Computed tomography-based machine learning for donor lung screening before transplantation.

Computed tomography Dictionary learning Donor assessment Donor lung screening Lung transplantation Machine learning Primary graft dysfunction

Journal

The Journal of heart and lung transplantation : the official publication of the International Society for Heart Transplantation
ISSN: 1557-3117
Titre abrégé: J Heart Lung Transplant
Pays: United States
ID NLM: 9102703

Informations de publication

Date de publication:
29 Sep 2023
Historique:
received: 22 03 2023
revised: 21 09 2023
accepted: 25 09 2023
pubmed: 2 10 2023
medline: 2 10 2023
entrez: 1 10 2023
Statut: aheadofprint

Résumé

Assessment and selection of donor lungs remain largely subjective and experience based. Criteria to accept or decline lungs are poorly standardized and are not compliant with the current donor pool. Using ex vivo computed tomography (CT) images, we investigated the use of a CT-based machine learning algorithm for screening donor lungs before transplantation. Clinical measures and ex situ CT scans were collected from 100 cases as part of a prospective clinical trial. Following procurement, donor lungs were inflated, placed on ice according to routine clinical practice, and imaged using a clinical CT scanner before transplantation while stored in the icebox. We trained and tested a supervised machine learning method called dictionary learning, which uses CT scans and learns specific image patterns and features pertaining to each class for a classification task. The results were evaluated with donor and recipient clinical measures. Of the 100 lung pairs donated, 70 were considered acceptable for transplantation (based on standard clinical assessment) before CT screening and were consequently implanted. The remaining 30 pairs were screened but not transplanted. Our machine learning algorithm was able to detect pulmonary abnormalities on the CT scans. Among the patients who received donor lungs, our algorithm identified recipients who had extended stays in the intensive care unit and were at 19 times higher risk of developing chronic lung allograft dysfunction within 2 years posttransplant. We have created a strategy to ex vivo screen donor lungs using a CT-based machine learning algorithm. As the use of suboptimal donor lungs rises, it is important to have in place objective techniques that will assist physicians in accurately screening donor lungs to identify recipients most at risk of posttransplant complications.

Sections du résumé

BACKGROUND BACKGROUND
Assessment and selection of donor lungs remain largely subjective and experience based. Criteria to accept or decline lungs are poorly standardized and are not compliant with the current donor pool. Using ex vivo computed tomography (CT) images, we investigated the use of a CT-based machine learning algorithm for screening donor lungs before transplantation.
METHODS METHODS
Clinical measures and ex situ CT scans were collected from 100 cases as part of a prospective clinical trial. Following procurement, donor lungs were inflated, placed on ice according to routine clinical practice, and imaged using a clinical CT scanner before transplantation while stored in the icebox. We trained and tested a supervised machine learning method called dictionary learning, which uses CT scans and learns specific image patterns and features pertaining to each class for a classification task. The results were evaluated with donor and recipient clinical measures.
RESULTS RESULTS
Of the 100 lung pairs donated, 70 were considered acceptable for transplantation (based on standard clinical assessment) before CT screening and were consequently implanted. The remaining 30 pairs were screened but not transplanted. Our machine learning algorithm was able to detect pulmonary abnormalities on the CT scans. Among the patients who received donor lungs, our algorithm identified recipients who had extended stays in the intensive care unit and were at 19 times higher risk of developing chronic lung allograft dysfunction within 2 years posttransplant.
CONCLUSIONS CONCLUSIONS
We have created a strategy to ex vivo screen donor lungs using a CT-based machine learning algorithm. As the use of suboptimal donor lungs rises, it is important to have in place objective techniques that will assist physicians in accurately screening donor lungs to identify recipients most at risk of posttransplant complications.

