Transformative Non-Invasive/Minimally Invasive Technologies for Imaging the Olfactory System Across Scales

September 19-20, 2024
Virtual Workshop

• Agenda
• Participant List

Workshop Summary

On this page:

• Introduction
• Executive Summary
• Conclusions
• Day One Presentation Summaries
• Day Two Presentation Summaries
• Acknowledgements
• Contact Information

Introduction

In vivo imaging of the human olfactory system and advanced image analysis techniques are critical to improving the clinical diagnosis of olfactory disorders and evaluating and optimizing potential treatments. Currently, there are limited non-invasive technologies available for high spatial and temporal resolution imaging of the living human peripheral and central olfactory system. Imaging the olfactory system in the laboratory and in the clinic is essential to understanding underlying biological mechanisms as well as the etiology of an olfactory disorder. Such imaging has proven challenging due to unique anatomical and functional characteristics such as limited accessibility, multiple cell types, and low signal resolution.

Addressing these challenges will require a multimodal approach including leveraging existing advanced tools and technologies and developing new tools tailored for the olfactory system. Recent advances in minimally invasive and non-invasive imaging technologies, such as cell specific biomarkers, fluorescence imaging, specialized endoscopes, optical and acoustic imaging, molecular imaging, and hemodynamic and diffusion imaging have the potential to transform our ability to image and investigate the peripheral and central olfactory system in the living human across scales/aspects with superior temporal and spatial resolution.

NIDCD, in collaboration with the National Institute of Biomedical Imaging and Bioengineering (NIBIB), hosted a two-day virtual workshop bringing together leaders in the fields of olfactory science and imaging tools and technologies to:

• Review the current state of the science.
• Identify gaps in knowledge.
• Explore new opportunities and challenges for imaging the olfactory system.

Executive Summary

Unique nature of the olfactory system

• Model system exploration: The olfactory system is seen as a unique model for understanding sensory processing, necessitating new frameworks for studying how the brain interprets olfactory signals.
• Understanding normal function: Gaining insights into the typical functioning of the olfactory system is essential for diagnosing changes associated with aging and disease.

Technological integration in olfactory research

• Advanced tools and techniques: The integration of optical and acoustic technologies with MRI and computational tools can provide new insights into how the olfactory system functions, both peripherally and centrally. These tools could enable detailed visualization and characterization of olfactory targets and pathways.
• Smart diagnostic systems: There is a need for developing sophisticated sensors and probes and other minimally invasive diagnostic tools to objectively assess olfactory status and dysfunction, moving beyond traditional methods (e.g., scratch and sniff).
• Multi-scale imaging: The field needs multi-scale studies that incorporate various levels of analysis, from microscopic to macroscopic scales, and a temporal dimension to understand olfactory changes over time.
• Spatial transcriptomics: Advanced techniques such as spatial transcriptomics offer the potential to characterize the human olfactory system at a molecular level. This approach could provide a more detailed understanding of the genetic underpinnings of olfactory dysfunction and the system’s response to aging or disease, offering a path toward personalized medical interventions.

Need for interdisciplinary collaborations

• Bridging gaps: The workshop emphasized the importance of integrating expertise from engineers and biologists to explore the olfactory system's complexities and share clinical and technological expertise. Collaborations between engineers, biologists, and clinicians are essential for developing innovative tools and diagnostic techniques. This could lead to significant advancements in both basic research and clinical applications, including early detection of neurodegenerative disease and olfactory dysfunction.
• Symposia and workshops: The need for organized events, such as symposia and workshops, was recognized in order to facilitate knowledge exchange between olfactory scientists and engineers, leveraging existing conferences to address challenges and innovations in the field.

Conclusions

Need for non-invasive high-resolution imaging of peripheral olfactory targets

Current methods for diagnosing the underlying pathophysiology of major forms of olfactory loss remain limited. Clinical practice primarily involves invasive endoscopic biopsy procedures to assess the olfactory epithelium within the nasal cavity. However, these biopsies come with several challenges, including invasiveness; difficulty in accurately sampling relevant regions; anatomical variability; and significant time, skill, and cost requirements. There is an urgent need for non-invasive, high-resolution techniques that allow real-time visualization of the entire olfactory epithelium in humans to help distinguish between peripheral and central causes of olfactory dysfunction.

