Your browser doesn't support javascript.
Eight pruning deep learning models for low storage and high-speed COVID-19 computed tomography lung segmentation and heatmap-based lesion localization: A multicenter study using COVLIAS 2.0.
Agarwal, Mohit; Agarwal, Sushant; Saba, Luca; Chabert, Gian Luca; Gupta, Suneet; Carriero, Alessandro; Pasche, Alessio; Danna, Pietro; Mehmedovic, Armin; Faa, Gavino; Shrivastava, Saurabh; Jain, Kanishka; Jain, Harsh; Jujaray, Tanay; Singh, Inder M; Turk, Monika; Chadha, Paramjit S; Johri, Amer M; Khanna, Narendra N; Mavrogeni, Sophie; Laird, John R; Sobel, David W; Miner, Martin; Balestrieri, Antonella; Sfikakis, Petros P; Tsoulfas, George; Misra, Durga Prasanna; Agarwal, Vikas; Kitas, George D; Teji, Jagjit S; Al-Maini, Mustafa; Dhanjil, Surinder K; Nicolaides, Andrew; Sharma, Aditya; Rathore, Vijay; Fatemi, Mostafa; Alizad, Azra; Krishnan, Pudukode R; Yadav, Rajanikant R; Nagy, Frence; Kincses, Zsigmond Tamás; Ruzsa, Zoltan; Naidu, Subbaram; Viskovic, Klaudija; Kalra, Manudeep K; Suri, Jasjit S.
  • Agarwal M; Department of Computer Science Engineering, Bennett University, India.
  • Agarwal S; Department of Computer Science Engineering, PSIT, Kanpur, India; Advanced Knowledge Engineering Centre, Global Biomedical Technologies, Inc., Roseville, CA 95661, USA.
  • Saba L; Department of Radiology, Azienda Ospedaliero Universitaria (A.O.U.), Cagliari, Italy.
  • Chabert GL; Department of Radiology, Azienda Ospedaliero Universitaria (A.O.U.), Cagliari, Italy.
  • Gupta S; Department of Computer Science Engineering, Bennett University, India.
  • Carriero A; Department of Radiology, Azienda Ospedaliero Universitaria (A.O.U.), Cagliari, Italy.
  • Pasche A; Depart of Radiology, "Maggiore della Carità" Hospital, University of Piemonte Orientale, Via Solaroli 17, 28100, Novara, Italy.
  • Danna P; Depart of Radiology, "Maggiore della Carità" Hospital, University of Piemonte Orientale, Via Solaroli 17, 28100, Novara, Italy.
  • Mehmedovic A; University Hospital for Infectious Diseases, Zagreb, Croatia.
  • Faa G; Department of Pathology - AOU of Cagliari, Italy.
  • Shrivastava S; College of Computing Sciences and IT, Teerthanker Mahaveer University, Moradabad, 244001, India.
  • Jain K; College of Computing Sciences and IT, Teerthanker Mahaveer University, Moradabad, 244001, India.
  • Jain H; College of Computing Sciences and IT, Teerthanker Mahaveer University, Moradabad, 244001, India.
  • Jujaray T; Dept of Molecular, Cell and Developmental Biology, University of California, Santa Cruz, CA, USA.
  • Singh IM; AtheroPoint LLC, CA, USA.
  • Turk M; The Hanse-Wissenschaftskolleg Institute for Advanced Study, Delmenhorst, Germany.
  • Chadha PS; AtheroPoint LLC, CA, USA.
  • Johri AM; Division of Cardiology, Queen's University, Kingston, Ontario, Canada.
  • Khanna NN; Department of Cardiology, Indraprastha APOLLO Hospitals, New Delhi, India.
  • Mavrogeni S; Cardiology Clinic, Onassis Cardiac Surgery Center, Athens, Greece.
  • Laird JR; Heart and Vascular Institute, Adventist Health St. Helena, St Helena, CA, USA.
  • Sobel DW; Minimally Invasive Urology Institute, Brown University, Providence, RI, USA.
  • Miner M; Men's Health Center, Miriam Hospital Providence, Rhode Island, USA.
  • Balestrieri A; Department of Radiology, Azienda Ospedaliero Universitaria (A.O.U.), Cagliari, Italy.
  • Sfikakis PP; Rheumatology Unit, National Kapodistrian University of Athens, Greece.
  • Tsoulfas G; Aristoteleion University of Thessaloniki, Thessaloniki, Greece.
  • Misra DP; Dept. of Immunology, SGPIMS, Lucknow, UP, India.
  • Agarwal V; Dept. of Immunology, SGPIMS, Lucknow, UP, India.
  • Kitas GD; Academic Affairs, Dudley Group NHS Foundation Trust, Dudley, UK; Arthritis Research UK Epidemiology Unit, Manchester University, Manchester, UK.
  • Teji JS; Ann and Robert H. Lurie Children's Hospital of Chicago, Chicago, USA.
  • Al-Maini M; Allergy, Clinical Immunology and Rheumatology Institute, Toronto, Canada.
  • Dhanjil SK; AtheroPoint LLC, CA, USA.
  • Nicolaides A; Vascular Screening and Diagnostic Centre and Univ. of Nicosia Medical School, Cyprus.
  • Sharma A; Division of Cardiovascular Medicine, University of Virginia, Charlottesville, VA, USA.
  • Rathore V; AtheroPoint LLC, CA, USA.
  • Fatemi M; Dept. of Physiology & Biomedical Engg., Mayo Clinic College of Medicine and Science, MN, USA.
  • Alizad A; Dept. of Radiology, Mayo Clinic College of Medicine and Science, MN, USA.
  • Krishnan PR; Neurology Department, Fortis Hospital, Bangalore, India.
  • Yadav RR; Radiodiagnosis, SGPIMS, Lucknow, Uttar Pradesh, India.
  • Nagy F; Department of Radiology, University of Szeged, 6725, Hungary.
  • Kincses ZT; Department of Radiology, University of Szeged, 6725, Hungary.
  • Ruzsa Z; Invasive Cardiology Division, University of Szeged, Budapest, Hungary.
  • Naidu S; Electrical Engineering Department, University of Minnesota, Duluth, MN, USA.
  • Viskovic K; University Hospital for Infectious Diseases, Zagreb, Croatia.
  • Kalra MK; Department of Radiology, Massachusetts General Hospital, Boston, MA, USA.
  • Suri JS; College of Computing Sciences and IT, Teerthanker Mahaveer University, Moradabad, 244001, India; Stroke Diagnostic and Monitoring Division, AtheroPoint™, Roseville, CA, USA. Electronic address: jasjit.suri@atheropoint.com.
Comput Biol Med ; 146: 105571, 2022 07.
Article in English | MEDLINE | ID: covidwho-1850900
ABSTRACT

