FDFrédérick Dallaire

Frédérick Dallaire

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Frédérick Dallaire has a total of 150 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Label-free wide-field imaging of brain tumors using near-infrared endogenous fluorescence and reflectance normalization

    by Gabriel Beaudoin, Victor Blanquez-Yeste, Frédérick Dallaire, Seung Soo Lee, Romain Cayrol, Marie-Christine Guiot, Mark A. Trifiro, Miltiadis Paliouras, Kevin Petrecca, Frédéric Leblond

    SignificanceMaximizing safe tumor resection remains a major challenge in brain tumor surgery due to the lack of reliable real-time intraoperative contrast between tumor and healthy brain tissue. Label-free optical methods based on endogenous tissue fluorescence could provide simple, noninvasive feedback to guide resection while avoiding the logistical and regulatory limitations of exogenous fluorophores.AimThe study aims to investigate whether endogenous near-infrared (NIR) fluorescence can differentiate tumor from normal brain tissue in vivo and to develop a wide-field fluorescence and reflectance imaging system for label-free visualization of brain tumor specimens ex vivo.ApproachWe analyzed in vivo data acquired with a hand-held near-infrared spectroscopy probe from 26 patients (430 measurements) to quantify endogenous fluorescence differences between glioblastoma or astrocytoma and normal brain. Based on these findings, we built a wide-field imaging prototype using 808-nm laser excitation, red-shifted fluorescence detection with an InGaAs short-wave infrared camera, and co-registered reflectance imaging for ratiometric normalization. The system was validated with ex vivo human surgical specimens (14 samples from five patients).ResultsEndogenous fluorescence intensity measured in vivo was significantly higher in normal brain tissue compared with tumor, yielding positive predictive values of 85 to 88% for tumor and a negative predictive value of 96% for normal tissue. Ex vivo wide-field imaging reproduced this contrast: normalized fluorescence was lower in tumor regions than in histologically confirmed normal areas, consistent with metabolic differences observed in vivo.ConclusionsEndogenous NIR fluorescence provides intrinsic contrast between normal and tumoral brain tissue. The demonstrated corrected wide-field fluorescence imaging approach could offer a feasible, label-free method for real-time visualization of tumor margins, supporting its future clinical translation as an intraoperative guidance tool.

  • Combining label-free Raman spectroscopy with machine learning to monitor COVID-19 disease from acute infection to recovery

    by Frédéric Leblond, Frédérick Dallaire, Guillaume Sheehy, Hugo Soudeyns, Sze Man Tse, Caroline Quach, Daniel E. Kaufmann, Katherine Ember, Mathieu Dehaes

    SignificanceMonitoring COVID-19 disease from acute infection to recovery is critical to understand biochemical dysregulation and COVID-19 heterogeneity over time.AimOur aim is to develop an approach combining label-free Raman spectroscopy and machine learning modeling to enable sensitive biomolecular detection of COVID-19 over time.ApproachHospitalized patients infected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) were recruited and stratified based on respiratory support (critical and non-critical). Controls had a negative SARS-CoV-2 test. Blood was collected in the acute and recovery phases and was analyzed with Raman spectroscopy. Four machine learning models based on Raman spectra were developed to differentiate critical and non-critical patients in the acute and recovery phases from controls. For each group of patients, two additional models also classified the patient status (acute versus recovery) using cross-sectional and longitudinal analyses.ResultsRaman peaks assigned to proteins, glucose, fatty acids, lactic acid, vitamin A, and lipids were identified in models. Overall, area under the receiver operating characteristic curve values were between 0.83 and 1.00 with sensitivities, specificities, and accuracies between 73% and 100%, 77% and 100%, and 78% and 100%, respectively.ConclusionsThese results highlight the capability of combined Raman spectroscopy and machine learning modeling to stratify patients at admission, monitor recovery after discharge, and support strategies to potentially reduce the risk of long-COVID.