Additionally, outside of measurements of smell function with psychophysical testing, there is no instrument for observing and quantifying peripheral changes associated with potential treatments of olfactory dysfunction. Advances in photoacoustic tomography, dynamic micro-optical coherence tomography, and fluorescence probes offer promising non-invasive approaches for imaging shallow peripheral targets including the nasal and olfactory epithelium, potentially reducing the need for invasive biopsies and improving diagnostic accuracy.

Need for non-invasive imaging providing cellular specificity

Visualizing the differences between the olfactory and non-olfactory epithelium in living humans remains a significant challenge. The mucosal epithelium contains specific enzymes that are unique to different epithelial types (e.g., respiratory vs. olfactory) and cell types (e.g., neuronal vs. supporting). These enzymes can convert non-fluorescent compounds into fluorescent forms, offering a potential approach for imaging the olfactory epithelium. Although animal studies have shown promise in fluorescent imaging of the olfactory epithelium, this technique has yet to be translated to humans.

A complementary technique, virtual staining using deep learning, shows great potential. This technology allows the generation of histological stains (including immunohistochemical stains) from label-free tissue images captured with autofluorescence microscopy, quantitative phase imaging, and reflectance confocal microscopy. Combining these deep learning methods with new fluorescence probes could enhance preclinical research and inform the development of diagnostic tools for olfactory loss.

Need for non-invasive high-resolution imaging of olfactory targets beyond the periphery

Olfactory dysfunction may also arise from changes in structures beyond the peripheral olfactory system. Recent advancements in brain imaging technologies show promise for visualizing targets in the central nervous system, such as the olfactory bulb. For instance, f/MRI is being optimized to study the unique laminar anatomy and layer-specific function of the olfactory bulb. Nevertheless, the bulb is challenging to image due to low signal intensity related to its small volume and deep location, and high levels of noise related to surrounding air pockets and bone. Brain imaging methodologies that are highly valuable to a holistic understanding of olfactory function, such as metabolic imaging, molecular imaging, and diffusion imaging, are yet to be explored. There are opportunities for custom development of these imaging methods to span a wide range of resolutions, sensitivities, coverage, and chemical specificity, which will be essential for the comprehensive study of the olfactory system.

Day One Presentation Summaries

Keynote

Bradley Goldstein, M.D., Ph.D., Duke University, “Imaging the Olfactory System Across Scales: Overview and Challenges”

Dr. Goldstein discussed the need for a better understanding of the cellular mechanisms and potential targets for the treatment of smell disorders. He detailed the anatomy of the olfactory system, noting variability in the number of glomeruli among individuals and that some may have a sense of smell despite lacking an olfactory bulb. He discussed techniques such as endoscopic brush biopsy and surgical biopsy for probing cell types within the nasal cavity. Despite the potential of these methods, several challenges remain, including the invasiveness of the techniques; individual anatomic variability; and the time, skill set, access, and cost required. Dr. Goldstein compared olfactory research to ophthalmology, which uses more advanced techniques, with better specificity and resolution, and emphasized that smell research could eventually attain the same level of advancement. To enable development/translation of effective therapies, several issues need to be addressed: what is being treated, the location of the damage/lesion, the cells that are impacted, and the signaling pathways or targets in these cells that could be modulated therapeutically.

Session 1: Endoscopy / Microscopy / Fluorescence Staining

Eric Holbrook, M.D., M.S., Massachusetts Eye and Ear, Harvard Medical School, Session Chair, “Utilizing Differential Expression of Cellular Enzymes to Visualize Olfactory Epithelium in Vivo”

Dr. Holbrook addressed the difficulties in imaging the olfactory epithelium in live humans, highlighting the need for effective techniques for diagnosis and monitoring. He discussed novel methods to selectively stain olfactory tissue that have been used successfully in mouse models but need further development for use in humans. He called for collaborations between biochemists and engineers to enhance imaging techniques for better diagnosis of olfactory dysfunction.

