Projects

Current Research Projects

Project 1: Neuroimaging Markers for Predicting the Outcome of Brain Tumor Surgery

Research Project Leader: Dr. Han Yuan. Mentors: Dr. Michael Wenger and Dr. Ian Dunn.

Illustration of multimodal imaging. (A) showing rsfMRI connectivity in two patients seeded at tumor, near tumor and contralateral to tumor, obtained from our pilot data at OUHSC. (B) nTMS map in a patient. Orange triangles indicate the points with motor response recorded in the nTMS mapping. Red circles indicate points of motor response in direct cortical electrical stimulation

The objective of this proposal is to develop an intelligent and multimodal strategy for identifying and predicting plasticity based on images of brain connectivity that relates to the neurological deficits due to brain tumor surgery in the patients with focal brain gliomas involving motor or language regions. Success of this project will demonstrate feasibility and advantages of this new strategy, and provide important preliminary data to support applying for a more comprehensive research project (i.e., NIH R01) to further optimize and validate this new strategy and multimodality imaging markers of plasticity, which can be leveraged into surgery planning to improve overall survival of patients by increasing the extent of tumor resection without compromising patient safety or long-term functional outcome.

Neuroimaging Markers for Predicting the Outcome of Brain Tumor Surgery-Abstract (.pdf)
More details can be read by downloading the above PDF file of the project abstract.

Project 2: New Ultrastructural 3D Optical Imaging of Tumor Endothelium for Cancer Nanomedicine Development

Research Project Leader: Dr. Stefan Wilhelm, Mentors: Dr. Wei Chen and Dr. Priyabrata Mukherjee.

The objective of this project is to demonstrate the feasibility of applying a novel 3D super-resolution optical imaging in combination with label-free scattered light imaging to visualize and quantify the intracellular nanoparticle transport in breast cancer associated endothelial cells, which enables researchers to better understand the intracellular pathways that the nanoparticles take during transcytosis. If successful, this project will provide essential preliminary study data to support a future NIH R01 grant application to further investigate and improve therapeutic efficacy of using nanoparticles for optimal drug delivery to treat breast cancers.

Figure on the left illustrates how nanoparticles extravasate from tumor blood vessels via transcytosis. Transmission electron micrographs (TEM) of breast cancer blood vessel lumen surrounded by endothelial cells (rose) and tumor tissue (grey). Nanoparticles (black dots) are seen in intracellular compartments in endothelial cells (blue arrowheads, a, b) and exit endothelial cells to transport into tumor tissue (red arrows, b). Bar: 2 µm.

New Ultrastructural 3D Optical Imaging of Tumor Endothelium for Cancer Nanomedicine Development -Abstract (.pdf)
More details can be read by downloading the above PDF file of the project abstract.

Project 3: Early Stage Evaluation of Ovarian Cancer Prognosis by Fusing Radiographic and Histopathologic Imaging Information

Research Project Leader: Dr. Yuchen Qiu, Mentors: Dr. Javier Jo and Dr. Robert Mannel.

Illustration of tumor segmentation (left), extraction of regions of interest from digital histopathology images (middle) and architecture of a deep learning network for the epithelium and stroma classification.

The overarching objective of this project is to demonstrate the feasibility of developing a novel decision-making support tool based on the fusion of prognostic information extracted from radiology (CT) and digital histopathology images, which can more accurately stratify ovarian cancer patients for receiving optimal chemotherapy. If successful, this project can produce the essential preliminary study data to support applying for a more comprehensive research project (i.e., NIH R01) to further optimize and validate this new decision-making support tool to help clinicians determine the optimal cancer treatment strategy for different patients.

Early Stage Evaluation of Ovarian Cancer Prognosis by Fusing Radiographic and Histopathologic Imaging Information-Abstract (.pdf)
More details can be read by downloading the above PDF file of the project abstract.

Project 4: Use of 3D Quantitative Optical Methods to Characterize Mebendazole Treatment of Ovarian Cancer

Research Project Leader: Dr. Lauren Dockery, Co-investigator: Dr. Qinggong Tang, Mentors: Dr. Hong Liu and Dr. Kathleen Moore.

