The Charter of the French Language and its regulations govern the consultation of English-language content.
2301 Avenue d’Estimauville Québec (Québec) G1E 0N5 Canada
Médecin clinicien enseignant sous octroi adjoint, Faculté de médecine, Médecine-Département de radiologie, Université Laval
Chercheur, CERVO
Axe de recherche:
Neurosciences cliniques et cognitives
Louis Gagnon est radiologiste au CHU de Québec et professeur au Département de radiologie et médecine moléculaire de l’Université Laval depuis juillet 2023. Ses projets de recherche se concentrent en neuro imagerie, notamment sur la modélisation et les principes physiques de même que sur l’intelligence artificielle en neuro-radiologie.
Louis Gagnon a complété sa formation en génie physique à l’École Polytechnique de Montréal et en physique des particules à l’accélérateur TRIUMF à Vancouver. Il a par la suite réalisé des maîtrises en génie biomédical à l’École Polytechnique et en génie électrique au MIT. Il a complété son doctorat (PhD) à la Harvard-MIT Division of Health Sciences and Technology. Il est revenu à l’Université Laval pour compléter son doctorat en médecine (MD) et sa résidence en radiologie diagnostique avant de sur-spécialiser en IRM à l’Université de Californie à San Diego.
Taha Belbadaoui; Andrew Forester; Farzad Khalvati; Louis Gagnon
Dans: Neurooncol Adv, vol. 8, no 1, p. vdag178, 2026, ISSN: 2632-2498.
@article{pmid42487901,
title = {Integrating chemotherapy, radiotherapy, and O6-methylguanine-DNA methyltransferase (MGMT) status with deep-learning cellular tumor volumetry sharpens prediction of glioblastoma recurrence on postoperative MRI},
author = {Taha Belbadaoui and Andrew Forester and Farzad Khalvati and Louis Gagnon},
doi = {10.1093/noajnl/vdag178},
issn = {2632-2498},
year = {2026},
date = {2026-01-01},
journal = {Neurooncol Adv},
volume = {8},
number = {1},
pages = {vdag178},
abstract = {BACKGROUND: Distinguishing glioblastoma recurrence from posttreatment effects on magnetic resonance imaging (MRI) remains a major diagnostic challenge. While recent models consider O6-methylguanine-DNA methyltransferase (MGMT) status, most machine learning models addressing this problem assume uniform chemotherapy and radiotherapy exposure, overlooking real-world treatment variability. This study aimed to determine whether integrating chemotherapy, radiotherapy, and MGMT status together with automated cellular tumor volume (VolCT) improves predictive accuracy in posttreatment glioblastoma assessment.nnMETHODS: We retrospectively analyzed 135 postsurgical MRI examinations from 97 glioblastoma patients treated between January 2008 and December 2022. The dataset included 105 confirmed recurrences and 30 cases of treatment-related change. A total of 8465 radiomic features were extracted from 5 MRI sequences (T1-weighted pre/postcontrast, T2-weighted, FLAIR, and restricted spectrum imaging cellularity maps). Cellular tumor volumes were automatically segmented using nnU-Net and combined with chemotherapy radiotherapy and MGMT status. Models were trained using Extremely Randomized Trees with nested Monte Carlo cross-validation.nnRESULTS: The baseline Top10 radiomic model achieved an area under the curve (AUC) of 0.765. Incorporation of nnU-Net-derived VolCT significantly improved performance to 0.809 ( = 2.84 × 10). Adding chemotherapy, radiotherapy, or MGMT status to the Top10+VolCT model yielded further gains, with AUCs ranging from 0.828 to 0.830 (all ≤ .032). The full model combining VolCT with chemotherapy, radiotherapy, and MGMT achieved the best performance, with an AUC of 0.853 ( = .0006 vs Top10+VolCT).nnCONCLUSIONS: Incorporating chemotherapy, radiotherapy, and MGMT status improves posttreatment glioblastoma classification. Deep learning-derived cellular tumor volumetry further enhances radiomics-based performance, highlighting the value of combining clinical context with advanced computational imaging.