  • Open Access

    Raman Spectroscopy Combined with Machine Learning Reveals Myalgic Encephalomyelitis–Associated Biomolecular Signatures at Rest and After Standardized Stress

    by Maryam Heidarifard, Atefeh Moezzi, Frédérick Dallaire, Katherine Ember, Wesam Elremaly, Iurie Caraus, Anita Franco, Frédéric Leblond, Alain Moreau, Mathieu Dehaes

    Myalgic encephalomyelitis (ME) is characterized by profound fatigue, post-exertional malaise (PEM), and cognitive dysfunction. Despite its clinical significance, the pathophysiology of PEM and disease heterogeneity remain unclear, and no validated biomarkers are available for rapid diagnosis or monitoring. We aimed to develop a screening approach combining label-free Raman spectroscopy (RS) and machine learning modeling (ML) to detect biomolecular changes in blood plasma and differentiate patients with ME from sedentary healthy controls. Blood plasma was collected from 115 patients with ME and 45 controls at rest (T0) and 90 min after a standardized, non-invasive stress test designed to induce PEM. Plasma samples were analyzed by RS, and ML models were developed independently at each time point to differentiate patients with ME and controls. The RS-ML models identified spectral features consistent with contributions from proteins, lipids, and low-molecular-weight metabolites. At T0 and T90, the area under the receiver operating characteristic curve, accuracy, specificity and sensitivity were 0.85 and 0.83, 79% and 84%, 82% and 90%, and 73% and 69%, respectively. RS-ML provides a rapid, low-cost approach to detect ME-associated biomolecular signatures in plasma and capture biochemical alterations associated with standardized stress.

  • Open Access

    Combining label-free Raman spectroscopy and machine learning to identify early biomarkers of COVID-19 disease severity and mortality

    by Maryam Heidarifard, Katherine Ember, Frédérick Dallaire, Elsa Brunet-Ratnasingham, Yiheng Chen, Nassim Ksantini, Myriam Mahfoud, Guillaume Sheehy, Hugo Soudeyns, Philippe Jouvet, et al.

    Early prediction of COVID-19 severity and mortality is crucial for optimizing clinical care and patient outcomes, but remains challenging. We aim to develop a screening tool combining label-free Raman spectroscopy and machine learning modeling to predict COVID-19 severity and mortality. Patients infected by SARS-CoV-2 (N=58) were recruited during the first wave of COVID-19 and stratified based on respiratory support. Blood samples were collected during hospitalization and analyzed using Raman spectroscopy and metabolomics. Machine learning models based on Raman spectra were developed to classify (1) survivors versus nonsurvivors, (2) critical patients with noninvasive versus invasive ventilation, and (3) noncritical (no respiratory support or oxygen via nasal cannula) versus critical patients. Raman peaks assigned to proteins, glucose, lactic acid, fatty acids, urea, and lipids were extracted by the models. Area under the receiver operating characteristic curve ranged between 0.83 and 0.94, with sensitivities and specificities ranging between 80% and 83% and 75% and 92%, respectively. Accuracy for detecting mortality, invasive ventilation, and critical disease was 90%, 87%, and 78%. A complementary metabolomic analysis confirmed some molecular differences between groups. These results suggest the potential of Raman spectroscopy and machine learning modeling to stratify COVID-19 patients at admission, individualize care, and improve survival rates.


  • Open Access

    Deployment of a real-time prostate cancer confirmation system with Raman spectroscopy: fine-tuning versus test-time adaptation of 1D CNNs

    by David Grajales, William T. Le, Victor Blanquez-Yeste, Frédérick Dallaire, Feryel Azzi, Guila Delouya, Dominique Trudel, Frédéric Leblond, Cynthia Ménard, Samuel Kadoury