Identifiants

pubmed: 37778525
pii: S1053-2498(23)02031-4
doi: 10.1016/j.healun.2023.09.018
pii:
doi:

Types de publication

Journal Article

Langues

eng

Sous-ensembles de citation

IM

Commentaires et corrections

Type : UpdateOf

Informations de copyright

Copyright © 2023 International Society for the Heart and Lung Transplantation. Published by Elsevier Inc. All rights reserved.

Auteurs

Sundaresh Ram (S)

Department of Radiology, University of Michigan, Ann Arbor, Michigan; Department of Biomedical Engineering, University of Michigan, Ann Arbor, Michigan.

Stijn E Verleden (SE)

Lung Transplant Unit, Department of Chronic Diseases and Metabolism, Laboratory of Respiratory Diseases and Thoracic Surgery (BREATHE), KU Leuven, Leuven, Belgium; Department of ASTARC, University of Antwerp, Wilrijk, Belgium.

Madhav Kumar (M)

Department of Biomedical Engineering, University of Michigan, Ann Arbor, Michigan.

Alexander J Bell (AJ)

Department of Radiology, University of Michigan, Ann Arbor, Michigan.

Ravi Pal (R)

Department of Radiology, University of Michigan, Ann Arbor, Michigan.

Sofie Ordies (S)

Lung Transplant Unit, Department of Chronic Diseases and Metabolism, Laboratory of Respiratory Diseases and Thoracic Surgery (BREATHE), KU Leuven, Leuven, Belgium.

Arno Vanstapel (A)

Lung Transplant Unit, Department of Chronic Diseases and Metabolism, Laboratory of Respiratory Diseases and Thoracic Surgery (BREATHE), KU Leuven, Leuven, Belgium; Department of Imaging & Pathology, KU Leuven, Leuven, Belgium.

Adriana Dubbeldam (A)

Department of Imaging & Pathology, KU Leuven, Leuven, Belgium.

Robin Vos (R)

Lung Transplant Unit, Department of Chronic Diseases and Metabolism, Laboratory of Respiratory Diseases and Thoracic Surgery (BREATHE), KU Leuven, Leuven, Belgium.

Stefanie Galban (S)

Department of Radiology, University of Michigan, Ann Arbor, Michigan.

Laurens J Ceulemans (LJ)

Lung Transplant Unit, Department of Chronic Diseases and Metabolism, Laboratory of Respiratory Diseases and Thoracic Surgery (BREATHE), KU Leuven, Leuven, Belgium.

Anna E Frick (AE)

Lung Transplant Unit, Department of Chronic Diseases and Metabolism, Laboratory of Respiratory Diseases and Thoracic Surgery (BREATHE), KU Leuven, Leuven, Belgium.

Dirk E Van Raemdonck (DE)

Lung Transplant Unit, Department of Chronic Diseases and Metabolism, Laboratory of Respiratory Diseases and Thoracic Surgery (BREATHE), KU Leuven, Leuven, Belgium.

Johny Verschakelen (J)

Department of Imaging & Pathology, KU Leuven, Leuven, Belgium.

Bart M Vanaudenaerde (BM)

Lung Transplant Unit, Department of Chronic Diseases and Metabolism, Laboratory of Respiratory Diseases and Thoracic Surgery (BREATHE), KU Leuven, Leuven, Belgium.

Geert M Verleden (GM)

Lung Transplant Unit, Department of Chronic Diseases and Metabolism, Laboratory of Respiratory Diseases and Thoracic Surgery (BREATHE), KU Leuven, Leuven, Belgium.

Vibha N Lama (VN)

Pulmonary and Critical Care Medicine, University of Michigan, Ann Arbor, Michigan.

Arne P Neyrinck (AP)

Department of Cardiovascular Sciences, KU Leuven, Leuven, Belgium.

Craig J Galban (CJ)

Department of Radiology, University of Michigan, Ann Arbor, Michigan; Department of Biomedical Engineering, University of Michigan, Ann Arbor, Michigan. Electronic address: cgalban@med.umich.edu.

Classifications MeSH