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Dominic LaBella; Katherine Schumacher; Michael Mix; Kevin Leu; Shan McBurney-Lin; Pierre Nedelec; Javier Villanueva-Meyer; David R Raleigh; Jonathan Shapey; Tom Vercauteren; Kazumi Chia; Marina Ivory; Theodore Barfoot; Omar Al-Salihi; Justin Leu; Lia M Halasz; Yury Velichko; Chunhao Wang; John P Kirkpatrick; Scott R Floyd; Zachary J Reitman; Trey C Mullikin; Eugene J Vaios; Ulas Bagci; Sean Sachdev; Jona A Hattangadi-Gluth; Tyler M Seibert; Nikdokht Farid; Connor Puett; Matthew W Pease; Kevin Shiue; Syed M Anwar; Shahriar Faghani; Peter Taylor; Pranav Warman; Jake Albrecht; András Jakab; Mana Moassefi; Verena Chung; Rong Chai; Alejandro Aristizabal; Alexandros Karargyris; Hasan Kassem; Sarthak Pati; Micah Sheller; Nazanin Maleki; Rachit Saluja; Florian Kofler; Christopher G Schwarz; Philipp Lohmann; Phillipp Vollmuth; Louis Gagnon; Maruf Adewole; Li Hongwei B; Anahita Fathi Kazerooni; Nourel H Tahon; Udunna Anazodo; Ahmed W Moawad; Bjoern Menze; Marius G Linguraru; Mariam Aboian; Benedikt Wiestler; Ujjwal Baid; Gian-Marco Conte; Andreas M Rauschecker; Ayman Nada; Aly H Abayazeed; Raymond Huang; Maria Correia de Verdier; Jeffrey D Rudie; Spyridon Bakas; Evan Calabrese
The 2024 Brain Tumor Segmentation Challenge Meningioma Radiotherapy (BraTS-MEN-RT) dataset Article de journal
Dans: Sci Data, vol. 13, no 1, 2026, ISSN: 2052-4463.
@article{pmid41593091,
title = {The 2024 Brain Tumor Segmentation Challenge Meningioma Radiotherapy (BraTS-MEN-RT) dataset},
author = {Dominic LaBella and Katherine Schumacher and Michael Mix and Kevin Leu and Shan McBurney-Lin and Pierre Nedelec and Javier Villanueva-Meyer and David R Raleigh and Jonathan Shapey and Tom Vercauteren and Kazumi Chia and Marina Ivory and Theodore Barfoot and Omar Al-Salihi and Justin Leu and Lia M Halasz and Yury Velichko and Chunhao Wang and John P Kirkpatrick and Scott R Floyd and Zachary J Reitman and Trey C Mullikin and Eugene J Vaios and Ulas Bagci and Sean Sachdev and Jona A Hattangadi-Gluth and Tyler M Seibert and Nikdokht Farid and Connor Puett and Matthew W Pease and Kevin Shiue and Syed M Anwar and Shahriar Faghani and Peter Taylor and Pranav Warman and Jake Albrecht and András Jakab and Mana Moassefi and Verena Chung and Rong Chai and Alejandro Aristizabal and Alexandros Karargyris and Hasan Kassem and Sarthak Pati and Micah Sheller and Nazanin Maleki and Rachit Saluja and Florian Kofler and Christopher G Schwarz and Philipp Lohmann and Phillipp Vollmuth and Louis Gagnon and Maruf Adewole and Li Hongwei B and Anahita Fathi Kazerooni and Nourel H Tahon and Udunna Anazodo and Ahmed W Moawad and Bjoern Menze and Marius G Linguraru and Mariam Aboian and Benedikt Wiestler and Ujjwal Baid and Gian-Marco Conte and Andreas M Rauschecker and Ayman Nada and Aly H Abayazeed and Raymond Huang and Maria Correia de Verdier and Jeffrey D Rudie and Spyridon Bakas and Evan Calabrese},
doi = {10.1038/s41597-026-06649-x},
issn = {2052-4463},
year = {2026},
date = {2026-01-01},
journal = {Sci Data},
volume = {13},
number = {1},
abstract = {Meningiomas are the most common primary intracranial tumors, frequently requiring radiotherapy as a part of management. Effective radiotherapy planning for meningiomas necessitates accurate and consistent segmentation of target volumes on MRI, a process that is complex, labor-intensive, and dependent on expert expertise. The 2024 Brain Tumor Segmentation Challenge Meningioma Radiotherapy (BraTS-MEN-RT) Dataset addresses this problem by providing the largest multi-institutional collection of systematically annotated radiotherapy planning MRIs for meningiomas. Publicly accessible, this dataset comprises 570 radiotherapy planning 3D T1-weighted post-contrast MRIs at native resolutions, with 500 cases featuring expert-annotated gross tumor volumes (GTV). Annotations follow standardized radiotherapy planning protocols and include both intact and postoperative meningioma cases, ensuring wide clinical relevance. Contributions from seven diverse medical centers across the United States and the United Kingdom enhance the dataset's generalizability. The dataset aims to accelerate the development of automated segmentation methods for radiotherapy planning, improving workflow efficiency, reducing interobserver variability, and ultimately enhancing patient outcomes.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Louis Gagnon; Diviya Gupta; Uyen Nguyen; Maria Correia de Verdier; Rachit Saluja; George Mastorakos; Nathan White; Vanessa Goodwill; Carrie R McDonald; Thomas Beaumont; Christopher Conlin; Tyler M Seibert; Jona Hattangadi-Gluth; Santosh Kesari; Jessica D Schulte; David Piccioni; Kathleen M Schmainda; Nikdokht Farid; Anders M Dale; Jeffrey D Rudie
The University of California San Diego Post-Treatment Glioblastoma (UCSD-PTGBM) annotated multimodal MRI Dataset Article de journal
Dans: Sci Data, vol. 13, no 1, p. 183, 2026, ISSN: 2052-4463.