    SignificanceProstate cancer (PCa) confirmation during needle-based procedures is limited by the lack of intraoperative diagnostic tools. Raman spectroscopy (RS), combined with classification models, offers a promising solution for real-time tissue characterization, potentially improving sampling accuracy and therapy guidance. However, such models require tissue- and organ-specific data, making deployment in studies challenging due to limited data availability.AimThe aim is to develop a one-dimensional convolutional neural network (1D-CNN) for real-time PCa detection using RS on prospectively collected ex vivo data, leveraging multi-organ pre-training and evaluating two domain adaptation strategies.ApproachA ResNet-based 1D-CNN was trained for binary cancer/normal tissue classification. We implemented a pre-training strategy using retrospective RS data from brain, breast, and prostate (202 patients), along with pre-trained bacterial models, followed by efficient fine-tuning and test-time adaptation (TTA) to adapt to unseen domains.ResultsProspective RS data were acquired using a robotic system from 10 PCa patients (two to five biopsies each). The fine-tuned model achieved 0.76 area under the receiver operating characteristic curve, 0.79 accuracy, 0.83 sensitivity, and 0.72 specificity, outperforming support vector machines. TTA improved predictions when labels were unavailable.ConclusionsPre-trained 1D-CNNs combined with efficient fine-tuning or TTA enable accurate PCa detection in small-cohort settings using real-time RS.


  • Open Access

    Development and preclinical evaluation of an endonasal Raman spectroscopy probe for transsphenoidal pituitary adenoma surgery

    by Victor Blanquez-Yeste, Félix Janelle, Trang Tran, Katherine Ember, Guillaume Sheehy, Frédérick Dallaire, Eric Marple, Kirk Urmey, Frédéric Labidi, Frédéric Leblond

    Several tumors originate in the pituitary gland, which regulates hormonal balance and controls endocrine functions. This gland is located in the sellar region at the base of the brain. Pituitary adenomas represent ∼15% of all intracranial tumors and can have several consequences on human health, including local nerve compression, visual impairment, endocrine dysfunctions, and diseases, such as acromegaly, gigantism, or Cushing’s disease.1 The symptoms are contingent upon the size of the tumor and whether it secretes hormones.1,2 Endoscopic transsphenoidal resection surgery is a common treatment approach, during which pituitary adenomas and other tumors of the sellar region are removed through the nasal cavity and the sphenoid sinuses, using a rigid endoscope.3 The endoscope is used for visualization, while the patient is under general anesthesia. During the procedure, the anterior wall of the sphenoid sinus is identified and removed. Subsequently, the surgeon removes any remaining septations within the sinus and exposes the sella turcica (ST), the bony cavity that houses the pituitary gland. The bony wall of the sella (sellar floor) is meticulously removed with drills or punches.4 Subsequently, the dura mater is incised to expose the pituitary gland and the tumor. The tumor is then removed in pieces using microinstruments or suction devices. Great care is taken to preserve the normal pituitary tissue and to avoid damage to the surrounding structures, including the optic nerves superiorly and the carotid arteries laterally. This approach has become the standard for most pituitary adenomas and lesions confined to the sella turcica because of its precision, safety, and minimal invasiveness.5


  • Open Access

    Single-cell Raman spectroscopy detects pediatric focal cortical dysplasia

    by Trang Tran, Frederick Dallaire, Joshua Sonnen, Romain Cayrol, Frédéric Leblond, Roy W. R. Dudley

    More than 50 million people worldwide suffer from epilepsy with approximately half being children.1 Over one-third of this population does not respond to medications, leaving surgical resection the only option for seizure control. Seizure onset is focal for 60% of all drug-refractory cases, and surgery can be curative when the epileptogenic zone (EZ) is removed completely.2,3 The most common cause of focal epilepsy in children is focal cortical dysplasia (FCD), with FCD type II being the most prevalent.2 FCD is classified into different neuropathological subtypes based on the degree of cytoarchitectural disruption visible in histology. In type I, the cortical layers are disorganized (dyslamination) without abnormal or dysmorphic cells.4 Type II is characterized either by large dysmorphic neurons without balloon cells (type IIa) or with balloon cells (type IIb). Type III exhibits cortical lamination abnormalities akin to type I, yet they are associated with other pathologies: hippocampal sclerosis (type IIIa), tumors (type IIIb), vascular malformations (type IIIc), and acquired lesions including strokes or previous trauma (type IIId).5

  • Open Access

    Quantitative assessment of the generalizability of a brain tumor Raman spectroscopy machine learning model to various tumor types including astrocytoma and oligodendroglioma

    by Frédéric Leblond, Frédérick Dallaire, Katherine Ember, Alice Le Moël, Victor Blanquez-Yeste, Hugo Tavera, Guillaume Sheehy, Trang Tran, Marie-Christine Guiot, Alexander G. Weil, et al.

