HVB and common words are captured during spoken and imagined language while participants wear EEG. By comparing EEG signals recorded during these two conditions, the study aims to identify distinct neural activation patterns associated with each. Understanding these differences could contribute to advancements in brain-computer interfaces.

Optical coherence tomography (OCT) image analysis plays a crucial role in detecting and monitoring retinal diseases such as diabetic macular edema and age-related macular degeneration. In this project, we developed a retinal layer segmentation algorithm for OCT images, organizing the retina into four primary layer groups. To improve segmentation accuracy, we designed new preprocessing techniques, including background adjustment, adaptive denoising, and rotational correction. We then built an interdependent segmentation model using Otsu’s method, edge detection, and morphological operators. To evaluate the effectiveness of the segmented layers, we extracted statistical and texture-based features to support further analysis.

Ambulatory EEG has become a valuable diagnostic tool for neurological disorders such as epilepsy, but conventional EEG caps with wired electrodes are intrusive and socially conspicuous, limiting their suitability for continuous daily monitoring. This project explores conductive tattoo ink as a possible replacement for traditional EEG electrodes and wiring, aiming to make long-term data collection less obtrusive and more practical for everyday use, similar to how continuous glucose monitors have transformed diabetes care.

Using PEDOT:PSS-based ink formulations, including low-impedance, high-impedance, and modified high ion concentration inks, the study evaluates key electrical properties such as voltage, current, and resistance as a function of trace length over a 24-hour drying period. Results show that conductivity decreases with increasing trace length, with the higher ion concentration formulation demonstrating improved electrode performance. Stereomicroscopic imaging was also used to assess ink deposition and confirm feasible conductive pathways within synthetic skin.

Future work will compare tattoo-based EEG recordings with conventional EEG systems and evaluate performance on curved scalp geometries, with the long-term goal of supporting more practical seizure prediction technology.

Inflammatory breast cancer (IBC) is a rare and aggressive clinicopathologic form of breast cancer. It has a roughly 55% 5-year survival rate, much lower than most other types of breast cancer, which is about a 90% 5-year survival rate in the United States. The American Joint Committee on Cancer (AJCC) defines the IBC with erythema (redness), edema, and peau d’orange over at least a third of the breast—often with no underlying tumor mass—with a duration of the first symptom to the diagnosis fewer than six months. In addition, the histopathology of tumor emboli in the breast biopsy indicates IBC. The diagnosis of IBC based on the AJCC criteria is challenging because the presence and extent of erythema, edema, and peau d’orange vary between IBC patients, and there are no solid pathologic criteria to confirm the diagnosis of the diseases. In addition, the AJCC clinical signs are not present in many IBC cases.