@article{pmid41495087,
title = {The University of California San Diego Post-Treatment Glioblastoma (UCSD-PTGBM) annotated multimodal MRI Dataset},
author = {Louis Gagnon and Diviya Gupta and Uyen Nguyen and Maria Correia de Verdier and Rachit Saluja and George Mastorakos and Nathan White and Vanessa Goodwill and Carrie R McDonald and Thomas Beaumont and Christopher Conlin and Tyler M Seibert and Jona Hattangadi-Gluth and Santosh Kesari and Jessica D Schulte and David Piccioni and Kathleen M Schmainda and Nikdokht Farid and Anders M Dale and Jeffrey D Rudie},
doi = {10.1038/s41597-025-06499-z},
issn = {2052-4463},
year = {2026},
date = {2026-01-01},
journal = {Sci Data},
volume = {13},
number = {1},
pages = {183},
abstract = {We present the University of California San Diego post-treatment glioblastoma (UCSD-PTGBM) annotated multimodal MRI dataset. The UCSD-PTGBM dataset includes 243 timepoints on 178 subjects with histopathologically-proven glioblastoma who were imaged with an advanced brain tumor protocol on 3 Tesla MRI scanners. Sequences include standard 3D imaging, as well as multishell diffusion (Restricted Spectrum Imaging, RSI) and perfusion imaging techniques (Arterial Spin Labelling, ASL and Dynamic Susceptibility Contrast, DSC), and neuroradiologist approved voxelwise tumor segmentations for both traditional segmentation masks and cellular tumor segmentations. The dataset also includes isocitrate dehydrogenase (IDH) mutation status and O6-methylguianine-DNA methyl-transferase (MGMT) promotor methylation status, as well as overall survival and progression free survival information for a subset of cases. We hope that researchers around the world will use these data to continue to improve analysis of post-operative MRI on glioblastoma patients, translate their findings into clinical practice and improve the management and outcome for these patients.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Louis Gagnon; Diviya Gupta; George Mastorakos; Nathan White; Vanessa Goodwill; Carrie R McDonald; Thomas Beaumont; Christopher Conlin; Tyler M Seibert; Uyen Nguyen; Jona Hattangadi-Gluth; Santosh Kesari; Jessica D Schulte; David Piccioni; Divya S Bolar; Anders M Dale; Nikdokht Farid; Jeffrey D Rudie
Dans: Neurooncol Adv, vol. 7, no 1, p. vdaf219, 2025, ISSN: 2632-2498.