BACKGROUND:

COVLIAS 1.0 an automated lung segmentation was designed for COVID-19 diagnosis. It has issues related to storage space and speed. This study shows that COVLIAS 2.0 uses pruned AI (PAI) networks for improving both storage and speed, wiliest high performance on lung segmentation and lesion localization.

METHOD:

ology The proposed study uses multicenter ∼9,000 CT slices from two different nations, namely, CroMed from Croatia (80 patients, experimental data), and NovMed from Italy (72 patients, validation data). We hypothesize that by using pruning and evolutionary optimization algorithms, the size of the AI models can be reduced significantly, ensuring optimal performance. Eight different pruning techniques (i) differential evolution (DE), (ii) genetic algorithm (GA), (iii) particle swarm optimization algorithm (PSO), and (iv) whale optimization algorithm (WO) in two deep learning frameworks (i) Fully connected network (FCN) and (ii) SegNet were designed. COVLIAS 2.0 was validated using "Unseen NovMed" and benchmarked against MedSeg. Statistical tests for stability and reliability were also conducted.

RESULTS:

Pruning algorithms (i) FCN-DE, (ii) FCN-GA, (iii) FCN-PSO, and (iv) FCN-WO showed improvement in storage by 92.4%, 95.3%, 98.7%, and 99.8% respectively when compared against solo FCN, and (v) SegNet-DE, (vi) SegNet-GA, (vii) SegNet-PSO, and (viii) SegNet-WO showed improvement by 97.1%, 97.9%, 98.8%, and 99.2% respectively when compared against solo SegNet. AUC > 0.94 (p < 0.0001) on CroMed and > 0.86 (p < 0.0001) on NovMed data set for all eight EA model. PAI <0.25 s per image. DenseNet-121-based Grad-CAM heatmaps showed validation on glass ground opacity lesions.

CONCLUSIONS:

Eight PAI networks that were successfully validated are five times faster, storage efficient, and could be used in clinical settings.
Subject(s)
Keywords
Fulltext
Print
XML
PubMed Links
Search on Google

Full text: Available Collection: International databases Database: MEDLINE Main subject: Deep Learning / COVID-19 Type of study: Diagnostic study / Experimental Studies / Prognostic study Limits: Humans Language: English Journal: Comput Biol Med Year: 2022 Document Type: Article Affiliation country: J.compbiomed.2022.105571

Similar

MEDLINE
LILACS
LIS
Fulltext
Print
XML
PubMed Links
Search on Google

Full text: Available Collection: International databases Database: MEDLINE Main subject: Deep Learning / COVID-19 Type of study: Diagnostic study / Experimental Studies / Prognostic study Limits: Humans Language: English Journal: Comput Biol Med Year: 2022 Document Type: Article Affiliation country: J.compbiomed.2022.105571