Stroke is devastating when left untreated. Early treatment significantly improves outcomes, and can be reliably detected using the FAST exam, which specifically assesses facial asymmetry, arm weakness, and speech deficits for signs of stroke. Our project aims to use smartphone technology to build a stroke detection algorithm based on this exam. We have collected video data of hospitalized stroke patients performing aspects of the standard neurology exam, which includes the factors previously mentioned. The next step is to build an algorithm that can detect facial asymmetry using this data.

Continue reading

Stroke is devastating when left untreated. Early treatment significantly improves outcomes, and can be reliably detected using the FAST exam. The exam looks for facial asymmetry, arm weakness, and speech deficits as signs of stroke. Our research project aims to use smartphone technology to build a stroke detection algorithm based on the FAST exam. We have collected speech data of hospitalized stroke patients. The next step is to build an algorithm that can detect abnormal speech.

Continue reading

While an normal semen sample contains hundreds of millions of sperm, men with azospermia have no sperm seen and azospermia represents an important cause of infertility. Men with azospermia typically have to use donor sperm in order to have a child. However, extended manual search of semen samples in men with azospermia reveal single-digit numbers of sperm in ~80% of cases. These few sperm are sufficient to fertility eggs in the IVF lab and allow the men to be biological fathers. Unfortunately, the manual search for sperm is only performed in a couple of sites in the world because of the enormous cost associated with having a specially trained andrologist search a sperm sample for roughly 8 hours. In this project, we aim to apply the semen sample to the custom designed microfluidics chip at a fixed flow rate. Next, several thousands of images will be captured at frame rate ranging between 4,000-40,000 fps using a high definition microscope camera as the sample flows through the microfluidics device. We will then build a training dataset of images with and without the rare cells of interest. These training datasets will be used to build an artificial intelligence (AI) model with high accuracy to detect rare cells of interest. The model will be tested for its accuracy on the ‘true’ sample containing rare cells. The Data Science Institute Scholar is expected to have basic programming skills (preferably in Python). Thus, if successful, there will literally be countless babies born as a result of this project.

Continue reading

We performed a prospective, non-randomized trial of patients undergoing surgical management of gastrointestinal, gynecologic, and urologic malignancies between 2017 and 2021 . Participants were instructed to take immunonutrition formula three times daily for five days before and after surgery once cleared by the surgeon for clear liquid diet. The dietician tracked their adherence. We are looking to perform statistical analysis to look at the impact of immunonutrition intake on unadjusted and adjusted postoperative surgical outcomes.

Continue reading

The Center for Behavioral Cardiovascular Health has been at the forefront of developing virtual interventions (e.g., home BP telemonitoring, iHeart enhanced depression screening app, and COVID remote care program) to improve the management of hypertension, depression, and COVID. We are looking for motivated mentees to help us organize and analyze a wide-breadth of patient data from the Columbia-New York Presbyterian data warehouse to understand the impact of the interventions on patient outcomes. The mentee would work alongside highly experienced biostatisticians and our Center’s professionalized data team managers. The mentee will develop expertise in analyzing health system data and will inform decisions to modify, sustain, or de-implement existing programs.

Continue reading

The aims of this project are two-fold. We seek to: 1) better understand the associations of discrimination with sleep and cardiovascular health among LGBTQ+ adults, and 2) compare various unsupervised learning techniques on their performance on clustering sleep patterns using actigraphy data. We have multiple datasets for conducting our analyses to delineate the pattern of associations between sleep and other variables. We will also explore various modeling techniques to quantify the within- and between-person variability in sleep patterns. We aim to complete multiple analyses in the Fall and Spring semesters to develop conference abstracts and manuscripts to be submitted in Spring 2022. Since our DSI seed grant is still ongoing, during the Fall semester the Scholar will work on data wrangling and analyses, in addition to data management from existing data collected from our Precision in Symptom Self-Management pilot studies. In January 2022, we will begin analyzing data from our DSI seed grant. We hope to collaborate with a student who is interested in machine learning, sleep, and/or health disparities. The Scholar will have the opportunity to contribute to all of these aspects of the project. The Scholar will also be able to propose and lead additional analyses.

Continue reading

This project is a collaboration between Columbia scholars and students and leadership at the Malcolm X & Dr. Betty Shabazz Center. The goal is to create programming and intellectual resources inspired by the legacy of Dr. Shabazz and Malcolm X through public events, workshops and data science activities. Part of this work will involve using computational methods to analyze existing publications around Malcolm X’s work, as well as develop a corpus of research materials by effectively digitizing the contents of the publicly-available FBI files about his life.

Continue reading