top of page

STEM Research 

SPACE WEATHER

Solar/Geomagnetic Activity & Space Weather Effects on Earth

MACHINE LEARNING

Automated Multiclass & Binary Classification of Premalignant Cervical Cells in Pap Smear Images Using a Support Vector Machine.

500_F_170545079_tmApvIqff2YnrQStHlMWpwpM

BREAST CANCER 

Effects of Single-Walled Carbon Nanotubes on Triple Negative Breast Cancer Cell Cytoskeleton

500_F_324564700_SKr91RFcJuk26znR4mNyfs40

SIGNAL PROCESSING

Signal Processing & Data Analysis from an Implantable Bioelectrical Medical Device

Machine Learning

MACHINE LEARNING

New York Institute of Technology | Bioengineering Master's Thesis | 2020

A master's thesis research and literature review focused on the "Automated Multiclass & Binary Classification of Premalignant Cervical Cells in Pap Smear Images Using a Support Vector Machine (SVM)". The study was primarily based on supervised learning, image processing using edge detection.

Performed segmentation and extraction of cervical cell nuclei of over 393 pap smear images using an Active Contour Model (ACM).

Wrote an SVM programme on MATLAB for the multiclass classification of the four classes of cells with an accuracy of over 80% based on the ground truth.

Classification Types

Binary Classification: Normal Cells and Abnormal Dysplasia.

Four-Class Multiclass Classification: Normal Cells, Mild, Moderate and Severe Dysplasia.

Software & Tools Used:

MATLAB

Active Contour Model for segmentation

Support Vector Machine for classification 

Relevance of Research

One key difference between normal and abnormal (precancerous) cervical cells is in the size of the nucleus, with the latter having an enlarged size.

This study can potentially reduce the strenuous work load of lab cytologists who examine and classify cervical cells through microscopic lens which can be prone to human error. With the use of a trained machine learning model or classifier, an automated detection of abnormal cells through image analysis can be achieved, thereby reducing the margin for human error.

Breast Cancer

BREAST CANCER

City University of New York - Queensborough Community College CUNY | Academic Research | 2018/2019

Conducted a one-year research study on the effects of single-walled carbon nanotubes (SWCNT) on triple negative breast cancer (TNBC) cells of line MDA- MB-468.

Presented a poster on the successful demonstration of the effects of SWCNT on the TNBC cells in reducing cell migration.

Skills, Equipment & Softwares Used

Mammalian cell culturing & optimization, migration assays, shake flasks, etc.

Confocal Microscopy

T-Scratch - to measure distance in cell migration assay.

Relevance of Research

According to the World Health Organization, breast cancer is the top cancer among women globally and it remains on the rise. This pilot study was based of the promise of targeted drug delivery (TDD) in cancer therapy through the use of nanotechnology.

 

If nanoparticles such as SWCNT can be transported directly and effectively to the site of the cancer tumor of a living organism (before metastasis) for administration, it could reduce the need for chemotherapy and radiation therapy which can have dire side effects.

Watch Dr. Elizabeth Wayne's Ted talk here.

Space Weather

City University of New York - Queensborough Community College CUNY | Academic Research | 2018

Co-authored a publication on the effects of the September 6, 2017 solar event that occurred in tandem with Hurricane Irma, such as a shortwave radio blackout across Europe and Africa.

Team member of a 4-person academic research class on the collation and data analysis of solar/geomagnetic activity in space weather and the impact on life on Earth.

Partly funded by the National Aeronautics & Space Administration (NASA).

Relevance of Research

Believe it or not but there's a weather in outer space and it can have a huge impact on our everyday lives, our own weather and even biological systems.

The same way we have weather storms on Earth, there are also storms that occur in space, and a big one took place on September 6th, 2017. Most people didn't realize the impact of this solar storm because Hurricane Irma was also in action. As a result of this X-class solar flare, countries in certain regions experienced high-frequency radio blackouts for up to an hour, i.e affecting essential communication in services such as ambulances. 

By studying solar signatures, observing the behavior and patterns of the sun, we can forecast solar storms and electromagnetic activity and prepare for impact on Earth. Same way you'd like to know if it will rain this week. 

Read my publication here. Also, find some NASA info here

SPACE WEATHER

Analyzed signals from the simulation of a wireless embedded chip/implantable stomach pacemaker that generates slow waves in the gastrointestinal (GI) tract to treat related disorders.

A supervised academic research.

Software & Tools Used:

MATLAB.

Relevance of Research 

We often take for granted the innumerable daily functions that our biological systems carry out in order to keep up with our daily lives. One of these important functions is the ability of our stomach to generate slow waves. Slow waves are rhythmic electrophysiological waves that are generated in our Interstitial cells of Cajal which are integral to our digestive system by triggering contraction and relaxation of the muscles as food travels through our gastrointestinal (GI) tract.

Simply put, for individuals with certain digestive health issues or GI related disorders, a stomach pacemaker could be a solution as it helps simulate the bioelectrical currents that trigger these slow waves similar to that of a healthy biological system.

Signal Processing

New York Institute of Technology | Academic Research | 2019

SIGNAL PROCESSING

bottom of page