TY - JOUR
T1 - A Machine Learning-Driven Virtual Biopsy System For Kidney Transplant Patients
AU - Yoo, Daniel
AU - Divard, Gillian
AU - Raynaud, Marc
AU - Cohen, Aaron
AU - Mone, Tom D.
AU - Rosenthal, John Thomas
AU - Bentall, Andrew J.
AU - Stegall, Mark D.
AU - Naesens, Maarten
AU - Zhang, Huanxi
AU - Wang, Changxi
AU - Gueguen, Juliette
AU - Kamar, Nassim
AU - Bouquegneau, Antoine
AU - Batal, Ibrahim
AU - Coley, Shana M.
AU - Gill, John S.
AU - Oppenheimer, Federico
AU - De Sousa-Amorim, Erika
AU - Kuypers, Dirk R.J.
AU - Durrbach, Antoine
AU - Seron, Daniel
AU - Rabant, Marion
AU - Van Huyen, Jean Paul Duong
AU - Campbell, Patricia
AU - Shojai, Soroush
AU - Mengel, Michael
AU - Bestard, Oriol
AU - Basic-Jukic, Nikolina
AU - Jurić, Ivana
AU - Boor, Peter
AU - Cornell, Lynn D.
AU - Alexander, Mariam P.
AU - Toby Coates, P.
AU - Legendre, Christophe
AU - Reese, Peter P.
AU - Lefaucheur, Carmen
AU - Aubert, Olivier
AU - Loupy, Alexandre
N1 - Publisher Copyright:
© 2024, The Author(s).
PY - 2024/12
Y1 - 2024/12
N2 - In kidney transplantation, day-zero biopsies are used to assess organ quality and discriminate between donor-inherited lesions and those acquired post-transplantation. However, many centers do not perform such biopsies since they are invasive, costly and may delay the transplant procedure. We aim to generate a non-invasive virtual biopsy system using routinely collected donor parameters. Using 14,032 day-zero kidney biopsies from 17 international centers, we develop a virtual biopsy system. 11 basic donor parameters are used to predict four Banff kidney lesions: arteriosclerosis, arteriolar hyalinosis, interstitial fibrosis and tubular atrophy, and the percentage of renal sclerotic glomeruli. Six machine learning models are aggregated into an ensemble model. The virtual biopsy system shows good performance in the internal and external validation sets. We confirm the generalizability of the system in various scenarios. This system could assist physicians in assessing organ quality, optimizing allograft allocation together with discriminating between donor derived and acquired lesions post-transplantation.
AB - In kidney transplantation, day-zero biopsies are used to assess organ quality and discriminate between donor-inherited lesions and those acquired post-transplantation. However, many centers do not perform such biopsies since they are invasive, costly and may delay the transplant procedure. We aim to generate a non-invasive virtual biopsy system using routinely collected donor parameters. Using 14,032 day-zero kidney biopsies from 17 international centers, we develop a virtual biopsy system. 11 basic donor parameters are used to predict four Banff kidney lesions: arteriosclerosis, arteriolar hyalinosis, interstitial fibrosis and tubular atrophy, and the percentage of renal sclerotic glomeruli. Six machine learning models are aggregated into an ensemble model. The virtual biopsy system shows good performance in the internal and external validation sets. We confirm the generalizability of the system in various scenarios. This system could assist physicians in assessing organ quality, optimizing allograft allocation together with discriminating between donor derived and acquired lesions post-transplantation.
UR - http://www.scopus.com/inward/record.url?scp=85182450270&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85182450270&partnerID=8YFLogxK
U2 - 10.1038/s41467-023-44595-z
DO - 10.1038/s41467-023-44595-z
M3 - Article
C2 - 38228634
AN - SCOPUS:85182450270
SN - 2041-1723
VL - 15
JO - Nature communications
JF - Nature communications
IS - 1
M1 - 554
ER -