Illustration of (A) a microstructural OCT image of tumor with tumor margin delineated and (B) a cross-sectional OCT image from the green dashed line in A (left), and a schematic diagram of a proposed endoscopic OCT/FLOT system (right), where FC: fiber coupler; PC: polarization controller; C: collimator, BD: balanced detector, MZI: Mach-Zehnder interferometer (frequency clocks), DAQ: data acquisition board, M: mirror, GSM: galvonometer scanning mirror, O: objective lens, LD: laser diode, P: polarizer, S: shutter, I: iris, CL: cylindrical lens, FW: filter wheel, GRIN: gradient Index lenses. DM: dichroic mirror.

The objective of this project is to investigate the feasibility of applying novel medical imaging technology to facilitate the discovery of new quantitative imaging markers and assessment methods to characterize the anti-cancer effects of a repurposed drug, mebendazole, for the effective treatment of recurrent ovarian cancer. The information gained from this project will add to our understanding of the anti-cancer activity of this promising drug, specifically regarding its effects on the tumor microenvironment, and will support future clinical studies by applying for a new NIH R01 project.

Use of 3D Quantitative Optical Methods to Characterize Mebendazole Treatment of Ovarian Cancer-Abstract (.pdf)
More details can be read by downloading the above PDF file of the project abstract.

Pilot Projects

Pilot Project 1: Novel Imaging Methods for Monitoring Peptide-Aggregation Induced Immunotherapy for Breast Cancer

Illustration of peptide-aggregation induced immunotherapy.

Illustration of [II] (0.4 mg/mL) treatment induces LDH release (A) and extracellular ATP release (B). Propidium iodide uptake of [II] treated OVCAR-8, B16F10, Panc02 and 3T3 cells for 6h (C). HMGB-1 and HSP90 secretion of [II] treated cells at 6h. The concentration optimizations for [II] were performed on EMT6 spheroids (E), green = live cells, red = dead cells (F).

The objective of this proposal is to test our recently developed immunotherapy tool, peptide-aggregation induced immunogenic membrane rupture technology, in breast cancer. The effectiveness will be monitored via a novel quantitative 3D imaging platform. We aim to identify the efficacy of our tool with in vitro breast cancer spheroid models and in vivo immunocompetent mouse models, which will guide our efforts to optimize the conditions for better imaging and higher effectiveness for faster clinical translation.

Specific Aim of Pilot Project 1.

Pilot Project 2: Fabrication of Hexagonal Boron Nitride-Graphene Quantum Dots Nanocomposites based Surface-Enhanced Raman Scattering Probes for Cancer Cell Imaging

Illustration of schematic representation of the enhanced Raman signal due to HBN-GQDs nano composite..

The research objective of this proposal is to elucidate how novel HBN-GQDs nanocomposites can be fabricated via electrostatic assembly. The new knowledge of their fluorescent property, stability, water solubility, and cytotoxicity will facilitate the creation of novel SERS platform. Overall, discovering a precise processingstructure–property relationship will enable SERS to become a powerful tool for cancer imaging. In pursuit of this goal, novel SERS labels based on hexagonal boron nitride HBN-GQDs nanocomposites will be developed for fast cellular probing and imaging.

Specific Aim of Pilot Project 2.

Pilot Project 3: Cancer Prognosis Prediction by Integrating Clinical, Imaging, and Multi-omics Data Using Deep Learning

Illustation of an interpretable deep learning framework.

Summary of dataset (left) and proposed interpretable deep learning framework (right) including (a) VAE for latent representative feature extraction from the 𝑖𝑖-th imaging or genomic data; and (b) Concatenated latent feature and clinical data for risk of progression. DeepSHAP was used to interpret influential factors.

Accurate and reliable cancer prognosis prediction is essential and of great practical value to provide decision support on personalized and precision therapy for both cancer patients and healthcare providers. Although te plea for multimodal integrative analysis is urgent to permit a more accurate prediction of cancer patient’s prognosis and complete understanding about the interplay of the predictive risk factors, models integrating all types of the data and their complex interactions are not sufficiently investigated in existing cancer prognosis prediction studies. Thus, the objective of this project is to develop an accurate, robust, and interpretable framework for the prediction of cancer patient’s progression by integrative analysis of clinical records, histopathology image, and multi-(ge)omics profile.

Specific Aim of Pilot Project 3.

In summary, the medical imaging COBRE currently includes 4 research projects and 3 pilot projects that develop and/or apply the advanced medical imaging modalities and quantitative imaging markers to investigate broad research topics for promoting translational cancer research in two OU campuses.