Advances in data collection technologies in neuroscience has resulted in a deluge of high-quality data that needs to be analyzed, and presented to the experimentalist in a meaningful way. Usually the “data analysis and visualization”-pipeline is built from scratch for each new experiment resulting in a significant amount of code duplication and wasted effort in rebuilding the analysis tools. There is a growing need for a unified system to automate much of the repetitive tasks and aid biologists in understanding their data more efficiently.

Continue reading

We aim to augment recovery in spinal cord (SC) injured patients. Electrical stimulation of the SC can facilitate recovery, but the mechanisms are not yet understood. One knowledge gap lies in the exact pathways that are recruited by stimulation. To close this gap, we have tested the effects of SC stimulation in people undergoing clinically indicated surgery. By testing the distribution and size of muscle responses to SC stimulation, we can infer which circuits are activated. We are also examining how SC injury changes those responses. We propose to use Bayesian methods to understand the interaction between muscle responses to stimulation and the MRI indicated pattern of damage. The project will involve construction of models linking multiple data modalities that predict muscle activity, followed by the modification of these models to account for patterns of damage. Construction of such models would enable a deeper understanding of SC stimulation leading to more effective stimulation paradigms.

Continue reading

Our goal is to use deep learning networks to understand which neurons in the brain encode fine motor movements in mice. We collected large datasets entailing calcium imaging data of active neurons and high-resolution videos when mice perform motor tasks. We want to use recent advances in deep learning to (1) estimate the poses of mouse body parts at a high spatiotemporal resolution (2) extract behaviorally-relevant information and (3) align them with neural activity data. Behavioral video analysis is made possible by transfer learning, the ability to take a network that was trained on a task with a large supervised dataset and utilize it on a small supervised dataset. This has been used e.g. in a human pose–estimation algorithm called DeeperCut. Recently, such algorithms were tailored for use in the laboratory in a Python-based toolbox known as DeepLabCut, providing a tool for high-throughput behavioral video analysis.

Continue reading

This position is associated with a new research project co-funded by the Earth Institute Frontiers seed funding and the UN Development Programme in Guinea. Its goal is to develop remote sensing technologies to track environmental impact from bauxite mining in rural communities and to help establish protocols for their use. We seek a research assistant who can (1) develop image classification to be applied to satellite imagery to determine extent and abundance of bauxite dust within specific communities in the Boké region (2) develop methods to make classified images available for download to smartphones in the field and (3) adapt existing application technologies to allow users in the field to upload field photographs and locations of potential impacts they identify. The resulting map would represent a composite of satellite imaged areas of impact verified in the field and documentation of other areas of impact not visible by satellite but already identified in the field.

Continue reading

Research on: (i) COSMOS cloud connected vehicles, (ii) Monitoring of traffic intersections, using bird’s eye cameras, supported by ultra-low latency computational/communications hubs; (iii) Simultaneous video-based tracking of cars and pedestrians, and prediction of movement based on long-term observations of the intersection; (iv) Real-time computational processing, using deep learning, utilizing GPUs, in support of COSMOS applications; (v) Sub-10ms latency communication between all vehicles and the edge cloud computational/communication hub, to be used in support of autonomous vehicle navigation. The research is performed using the pilot node of project COSMOS infrastructure.

Continue reading

The objective of this project is to construct linkages across disparate public health data systems using machine learning tools and assess them for bias and equitable representation of subpopulations defined by demographic and socioeconomic factors.

Continue reading

42% of New York City greenhouse gas emissions result from on-site fossil fuel combustion in residential and commercial buildings; space heating is, by far, the majority contributor. Both New York State and NYC have policies to dramatically reduce emissions that will require a transformation in the way buildings are heated, including major efforts in existing buildings. This transition is inextricably linked to existing energy equity issues that we believe significantly overlap across NYC (and elsewhere). These include unreliable heating in the winter, susceptibility to extreme heat (an increasing occurrence with climate change) and struggles to afford energy needs. Various known data sources for NYC are available, though they are disparate and have not been analyzed holistically. Further, we believe there are potential engineering and policy solutions to these challenges. In this project, the DSI scholar will access (and search for where not yet known to qSEL researchers) relevant data sets, analyze those data sets to identify communities exposed to all or a subset of these issues, and assist qSEL researchers in developing models to evaluate possible solutions. The project has the possibility of extending through Summer 2020, subject to fundraising efforts and the success of the Spring 2020 project.

Continue reading