Lu Xu, Ph.D. | Stuart Firestein, Ph.D., Columbia University, “High-Speed Volumetric Imaging of the Olfactory System: From Periphery to Cortex”

Dr. Xu presented her research on using SCAPE (Swept Confocally-Aligned Planar Excitation) microscopy to study the olfactory system. She outlined two applications of the technique. The first involves imaging the olfactory epithelium in mice to visualize real-time calcium responses in sensory neurons, revealing complex odor encoding patterns that differ between single odors and mixtures. Another application examines central olfactory processing in salamanders, using in vivo imaging to analyze neuronal activity and morphology in response to odors. Dr. Xu emphasized the potential of SCAPE microscopy for simultaneous recording of large numbers of neurons, linking their activity with morphology across substantial brain areas.

Emily Gibson, Ph.D., University of Colorado, “Development of Multiphoton Endoscopy Towards Future Clinical Imaging of Olfactory Epithelium”

Dr. Gibson discussed multi-photon microscopy and its applications, particularly in clinical endoscopy. She highlighted its use in brain imaging, such as locating the subthalamic nucleus and distinguishing amyloid plaques through autofluorescence. Dr. Gibson's current research focuses on the olfactory epithelium and the development of miniaturized, fiber-tethered microscopes for use in freely moving animals, as well as adapting these technologies for clinical endoscopy. Future directions include a plan to combine advanced imaging capabilities with other modalities such as functional MRI.

Aydogan Ozcan, Ph.D., University of California, Los Angeles, “Virtual Staining of Label-Free Tissue Using Deep Learning”

Dr. Ozcan discussed the use of deep learning in microscopy for label-free staining of tissue, transforming autofluorescence images into synthetic stained images without chemical labels. This method effectively replicates tissue staining and allows for the creation of multiple virtual “stains” from a single tissue section, preserving samples for further analysis. Dr. Ozcan demonstrated application of deep learning in biopsy-free operations, enabling diagnostic-quality imaging in vivo. This advancement has the potential to enhance clinical pathology practices.

Session 2: Optical and Acoustic Imaging

Lihong Wang, Ph.D., California Institute of Technology, Session Chair, “Multiscale Photoacoustic Imaging from Organelles to Patients”

Dr. Wang presented on multi-scale photoacoustic imaging, which integrates light and sound for non-invasive visualization of biological structures, from organelles to patients. Initially demonstrated in small animals, this technology enables non-invasive imaging by detecting optical molecular absorption, providing advantages such as deeper tissue penetration and enhanced contrast compared to traditional high-resolution optical imaging methods such as two-photon microscopy. Recent innovations include handheld devices and improved imaging capabilities through skulls as well as real-time pathology without staining. The technique's versatility enables it to target various biomolecules for metabolic and molecular imaging, indicating broad potential applications, including imaging of olfactory structures.

Junjie Yao, Ph.D., Duke University, “From Light to Sound: Photoacoustic and Ultrasound Imaging of Deep Brains in Action”

Dr. Yao presented on various implementations of photoacoustic imaging, emphasizing its adaptability for different resolutions and penetration depths in brain studies. He discussed three technologies: ultrafast photoacoustic microscopy for rapid imaging of brain functions, a combination of ultrasound localization microscopy and photoacoustic imaging for deeper brain analysis, and a soft ultrasonic array for improved imaging through the skull. These technologies allow for high-resolution, high-speed imaging and potential therapeutic applications that can enhance the understanding of brain dynamics and olfactory processing.

Joseph Culver, Ph.D., Washington University in St. Louis, “Wearable Naturalistic Brain Imaging with Diffuse Optical Tomography”

Dr. Culver discussed the potential of diffuse optical tomography (DOT) as an alternative to fMRI in studying brain responses. He introduced the DOT systems and the latest wearable high-density DOT, which provides high-resolution imaging while allowing for real-time monitoring in various settings. The advantages of DOT include higher spatial resolution and reduced motion artifacts compared to more traditional functional near-infrared spectroscopy (fNIRS) systems. He addressed the need for naturalistic paradigms in neuroimaging and discussed how his lab uses movies as a step towards more immersive naturalistic tasks. He then discussed both previous fNIRS and fMRI research on olfaction. The fNIRS research demonstrates how natural it is to conduct olfaction behavioral tasks with optical brain measurements, though with limited imaging. More expansive fMRI research shows olfaction responses in the parietal cortex and inferior frontal, supramarginal, and angular gyri. These regions are readily accessible to DOT. He emphasized the need for context in olfaction research and how naturalistic DOT can enable behavioral paradigms that include context.

