferenc Nagy; Zoltan Ruzsa; Mostafa M Fouda; Subbaram Naidu; Klaudija Viskovic; Mannudeep K Kalra

doi:10.3390/diagnostics12061482

COVLIAS 2.0-cXAI: Cloud-Based Explainable Deep Learning System for COVID-19 Lesion Localization in Computed Tomography Scans

Diagnostics (Basel). 2022 Jun 16;12(6):1482. doi: 10.3390/diagnostics12061482.

Authors

Jasjit S Suri^{1

2}, Sushant Agarwal^{2

3}, Gian Luca Chabert⁴, Alessandro Carriero⁵, Alessio Paschè⁴, Pietro S C Danna⁴, Luca Saba⁴, Armin Mehmedović⁶, Gavino Faa⁷, Inder M Singh¹, Monika Turk⁸, Paramjit S Chadha¹, Amer M Johri⁹, Narendra N Khanna¹⁰, Sophie Mavrogeni¹¹, John R Laird¹², Gyan Pareek¹³, Martin Miner¹⁴, David W Sobel¹³, Antonella Balestrieri⁴, Petros P Sfikakis¹⁵, George Tsoulfas¹⁶, Athanasios D Protogerou¹⁷, Durga Prasanna Misra¹⁸, Vikas Agarwal¹⁸, George D Kitas^{19

20}, Jagjit S Teji²¹, Mustafa Al-Maini²², Surinder K Dhanjil²³, Andrew Nicolaides²⁴, Aditya Sharma²⁵, Vijay Rathore²³, Mostafa Fatemi²⁶, Azra Alizad²⁷, Pudukode R Krishnan²⁸, Ferenc Nagy²⁹, Zoltan Ruzsa³⁰, Mostafa M Fouda³¹, Subbaram Naidu³², Klaudija Viskovic⁶, Mannudeep K Kalra³³

Affiliations

¹ Stroke Diagnostic and Monitoring Division, AtheroPoint™, Roseville, CA 95661, USA.
² Advanced Knowledge Engineering Centre, GBTI, Roseville, CA 95661, USA.
³ Department of Computer Science Engineering, PSIT, Kanpur 209305, India.
⁴ Department of Radiology, Azienda Ospedaliero Universitaria (A.O.U.), 09123 Cagliari, Italy.
⁵ Department of Radiology, "Maggiore della Carità" Hospital, University of Piemonte Orientale (UPO), Via Solaroli 17, 28100 Novara, Italy.
⁶ Department of Radiology, University Hospital for Infectious Diseases, 10000 Zagreb, Croatia.
⁷ Department of Pathology, Azienda Ospedaliero Universitaria (A.O.U.), 09124 Cagliari, Italy.
⁸ The Hanse-Wissenschaftskolleg Institute for Advanced Study, 27753 Delmenhorst, Germany.
⁹ Department of Medicine, Division of Cardiology, Queen's University, Kingston, ON K7L 3N6, Canada.
¹⁰ Department of Cardiology, Indraprastha APOLLO Hospitals, New Delhi 110076, India.
¹¹ Cardiology Clinic, Onassis Cardiac Surgery Center, 17674 Athens, Greece.
¹² Heart and Vascular Institute, Adventist Health St. Helena, St. Helena, CA 94574, USA.
¹³ Minimally Invasive Urology Institute, Brown University, Providence, RI 02912, USA.
¹⁴ Men's Health Center, Miriam Hospital, Providence, RI 02912, USA.
¹⁵ Rheumatology Unit, National Kapodistrian University of Athens, 17674 Athens, Greece.
¹⁶ Department of Surgery, Aristoteleion University of Thessaloniki, 54124 Thessaloniki, Greece.
¹⁷ Cardiovascular Prevention and Research Unit, Department of Pathophysiology, National & Kapodistrian University of Athens, 15772 Athens, Greece.
¹⁸ Department of Immunology, SGPIMS, Lucknow 226014, India.
¹⁹ Academic Affairs, Dudley Group NHS Foundation Trust, Dudley DY1 2HQ, UK.
²⁰ Arthritis Research UK Epidemiology Unit, Manchester University, Manchester M13 9PL, UK.
²¹ Ann and Robert H. Lurie Children's Hospital of Chicago, Chicago, IL 60611, USA.
²² Allergy, Clinical Immunology and Rheumatology Institute, Toronto, ON M5G 1N8, Canada.
²³ AtheroPoint LLC., Roseville, CA 95661, USA.
²⁴ Vascular Screening and Diagnostic Centre, University of Nicosia Medical School, Engomi 2408, Cyprus.
²⁵ Division of Cardiovascular Medicine, University of Virginia, Charlottesville, VA 22902, USA.
²⁶ Department of Physiology & Biomedical Engineering, Mayo Clinic College of Medicine and Science, Rochester, MN 55905, USA.
²⁷ Department of Radiology, Mayo Clinic College of Medicine and Science, Rochester, MN 55905, USA.
²⁸ Neurology Department, Fortis Hospital, Bengaluru 560076, India.
²⁹ Internal Medicine Department, University of Szeged, 6725 Szeged, Hungary.
³⁰ Invasive Cardiology Division, University of Szeged, 1122 Budapest, Hungary.
³¹ Department of ECE, Idaho State University, Pocatello, ID 83209, USA.
³² Electrical Engineering Department, University of Minnesota, Duluth, MN 55812, USA.
³³ Department of Radiology, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, USA.

Abstract

Background: The previous COVID-19 lung diagnosis system lacks both scientific validation and the role of explainable artificial intelligence (AI) for understanding lesion localization. This study presents a cloud-based explainable AI, the “COVLIAS 2.0-cXAI” system using four kinds of class activation maps (CAM) models. Methodology: Our cohort consisted of ~6000 CT slices from two sources (Croatia, 80 COVID-19 patients and Italy, 15 control patients). COVLIAS 2.0-cXAI design consisted of three stages: (i) automated lung segmentation using hybrid deep learning ResNet-UNet model by automatic adjustment of Hounsfield units, hyperparameter optimization, and parallel and distributed training, (ii) classification using three kinds of DenseNet (DN) models (DN-121, DN-169, DN-201), and (iii) validation using four kinds of CAM visualization techniques: gradient-weighted class activation mapping (Grad-CAM), Grad-CAM++, score-weighted CAM (Score-CAM), and FasterScore-CAM. The COVLIAS 2.0-cXAI was validated by three trained senior radiologists for its stability and reliability. The Friedman test was also performed on the scores of the three radiologists. Results: The ResNet-UNet segmentation model resulted in dice similarity of 0.96, Jaccard index of 0.93, a correlation coefficient of 0.99, with a figure-of-merit of 95.99%, while the classifier accuracies for the three DN nets (DN-121, DN-169, and DN-201) were 98%, 98%, and 99% with a loss of ~0.003, ~0.0025, and ~0.002 using 50 epochs, respectively. The mean AUC for all three DN models was 0.99 (p < 0.0001). The COVLIAS 2.0-cXAI showed 80% scans for mean alignment index (MAI) between heatmaps and gold standard, a score of four out of five, establishing the system for clinical settings. Conclusions: The COVLIAS 2.0-cXAI successfully showed a cloud-based explainable AI system for lesion localization in lung CT scans.

Keywords: COVID-19 lesion; FasterScore-CAM; GRAD-CAM; Grad-CAM++; Hounsfield units; Score-CAM; classification; explainable AI; glass ground opacities; hybrid deep learning; lung CT; segmentation.

Grants and funding

This research received no external funding.