@article{pmid41497450,
title = {Comparing the performance of dynamic susceptibility contrast and arterial spin labeling for detecting residual and recurrent glioblastoma with deep learning and multishell diffusion MRI},
author = {Louis Gagnon and Diviya Gupta and George Mastorakos and Nathan White and Vanessa Goodwill and Carrie R McDonald and Thomas Beaumont and Christopher Conlin and Tyler M Seibert and Uyen Nguyen and Jona Hattangadi-Gluth and Santosh Kesari and Jessica D Schulte and David Piccioni and Divya S Bolar and Anders M Dale and Nikdokht Farid and Jeffrey D Rudie},
doi = {10.1093/noajnl/vdaf219},
issn = {2632-2498},
year = {2025},
date = {2025-01-01},
journal = {Neurooncol Adv},
volume = {7},
number = {1},
pages = {vdaf219},
abstract = {BACKGROUND: Differentiating recurrent tumor from post-treatment changes remains a major challenge in glioblastoma (GBM) patients. In this work, we compared the performance of 2 different MR perfusion techniques, dynamic susceptibility contrast (DSC), and arterial spin labeling (ASL) to differentiate recurrent tumor and post-treatment changes from the volume of cellular tumor segmented from combined Deep Learning and multimodal MRI measurements, including multishell diffusion and perfusion.nnMETHODS: In this retrospective study, 137 MRIs from 107 patients with GBM were analyzed. Cellular tumor maps were segmented by 2 radiologists based on imaging, clinical history, and pathology. Multimodal MRI with perfusion and multishell diffusion were inputted into 5 nnU-Net Deep Learning models using either DSC or ASL with combination of multishell diffusion and standard MRI sequences to segment cellular tumor. Models with DSC and ASL were compared using segmentation performance (Dice score) and accuracy to detect recurrent tumor from post-treatment changes (area under the curve [AUC] under the receiver operating characteristic curve).nnRESULTS: Segmentation performances were similar in both cases, with a median Dice score of 0.75 (IQR: 0.53-0.84) for ASL and 0.76 (IQR: 0.57-0.84). AUC was 0.88 (CI 0.82-0.94) for ASL and 0.86 (CI, 0.80-0.92) for DSC, and this difference was statistically significant ( < .05, = 10 000 permutation test). In 11 individual cases, recurring disease was detected with ASL but missed with cerebral blood volume, including recurring tumor in the vicinity of a surgical cavity ( = 5), close to the skull base ( = 1), and adjacent to an Ommaya reservoir ( = 2).nnCONCLUSIONS: Our results demonstrate the utility of ASL in regions where susceptibility artifacts decrease the quality of DSC images.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Jordan Charest; Mathieu Walsh; Élie Genois; Emmanuelle Sévigny; Pierre-Olivier Schwarz; Louis Gagnon; Michèle Desjardins
Dans: NMR Biomed, vol. 37, no 12, p. e5252, 2024, ISSN: 1099-1492.
@article{pmid39245649,
title = {Comparison of compartmental analytical Blood-Oxygen-Level-Dependent functional Magnetic Resonance Imaging models against Monte Carlo simulations performed over cortical micro-angiograms},
author = {Jordan Charest and Mathieu Walsh and Élie Genois and Emmanuelle Sévigny and Pierre-Olivier Schwarz and Louis Gagnon and Michèle Desjardins},
doi = {10.1002/nbm.5252},
issn = {1099-1492},
year = {2024},
date = {2024-12-01},
journal = {NMR Biomed},
volume = {37},
number = {12},
pages = {e5252},
abstract = {Blood oxygen level-dependent functional magnetic resonance imaging (BOLD fMRI) arises from a physiological and physical cascade of events taking place at the level of the cortical microvasculature which constitutes a medium with complex geometry. Several analytical models of the BOLD contrast have been developed, but these have not been compared directly against detailed bottom-up modeling methods. Using a 3D modeling method based on experimentally measured images of mice microvasculature and Monte Carlo simulations, we quantified the accuracy of two analytical models to predict the amplitude of the BOLD response from 1.5 to 7 T, for different echo time (TE) and for both gradient echo and spin echo acquisition protocols. We also showed that accounting for the tridimensional structure of the microvasculature results in more accurate prediction of the BOLD amplitude, even if the values for SO were averaged across individual vascular compartments. A secondary finding is that modeling the venous compartment as two individual compartments results in more accurate prediction of the BOLD amplitude compared with standard homogenous venous modeling, arising from the bimodal distribution of venous SO across the microvasculature in our data.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Daniel Gourdeau; Olivier Potvin; Jason Henry Biem; Florence Cloutier; Lyna Abrougui; Patrick Archambault; Carl Chartrand-Lefebvre; Louis Dieumegarde; Christian Gagné; Louis Gagnon; Raphaelle Giguère; Alexandre Hains; Huy Le; Simon Lemieux; Marie-Hélène Lévesque; Simon Nepveu; Lorne Rosenbloom; An Tang; Issac Yang; Nathalie Duchesne; Simon Duchesne
Deep learning of chest X-rays can predict mechanical ventilation outcome in ICU-admitted COVID-19 patients Article de journal
Dans: Sci Rep, vol. 12, no 1, p. 6193, 2022, ISSN: 2045-2322.