    SignificanceMaximal safe resection of brain tumors can be performed by neurosurgeons through the use of accurate and practical guidance tools that provide real-time information during surgery. Current established adjuvant intraoperative technologies include neuronavigation guidance, intraoperative imaging (MRI and ultrasound), and 5-ALA for fluorescence-guided surgery.AimWe have developed intraoperative Raman spectroscopy as a real-time decision support system for neurosurgical guidance in brain tumors. Using a machine learning model, trained on data from a multicenter clinical study involving 67 patients, the device achieved diagnostic accuracies of 91% for glioblastoma, 97% for brain metastases, and 96% for meningiomas. Here, the aim is to assess the generalizability of a predictive model trained with data from this study to other types of brain tumors.ApproachA method was developed to assess the generalizability of the model, quantifying performance for tumors including astrocytoma, oligodendroglioma and ependymoma, pediatric glioblastoma, and classification of glioblastoma data acquired in the presence of 5-ALA induced fluorescence. Statistical analyses were conducted to assess the impact of vibrational bands beyond contributors identified in our previous research.ResultsA machine learning brain tumor detection model showed a positive predictive value (PPV) of 70% for astrocytoma, 74% for oligodendroglioma, and 100% for ependymoma. Furthermore, the PPV was 100% in classifying spectra from a pediatric glioblastoma and 90% for detecting adult glioblastoma labeled with 5-ALA-induced fluorescence. Univariate statistical analyses applied to individual vibrational bands demonstrated that the inclusion of Raman biomarkers unexploited to date had the potential to improve detectability, setting the stage for future advances.ConclusionsDeveloping predictive models relying on the inelastic scattering contrast from a wider pool of Raman bands may improve detection accuracy for astrocytoma and oligodendroglioma. To do so, larger tumor datasets and a higher Raman photon signal-to-noise ratio may be required.


  • Open Access

    Macroscopic inelastic scattering imaging using a hyperspectral line-scanning system identifies invasive breast cancer in lumpectomy and mastectomy specimens

    by Sandryne David, Hugo Tavera, Tran Trang, Frédérick Dallaire, François Daoust, Francine Tremblay, Lara Richer, Sarkis Meterissian, Frédéric Leblond

    SignificanceOf patients with early-stage breast cancer, 60% to 75% undergo breast-conserving surgery. Of those, 20% or more need a second surgery because of an incomplete tumor resection only discovered days after surgery. An intraoperative imaging technology allowing cancer detection on the margins of breast specimens could reduce re-excision procedure rates and improve patient survival.AimWe aimed to develop an experimental protocol using hyperspectral line-scanning Raman spectroscopy to image fresh breast specimens from cancer patients. Our objective was to determine whether macroscopic specimen images could be produced to distinguish invasive breast cancer from normal tissue structures.ApproachA hyperspectral inelastic scattering imaging instrument was used to interrogate eight specimens from six patients undergoing breast cancer surgery. Machine learning models trained with a different system to distinguish cancer from normal breast structures were used to produce tissue maps with a field-of-view of 1 cm2 classifying each pixel as either cancer, adipose, or other normal tissues. The predictive model results were compared with spatially correlated histology maps of the specimens.ResultsA total of eight specimens from six patients were imaged. Four of the hyperspectral images were associated with specimens containing cancer cells that were correctly identified by the new ex vivo pathology technique. The images associated with the remaining four specimens had no histologically detectable cancer cells, and this was also correctly predicted by the instrument.ConclusionsWe showed the potential of hyperspectral Raman imaging as an intraoperative breast cancer margin assessment technique that could help surgeons improve cosmesis and reduce the number of repeat procedures in breast cancer surgery.


  • Towards Real-Time Confirmation of Breast Cancer in the OR Using CNN-Based Raman Spectroscopy Classification

    by David Grajales, William Le, Frédérick Dallaire, Guillaume Sheehy, Sandryne David, Trang Tran, Frédéric Leblond, Cynthia Ménard, Samuel Kadoury