Guillermo (Gary) Tearney, M.D., Ph.D., Harvard Medical School, “Imaging the Nasal and Olfactory Epithelium with in Vivo Microscopy”

Dr. Tearney discussed optical coherence tomography (OCT), specifically micro-OCT, which offers cellular-level imaging with resolutions under 2 microns. This technology is used in various fields, including examining the organ of Corti for hearing-related research and assessing the human nasal epithelium in conditions like cystic fibrosis. A current advancement is dynamic micro-OCT, which captures real-time intracellular movements and metabolic activities. Dr. Tearney emphasized the potential of these techniques for studying the olfactory epithelium in vivo, providing label-free insights into tissue structure and function.

Session 3: Molecular Imaging Technologies

Robin de Graaf, Ph.D., Yale University School of Medicine, Session Chair, “MR-Based Imaging and Spectroscopy of the Olfactory System”

Dr. de Graaf discussed challenges in imaging and spectroscopy of the olfactory system— particularly the impact of air and bone on magnetic field homogeneity, which affects signal detection in both rodents and humans. Despite these challenges, studies show that robust BOLD activation can be achieved in the olfactory bulb in response to odorants, with unique neurochemical profiles identified through MR spectroscopy. Solutions such as advanced shim elements are being developed to improve magnetic field uniformity, and deuterium metabolic imaging (DMI) shows promise due to its insensitivity to field heterogeneity. Dr. de Graaf stressed that continued research is essential for advancing olfactory imaging techniques and understanding the olfactory system.

Ravinder Reddy, Ph.D., University of Pennsylvania, “Metabolite Weighted CEST Imaging of the Olfactory System”

Dr. Reddy described the chemical exchange saturation transfer (CEST) technique, which allows for measuring metabolites through magnetic resonance imaging, and discussed its use in studying glutamate levels in the context of schizophrenia and Alzheimer's disease. He discussed testing the glutamate hypothesis in schizophrenia using CEST to measure glutamate level (stylized as GluCEST), highlighting how elevated glutamate levels in the olfactory system may indicate risk for psychosis in clinically high-risk populations. Preliminary findings showed significant increases in glutamate in the olfactory cortex of at-risk youth compared to controls. Preliminary studies using GluCEST in a mouse model of Alzheimer's disease indicate elevated glutamate levels in the olfactory bulb, suggesting potential biomarkers for the neurodegenerative disease, though he noted that further research is needed to confirm these findings with larger cohorts.

Georges El Fakhri, Ph.D., DABR, Yale University School of Medicine, “High Resolution and Sensitivity Brain PET”

Dr. El Fakhri discussed the potential of using PET imaging in studying the olfactory system and its related disorders, specifically in the context of Alzheimer's disease. He highlighted advancements in PET technology that allow for high-resolution imaging of small brain structures, such as those associated with memory and olfactory function. Dr. El Fakhri introduced two new PET systems—the Savant and the Neuro Explorer—and shared their capabilities (compared to current PET systems) in achieving improved spatial resolution and sensitivity, enabling clearer visualization of brain structures. He underscored the importance of these advancements in understanding olfactory and memory-related disorders, emphasizing that these techniques can significantly enhance research opportunities in this field.

Naga Vara Kishore Pillarsetty, Ph.D., Memorial Sloan Kettering, “Non-Invasive Fluorescence Imaging of Na_v1.7 as a Diagnostic Tool for Anosmia”

Dr. Pillarsetty discussed the potential for voltage-gated sodium channel 1.7 (Na_v1.7) as a diagnostic tool for non-invasive imaging of anosmia. Na_v1.7 is linked to congenital insensitivity to pain and is expressed in both sensory and motor neurons. The co-localization of Na_v1.7 with olfactory markers in the mouse olfactory bulb indicates its role in olfactory function. Dr. Pillarsetty developed a peptide, derived from spider venom, that selectively targets Na_v1.7, therefore creating a fluorescent probe for imaging. Data from his team demonstrate that Na_v1.7 expression decreased in mice with methimazole-induced anosmia and in COVID-19 infected mice, correlating with a loss of olfactory function. His group also validated their findings in primates, finding that even after a few hours, the peptide was still bound in the olfactory bulb. Dr. Pillarsetty suggested that a Na_v1.7-targeting agent has potential for clinical use, providing semi-quantitative assessments of olfactory function and complementing traditional behavioral studies.

