Public health threats often trigger narratives of “othering" that cast blame upon marginalized groups and stoke further stigmatization. The record increase in hate crimes targeting Asian Americans since the onset of the pandemic is alarming but yet not surprising. As tragic as these events are, even those hate crimes that are reported to police likely represent just a subset of incidents ranging from the mundane to the extreme that Asian Americans have faced. According to AAPI Data, though Asian Americans have experienced hate incidents more than the general population during this pandemic, they are also among the least likely to say they are “very comfortable'' reporting hate crimes to authorities. Systematically documenting everyday manifestations of anti-Asian sentiments is immensely challenging.

Continue reading

Decoding behavioral signifiers for the brain state of vigilance can have far reaching implications for understanding the neural basis for actions and identifying disease. We are using high resolution video recordings of mice as they navigate a maze but have access to very few pre-determined behavioral signifiers. Several recent publications implemented computer vision to extract a variety of previously unreachable aspects of behavioral analysis, including animal pose estimation and distinguishable internal states. These descriptions allowed for the identification and characterization of dynamics, which then revealed an unprecedented richness to the behaviors that determine decision making. Applying such computational approaches mice during exploration and in the context of behaviors that have been validated to measure choice and memory can reveal dimensions of behavior that predict or even determine psychological constructs like vigilance. We are also obtaining neural signal data, which can be aligned with the behavioral signifiers. DSI scholars would use pose estimation analysis to evaluate behavioral signifiers for choice and memory and relate it to our real time concurrent measures of neural activity and transmitter release. The students would also have opportunity to examine the effect of disease models known to impair performance on our maze task on any identified signifier.

Continue reading

Columbia University Data Science Institute is pleased to announce that the Data Science Institute (DSI) and Data For Good Scholars programs for Fall 2022 are open for application.

The goal of the DSI Scholars Program is to engage Columbia University’s undergraduate and master’s students in data science research with Columbia faculty through a research internship. The program connects students with research projects across Columbia and provides student researchers with an additional learning experience and networking opportunities. Through unique enrichment activities, this program aims to foster a learning and collaborative community in data science at Columbia.

The Data For Good Scholars program connects student volunteers to organizations and individuals working for the social good whose projects have developed a need for data science expertise. As “real world” problems with real world data, these projects are excellent opportunities for students to learn how data science is practiced outside of the university setting and to learn how to work effectively with people for whom data science sits outside of their subject area.

Continue reading

This project is the first comprehensive examination of African North Americans who crossed one of the U.S.-Canada borders, going either direction, after the Underground Railroad, in the generation alive roughly 1865-1930. It analyzes census and other records to match individuals and families across the decades, despite changes or ambiguities in their names, ages, “color,” birthplace, or other details.

Continue reading

Tumor segmentation and size assessment are of great clinical value in cancer diagnosis and treatment monitoring. As an example, tumor size is used to monitor neoadjuvant chemotherapy (NAC) response. As the NAC changes the stiffness (i.e. elasticity) of the tumor, elasticity imaging can be used to monitor the NAC response. Single transducer harmonic motion imaging (ST-HMI) is an ultrasound elastography method to assess elasticity of tissue. In ST-HMI, a force from the ultrasound propagating wave is used to oscillate tissue at a particular frequency and another ultrasound beam is used to assess the oscillatory motion. Then, the elasticity of the tissue is inferred from the motion. Instead of collecting a single frequency oscillation at a time, ST-HMI is expanded to collected several frequencies of data in a single acquisition (i.e. multi-frequency ST-HMI). We have demonstrated that lesions with different sizes and stiffness can be detected by exploiting oscillation frequency without prior knowledge of the inclusion characteristics. However, the boundary delineation is performed manually. The objective of this work is to develop an artificial intelligence algorithm to delineate tumor boundaries in the multi-frequency ST-HMI images automatically. A large data set of multi-frequency ST-HMI images of phantom with and without the presence of noise and artifacts, in vivo, and ex vivo mouse tumors, and in vivo human breast tumors will be used to train and test the artificial intelligence algorithm. By learning these multi-frequency data with a variety of noise sources, the constructed network is expected to learn boundary effects and be able to separate tumor boundaries from the background.

Continue reading

This multi-scale, multi-modal study aims to characterize the development of executive functions (EF) over the 1st 1000 days of life across diverse cultures and geographies. The international and multidisciplinary team brings together experts in global child health, EFs neurodevelopment, neuroimaging, nutrition, genomics, and the microbiome to realize several ambitious goals. The goals of the project are: i) rich longitudinal characterization of the structural and functional development of brain networks that underlie EFs. We will use behavior in combination with: A. state-of-the-art and multi-modal high-density EEG and MRI; and B. scalable low-field portable MRI and lower-density EEG to establish utility worldwide and particularly in global regions with limited resources; ii) rich longitudinal characterization of environmental influences on structural and functional development of brain networks that underlie EFs. Complementing the imaging portion of the study, the multimodal multiscale assessments include state-of-the-art deep phenotypic measurements including comprehensive surveys, sleep wearables, psychosocial and sociodemographic contextual factors, and biospecimen data to characterize health and environmental factors. The student(s) will be required to rapidly become familiar with a variety of data types, e.g., questionnaire data, physiological high-dimensional signals, neurobehavioral clinical examinations, neurodevelopmental assessments, etc. This DSI project will focus on the deep phenotyping of sleep in early life by analyzing objective and subjective measures. From a methodological point of view, the student(s) will support the development of: i) pipelines for the standardized and automated preprocessing and cleaning of physiological signals collected during sleep; ii) machine learning techniques (data-drive clustering, prediction) and iii) advanced statistical methodologies (inference, mediation, multidimensional data integration) tailored to the investigation of the role of sleep in relationship to the emergence of EF. The student(s) will join a young and highly dynamic multidisciplinary team of experts, students, and RAs (some who originally joined the lab as DSI-students).

Continue reading

Neurodevelopmental disorders (NDDs) comprise of a group of disorders associated with abnormal brain development. Rare genetic variants have been shown to play a key role in their development, especially in those NDDs which are severe in nature. During the last decade, genetic testing has emerged as an important etiological diagnostic test for NDDs with a considerable impact on disease management and treatment. Yet, current genetic testing has a diagnostic rate of ~ 50%. Due to technical limitations in modern next-generation sequencing techniques, these techniques fail to asses a large part of the genome (2/3rd), missing critical regions which may have clinical significance. New methods now have emerged that can assess these regions better, can access repetitive regions and identify complex structural genomic events with more accuracy. This project will employ and integrate novel genomic technologies, including optical genome mapping and long read sequencing, to perform a comprehensive investigation of the human genome in parent-child trios which remained genetically unsolved after standard genomic approaches.

Continue reading

Author's picture

Columbia Data Science Institute (DSI) Scholars Program

The DSI Scholars Program is to engage and support undergraduate and master students in participating data science related research with Columbia faculty. The program’s unique enrichment activities will foster a learning and collaborative community in data science at Columbia.

Columbia University DSI

New York, NY