Research

Quantifying pathological changes to myelin with high-resolution birefringence microscopy and deep learning
RA: Alex Gray
Myelin is an essential neuronal structure that enables rapid propagation of nerve impulses and plays a vital role in sensory, motor and cognitive function in the human brain. Insults to myelin structure, and its parent oligodendrocytes, is prevalent in disease and has been correlated to impairments in both cognition and motor function. Despite interest in studying and quantifying changes to myelin structure, current methods for myelin imaging are insufficient for large-scale characterization of various disease models. Birefringence microscopy (BRM) is a camera-based imaging technique that enables high-resolution imaging of myelin structure in postmortem tissue sections. BRM has previously been applied for investigating changes in myelination in a rhesus monkey model for induced cortical injury but has yet to be applied for larger-scale studies of disease. My research aims to expand the utility of BRM for high-throughput imaging and quantification of myelin pathology. I am working to utilize BRM and deep learning techniques to quantify the structural integrity of myelin in a rhesus monkey model of age-related cognitive decline and circumscribed cortical injury. The completion of this work will establish a framework for large-scale imaging and quantification of myelin pathology in the context of disease. This will also demonstrate the ability to evaluate the efficacy of treatments that target the restoration and regrowth of myelin following injury or in disease.

 

Volumetric assessment of cerebral myelin in Alzheimer’s disease and Chronic Traumatic Encephalopathy using polarization-sensitive optical coherence tomography and birefringence microscopy

RA: Anna Novoseltseva
Alzheimer’s disease (AD) and Chronic Traumatic Encephalopathy (CTE) are two incurable neurodegenerative disorders characterized by unique patterns of hyperphosphorylated tau accumulations, which form neurofibrillary tangles. These tangles are associated with progressive axonal degeneration, and are accompanied by myelin degradation, which remains underexplored in AD and CTE pathogenesis. Recent studies have suggested that myelin abnormalities may play a significant role in the pathogenesis of these diseases and significantly contribute to cognitive decline. In this study, we aim to investigate myelin changes in AD and CTE using advanced imaging techniques and deep learning algorithms. Specifically, we will use birefringence microscopy (BRM) and polarization-sensitive optical coherence tomography (PSOCT) to study the changesin myelin morphology associated withdifferent stages of the diseases. We will also develop a deeplearning neural network model to automatically identify and quantifymyelin defects, facilitatingthestudy of their relationship with disease progression relative to the spreading of tau and amyloid pathology. Three specific aims: Aim 1 involves using BRM and serial-sectioning PSOCT to examine the myelin morphology in postmortem brain tissue samples from individuals with age-matched normal control, and late-stages of AD and CTE. Aim 2 involves developing a deep learning neural network(Faster R-CNN based model)to automatically detect different types of myelin defects in brain tissue samples. Aim 3 involves examining the relationship between myelin changes and disease progression relative to tau and amyloid pathology spreading. The outcomes of this study will provide insights into the underlying pathologies of AD and CTE and will help to identify potential targets for therapeutic interventions.

 

Increasing understanding in the role of cerebrovascular remodeling in Alzheimer’s disease and chronic traumatic encephalopathy using birefringence microscopy
RA: Meadow Bradsby
I am working on using birefringent microscopy to understand the impact of some neurodegenerative diseases (Alzheimer’s disease and chronic traumatic encephalopathy) on the structure of cerebral arteries. I have been collaborating with other labs at BU that are studying cerebral arteries using mechanical testing, multiphoton imaging, histology, and mass spectrometry.  The end goal of this collaboration is to build a more complete understanding of the relationship between cerebral arteries and neurodegenerative disease.

 

Multispectral Colonoscopy Imaging using Five Narrowband Wavelengths of Light

RA: Ethan Espinoza

As of 2020, colorectal cancer (CRC) was the second leading cause of death in the United States. Colonoscopic surveillance is a common procedure for screening and diagnosis of CRC. Colonoscopy can identify patients at higher risk for developing metachronous lesions based on pathologic review of polypectomy specimens.  An adenoma is the most conventional identification of colorectal carcinoma tissues and served as the basis for polypectomy surveillance guidelines. However, at least 20% of colorectal carcinomas occur not through adenomas, but through serrated polyps. Approximately 17-24% of polyps are missed during colonoscopy using white light. These missed lesions are often attributed to characteristics such as size and sessile, or flat, morphology. Currently, colonoscopy uses white light which only provides surface morphology of the rectal wall and does not resolve abnormal architecture and subsurface microvasculatureNew methods of imaging these polyps are required to obtain better results. One approach includes the use of several narrowband wavelengths of light. These different wavelengths are capable of differentiating between neoplastic and other types of polyps, such as non-neoplastic polyps, and allow for better characterization of polyps that ultimately affect the decision regarding polyp removal during colonoscopy.

Narrowband imaging (NBI) techniques rely on the properties of hemoglobin as a major tissue chromophore. NBI coupled with magnifying endoscopy has been shown to reveal subepithelial microvascular architecture and microsurface structure within the GI mucosa. NBI is thus able to reveal tissue features that are not seen with white light; this can include architectural features such as blood vessels from different depths using shorter wavelengths. Use of longer wavelengths, such as 650 nm and near-infrared (NIR) will penetrate deeper into the mucosal tissue, showing more detailed structure. Since the gastrointestinal tract is mainly composed of blood vessels and mucosa, NBI is able to separate these two layers well by changing the wavelengths and exploiting the absorption and scattering of hemoglobin. Use of different wavelengths enhances the visibility of vessels and other structures on or near the mucosal surface.