@article{pmid35418698,
title = {Deep learning of chest X-rays can predict mechanical ventilation outcome in ICU-admitted COVID-19 patients},
author = {Daniel Gourdeau and Olivier Potvin and Jason Henry Biem and Florence Cloutier and Lyna Abrougui and Patrick Archambault and Carl Chartrand-Lefebvre and Louis Dieumegarde and Christian Gagné and Louis Gagnon and Raphaelle Giguère and Alexandre Hains and Huy Le and Simon Lemieux and Marie-Hélène Lévesque and Simon Nepveu and Lorne Rosenbloom and An Tang and Issac Yang and Nathalie Duchesne and Simon Duchesne},
doi = {10.1038/s41598-022-10136-9},
issn = {2045-2322},
year = {2022},
date = {2022-04-01},
journal = {Sci Rep},
volume = {12},
number = {1},
pages = {6193},
abstract = {The COVID-19 pandemic repeatedly overwhelms healthcare systems capacity and forced the development and implementation of triage guidelines in ICU for scarce resources (e.g. mechanical ventilation). These guidelines were often based on known risk factors for COVID-19. It is proposed that image data, specifically bedside computed X-ray (CXR), provide additional predictive information on mortality following mechanical ventilation that can be incorporated in the guidelines. Deep transfer learning was used to extract convolutional features from a systematically collected, multi-institutional dataset of COVID-19 ICU patients. A model predicting outcome of mechanical ventilation (remission or mortality) was trained on the extracted features and compared to a model based on known, aggregated risk factors. The model reached a 0.702 area under the curve (95% CI 0.707-0.694) at predicting mechanical ventilation outcome from pre-intubation CXRs, higher than the risk factor model. Combining imaging data and risk factors increased model performance to 0.743 AUC (95% CI 0.746-0.732). Additionally, a post-hoc analysis showed an increase performance on high-quality than low-quality CXRs, suggesting that using only high-quality images would result in an even stronger model.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Élie Genois; Louis Gagnon; Michèle Desjardins
Modeling of vascular space occupancy and BOLD functional MRI from first principles using real microvascular angiograms Article de journal
Dans: Magn Reson Med, vol. 85, no 1, p. 456–468, 2021, ISSN: 1522-2594.
@article{pmid32726489,
title = {Modeling of vascular space occupancy and BOLD functional MRI from first principles using real microvascular angiograms},
author = {Élie Genois and Louis Gagnon and Michèle Desjardins},
doi = {10.1002/mrm.28429},
issn = {1522-2594},
year = {2021},
date = {2021-01-01},
journal = {Magn Reson Med},
volume = {85},
number = {1},
pages = {456--468},
abstract = {PURPOSE: The vascular space occupancy (VASO) is a functional MRI technique for probing cerebral blood volume changes noninvasively, including during neuronal activation in humans. An important consideration when implementing VASO is the BOLD effect in the signal. Assessing the physical origin of this BOLD contamination and the capabilities of correction methods could improve the quantification of cerebral blood volume changes with VASO.nnMETHODS: Given the heterogeneity of cerebral microvascular architecture, the vascular geometry within an MRI voxel can influence both BOLD and VASO signals. To investigate this effect, 3D high-resolution images of mouse cerebral vasculature measured with two-photon microscopy were used to model BOLD and VASO signals from first principles using Monte Carlo diffusion of water protons. Quantitative plots of VASO together with intravascular and extravascular BOLD signals as a function of TE at B fields 1.5 T to 14 T were obtained.nnRESULTS: The BOLD contamination of the VASO response was on the order of 50% for gradient echo and 5% for spin echo at 7 T and TE = 6 ms and significantly increased with TE and B . Two currently used correction schemes were shown to account for most of this contamination and recover accurate relative signal changes, with optimal correction obtained using TEs as short as possible.nnCONCLUSION: These results may provide useful information for optimizing sequence parameters in VASO and BOLD functional MRI, leading the way to a wider application of these techniques in healthy and diseased brain.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