Day Two Presentation Summaries

Session 4: Electrophysiology, Hemodynamic, and Diffusion Imaging Technologies

Christina Zelano, Ph.D., Northwestern University, Session Chair, “Functional Neuroimaging and Electrophysiology of the Human Olfactory System”

Dr. Zelano presented research on the olfactory system, focusing on fMRI and a new intranasal technique to measure local field potentials from the olfactory bulb. Functional neuroimaging research showed that each cortical olfactory primary area—including the anterior olfactory nucleus, the olfactory tubercle, and frontal and temporal piriform cortices—forms distinct whole brain networks, suggesting that each may play a unique role. Ongoing research is exploring how odors are processed in various contexts, including social and feeding behavior. Dr. Zelano highlighted the development of a minimally invasive intranasal technique for recording robust local field potentials and oscillations that are visible on single trials in the olfactory bulb of awake behaving humans, and emphasized potential clinical applications in anosmia diagnosis and treatment.

Jun Hua, M.D., Ph.D., Kennedy Krieger Institute, Johns Hopkins University, “High Resolution Structural and Functional MRI of the Human Olfactory System”

Dr. Hua presented on developing MRI technology to study olfactory dysfunction in the brain. He discussed the challenges of imaging the olfactory epithelium and olfactory bulb due to artifacts caused by the air-tissue boundary. He presented improvements in structural and functional laminar MRI using a 7 Tesla system, allowing for better visualization of small structures. His lab has developed methods to obtain robust functional signals from the olfactory bulb, despite issues such as high noise levels and rapid signal habituation. Dr. Hua emphasized the need for cross-modal validation of his findings and presented a study with early Parkinson's disease patients showing unexpected increased response in the olfactory bulb, underscoring the clinical relevance of such research.

Anastasia Yendiki, Ph.D., Harvard Medical School, “New Frontiers in Imaging Human Brain Circuitry with Diffusion MRI and Microscopy”

Dr. Yendiki presented methods for reconstructing white-matter pathways with diffusion MRI (dMRI) tractography. Traditional methods involve manually or automatically drawing regions of interest (ROIs) to identify pathways, but if the tractography does not capture these pathways, accurate identification is very difficult. To address this challenge, Dr. Yendiki and her team developed TRACULA (Tracts Constrained by Underlying Anatomy), which integrates anatomical information directly into the tractography process. This method leverages high-quality dMRI data collected on the Connectome scanner at MGH to train TRACULA, allowing reconstruction of bundles in lower-quality data collected on widely available MRI scanners. Dr. Yendiki mentioned a new initiative, the Large-Scale Imaging of Neural Circuits (LINC), involving nine institutions in the imaging of cortico-subcortical connections in both humans and non-human primates. This project is funded by the NIH BRAIN Initiative.

Jay Gottfried, M.D., Ph.D., University of Pennsylvania, “Using Your Nose to Find Your Way”

Dr. Gottfried discussed the human olfactory system's role in guiding behavior and the importance of understanding natural sensory stimuli, particularly odors, in facilitating behaviors such as maternal bonding, hunting, and mating. He highlighted how the olfactory system processes a complex array of odor molecules. Focusing on human olfactory navigation, Dr. Gottfried described a study in which participants navigated a virtual space defined solely by odor cues, learning to associate and remember the spatial relationships of different smells. Over time, participants improved their navigation skills, demonstrating the ability to create a cognitive map based on olfactory information. The study also showed that specific brain areas—the entorhinal cortex, the piriform cortex, and the amygdala—are involved in olfactory processing.

Acknowledgments

Evan Wicker, Ph.D.; Merav Sabri, Ph.D.; Shumin Wang, Ph.D.; and Susan Sullivan, Ph.D., contributed to this workshop summary. The organizers would like to thank all speakers for their contributions to the workshop.

Contact Information

Please contact Merav Sabri, Ph.D., with questions related to this workshop.