Diagnostics (Jun 2022)
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
Affiliations
- Jasjit S. Suri
- Stroke Diagnostic and Monitoring Division, AtheroPoint™, Roseville, CA 95661, USA
- Sushant Agarwal
- Advanced Knowledge Engineering Centre, GBTI, Roseville, CA 95661, USA
- Gian Luca Chabert
- Department of Radiology, Azienda Ospedaliero Universitaria (A.O.U.), 09123 Cagliari, Italy
- Alessandro Carriero
- Department of Radiology, “Maggiore della Carità” Hospital, University of Piemonte Orientale (UPO), Via Solaroli 17, 28100 Novara, Italy
- Alessio Paschè
- Department of Radiology, Azienda Ospedaliero Universitaria (A.O.U.), 09123 Cagliari, Italy
- Pietro S. C. Danna
- Department of Radiology, Azienda Ospedaliero Universitaria (A.O.U.), 09123 Cagliari, Italy
- Luca Saba
- Department of Radiology, Azienda Ospedaliero Universitaria (A.O.U.), 09123 Cagliari, Italy
- Armin Mehmedović
- Department of Radiology, University Hospital for Infectious Diseases, 10000 Zagreb, Croatia
- Gavino Faa
- Department of Pathology, Azienda Ospedaliero Universitaria (A.O.U.), 09124 Cagliari, Italy
- Inder M. Singh
- Stroke Diagnostic and Monitoring Division, AtheroPoint™, Roseville, CA 95661, USA
- Monika Turk
- The Hanse-Wissenschaftskolleg Institute for Advanced Study, 27753 Delmenhorst, Germany
- Paramjit S. Chadha
- Stroke Diagnostic and Monitoring Division, AtheroPoint™, Roseville, CA 95661, USA
- Amer M. Johri
- Department of Medicine, Division of Cardiology, Queen’s University, Kingston, ON K7L 3N6, Canada
- Narendra N. Khanna
- Department of Cardiology, Indraprastha APOLLO Hospitals, New Delhi 110076, India
- Sophie Mavrogeni
- Cardiology Clinic, Onassis Cardiac Surgery Center, 17674 Athens, Greece
- John R. Laird
- Heart and Vascular Institute, Adventist Health St. Helena, St. Helena, CA 94574, USA
- Gyan Pareek
- Minimally Invasive Urology Institute, Brown University, Providence, RI 02912, USA
- Martin Miner
- Men’s Health Center, Miriam Hospital, Providence, RI 02912, USA
- David W. Sobel
- Minimally Invasive Urology Institute, Brown University, Providence, RI 02912, USA
- Antonella Balestrieri
- Department of Radiology, Azienda Ospedaliero Universitaria (A.O.U.), 09123 Cagliari, Italy
- Petros P. Sfikakis
- Rheumatology Unit, National Kapodistrian University of Athens, 17674 Athens, Greece
- George Tsoulfas
- Department of Surgery, Aristoteleion University of Thessaloniki, 54124 Thessaloniki, Greece
- Athanasios D. Protogerou
- Cardiovascular Prevention and Research Unit, Department of Pathophysiology, National & Kapodistrian University of Athens, 15772 Athens, Greece
- Durga Prasanna Misra
- Department of Immunology, SGPIMS, Lucknow 226014, India
- Vikas Agarwal
- Department of Immunology, SGPIMS, Lucknow 226014, India
- George D. Kitas
- Academic Affairs, Dudley Group NHS Foundation Trust, Dudley DY1 2HQ, UK
- Jagjit S. Teji
- Ann and Robert H. Lurie Children’s Hospital of Chicago, Chicago, IL 60611, USA
- Mustafa Al-Maini
- Allergy, Clinical Immunology and Rheumatology Institute, Toronto, ON M5G 1N8, Canada
- Surinder K. Dhanjil
- AtheroPoint LLC., Roseville, CA 95661, USA
- Andrew Nicolaides
- Vascular Screening and Diagnostic Centre, University of Nicosia Medical School, Engomi 2408, Cyprus
- Aditya Sharma
- Division of Cardiovascular Medicine, University of Virginia, Charlottesville, VA 22902, USA
- Vijay Rathore
- AtheroPoint LLC., Roseville, CA 95661, USA
- Mostafa Fatemi
- Department of Physiology & Biomedical Engineering, Mayo Clinic College of Medicine and Science, Rochester, MN 55905, USA
- Azra Alizad
- Department of Radiology, Mayo Clinic College of Medicine and Science, Rochester, MN 55905, USA
- Pudukode R. Krishnan
- Neurology Department, Fortis Hospital, Bengaluru 560076, India
- Ferenc Nagy
- Internal Medicine Department, University of Szeged, 6725 Szeged, Hungary
- Zoltan Ruzsa
- Invasive Cardiology Division, University of Szeged, 1122 Budapest, Hungary
- Mostafa M. Fouda
- Department of ECE, Idaho State University, Pocatello, ID 83209, USA
- Subbaram Naidu
- Electrical Engineering Department, University of Minnesota, Duluth, MN 55812, USA
- Klaudija Viskovic
- Department of Radiology, University Hospital for Infectious Diseases, 10000 Zagreb, Croatia
- Mannudeep K. Kalra
- Department of Radiology, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, USA
- DOI
- https://doi.org/10.3390/diagnostics12061482
- Journal volume & issue
-
Vol. 12,
no. 6
p. 1482
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
