Physics-informed machine learning digital twin for reconstructing prostate cancer tumor growth via PSA tests

Abstract Existing prostate cancer monitoring methods, reliant on prostate-specific antigen (PSA) measurements in blood tests often fail to detect tumor growth. We develop a computational framework to reconstruct tumor growth from the PSA integrating physics-based modeling and machine learning in dig...

Full description

Saved in:
Bibliographic Details
Main Authors: Daniel Camacho-Gomez, Carlos Borau, Jose Manuel Garcia-Aznar, Maria Jose Gomez-Benito, Mark Girolami, Maria Angeles Perez
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:npj Digital Medicine
Online Access:https://doi.org/10.1038/s41746-025-01890-x
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849234373676630016
author Daniel Camacho-Gomez
Carlos Borau
Jose Manuel Garcia-Aznar
Maria Jose Gomez-Benito
Mark Girolami
Maria Angeles Perez
author_facet Daniel Camacho-Gomez
Carlos Borau
Jose Manuel Garcia-Aznar
Maria Jose Gomez-Benito
Mark Girolami
Maria Angeles Perez
author_sort Daniel Camacho-Gomez
collection DOAJ
description Abstract Existing prostate cancer monitoring methods, reliant on prostate-specific antigen (PSA) measurements in blood tests often fail to detect tumor growth. We develop a computational framework to reconstruct tumor growth from the PSA integrating physics-based modeling and machine learning in digital twins. The physics-based model considers PSA secretion and flux from tissue to blood, depending on local vascularity. This model is enhanced by deep learning, which regulates tumor growth dynamics through the patient’s PSA blood tests and 3D spatial interactions of physiological variables of the digital twin. We showcase our framework by reconstructing tumor growth in real patients over 2.5 years from diagnosis, with tumor volume relative errors ranging from 0.8% to 12.28%. Additionally, our results reveal scenarios of tumor growth despite no significant rise in PSA levels. Therefore, our framework serves as a promising tool for prostate cancer monitoring, supporting the advancement of personalized monitoring protocols.
format Article
id doaj-art-2cedeb56ee4b4eecbda4926ee8226768
institution Kabale University
issn 2398-6352
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series npj Digital Medicine
spelling doaj-art-2cedeb56ee4b4eecbda4926ee82267682025-08-20T04:03:11ZengNature Portfolionpj Digital Medicine2398-63522025-07-018111010.1038/s41746-025-01890-xPhysics-informed machine learning digital twin for reconstructing prostate cancer tumor growth via PSA testsDaniel Camacho-Gomez0Carlos Borau1Jose Manuel Garcia-Aznar2Maria Jose Gomez-Benito3Mark Girolami4Maria Angeles Perez5Department of Mechanical Engineering, Multiscale in Mechanical and Biological Engineering (M2BE), Aragon Institute of Engineering. Research (I3A), University of ZaragozaDepartment of Mechanical Engineering, Multiscale in Mechanical and Biological Engineering (M2BE), Aragon Institute of Engineering. Research (I3A), University of ZaragozaDepartment of Mechanical Engineering, Multiscale in Mechanical and Biological Engineering (M2BE), Aragon Institute of Engineering. Research (I3A), University of ZaragozaDepartment of Mechanical Engineering, Multiscale in Mechanical and Biological Engineering (M2BE), Aragon Institute of Engineering. Research (I3A), University of ZaragozaDepartment of Engineering, University of CambridgeDepartment of Mechanical Engineering, Multiscale in Mechanical and Biological Engineering (M2BE), Aragon Institute of Engineering. Research (I3A), University of ZaragozaAbstract Existing prostate cancer monitoring methods, reliant on prostate-specific antigen (PSA) measurements in blood tests often fail to detect tumor growth. We develop a computational framework to reconstruct tumor growth from the PSA integrating physics-based modeling and machine learning in digital twins. The physics-based model considers PSA secretion and flux from tissue to blood, depending on local vascularity. This model is enhanced by deep learning, which regulates tumor growth dynamics through the patient’s PSA blood tests and 3D spatial interactions of physiological variables of the digital twin. We showcase our framework by reconstructing tumor growth in real patients over 2.5 years from diagnosis, with tumor volume relative errors ranging from 0.8% to 12.28%. Additionally, our results reveal scenarios of tumor growth despite no significant rise in PSA levels. Therefore, our framework serves as a promising tool for prostate cancer monitoring, supporting the advancement of personalized monitoring protocols.https://doi.org/10.1038/s41746-025-01890-x
spellingShingle Daniel Camacho-Gomez
Carlos Borau
Jose Manuel Garcia-Aznar
Maria Jose Gomez-Benito
Mark Girolami
Maria Angeles Perez
Physics-informed machine learning digital twin for reconstructing prostate cancer tumor growth via PSA tests
npj Digital Medicine
title Physics-informed machine learning digital twin for reconstructing prostate cancer tumor growth via PSA tests
title_full Physics-informed machine learning digital twin for reconstructing prostate cancer tumor growth via PSA tests
title_fullStr Physics-informed machine learning digital twin for reconstructing prostate cancer tumor growth via PSA tests
title_full_unstemmed Physics-informed machine learning digital twin for reconstructing prostate cancer tumor growth via PSA tests
title_short Physics-informed machine learning digital twin for reconstructing prostate cancer tumor growth via PSA tests
title_sort physics informed machine learning digital twin for reconstructing prostate cancer tumor growth via psa tests
url https://doi.org/10.1038/s41746-025-01890-x
work_keys_str_mv AT danielcamachogomez physicsinformedmachinelearningdigitaltwinforreconstructingprostatecancertumorgrowthviapsatests
AT carlosborau physicsinformedmachinelearningdigitaltwinforreconstructingprostatecancertumorgrowthviapsatests
AT josemanuelgarciaaznar physicsinformedmachinelearningdigitaltwinforreconstructingprostatecancertumorgrowthviapsatests
AT mariajosegomezbenito physicsinformedmachinelearningdigitaltwinforreconstructingprostatecancertumorgrowthviapsatests
AT markgirolami physicsinformedmachinelearningdigitaltwinforreconstructingprostatecancertumorgrowthviapsatests
AT mariaangelesperez physicsinformedmachinelearningdigitaltwinforreconstructingprostatecancertumorgrowthviapsatests