2021 Grant Recipients

Winners of the 2023-2024 JustX Faculty Grants Program for Dartmouth faculty have been announced for one-year projects. The Neukom Institute received $100K in total requests and awarded $60K of financial support with an additional combination of programming support from Research Computing and the Neukom Scholars program.

Dartmouth College faculty including the undergraduate, graduate, and professional schools were eligible to apply for these competitive grants.

Note:

* indicates an award that is partnered with assistance from Dartmouth College Research Computing.

+ indicates an award that is partnered with an RA provided through the Scholars program.

Drone Lidar 2.0

casana_iraq2019_copy.jpg

Over the past decade, aerial lidar has become a transformative tool for archaeology, enabling the discovery and documentation of vast, previously unknown ancient cities, agricultural landscapes, and ritual installations around the world. Using a low energy laser to scan the surface of the earth, lidar produces high-resolution topographic data, penetrating tree canopy and other dense vegetation and helping to reveal otherwise hidden archaeological features. Yet most lidar data are collected by a costly aircraft-mounted instrument, making aerial lidar surveys extremely expensive to conduct and thereby limiting the ability of researchers to deploy this powerful technology. Thanks to rapid developments in compact, low-cost lidar systems alongside ever more sophisticated drone technology and processing software, it is now possible to collect very high-resolution lidar data using a drone-mounted sensor, but the utility of these new systems for archaeological research remains largely untested.

Building on previous Comp-X supported research, this project will use a new, ultra-compact DJI L1 lidar sensor deployed on the Matrice 300 drone to undertake lidar surveys at several key archaeological sites in forested regions.  We plan to use the lidar system to investigate 1) an ancestral Puebloan settlement at Picuris Pueblo, New Mexico, where stone built houses and agricultural terraces are hidden below pinon-juniper forest, 2) at the Menominee Reservation, Wisconsin, where remains ancient house basins, raised fields, and ceremonial mounds are present throughout forests, and 3) in the Upper Connecticut Valley in New Hampshire and Vermont where both historic and prehistoric archaeological features are obscured by mixed deciduous-evergreen forest.  The project will explore the opportunities and challenges of this new generation of drone lidar technology, revealing many previously undocumented archaeological features and facilitating a range of other projects by Dartmouth researchers.

Fast GPU-based Texture Analysis of FLASH Radiotherapy-Induced Skin Sparing

vdemidov_photo_copy.jpg

Radiation therapy (RT) is widely used to treat cancer, with about 50% of all cancer patients receiving RT during the course of their illness. While being highly effective, its outcomes are limited by radiation toxicity to surrounding healthy tissue. A few months ago, Norris Cotton Cancer Center at Dartmouth became one of the leaders in pioneering FLASH cutting-edge technology for cancer treatment improvement.

This CompX project will further develop a non-invasive imaging technique called optical coherence tomography (OCT) into a platform that can perform in-vivo measurements of skin microvascular remodeling following radiotherapy, detection of subtle structural skin changes, assessment of radiation-induced skin toxicity, label-free mapping of lymphatic and neural networks. During this project, we will enable fast computing of our recently developed skin image texture analysis algorithms by encoding them for processing on graphic processing units and integrating with commercial OCT software. This will allow for fast non-invasive detection and quantification of skin toxicity in preclinical studies, and be potentially translatable to clinical setting at Dartmouth using portable OCT technology.

Computational Modeling of Ambiguity Resolution Within and Across Individuals

esf_headshot1_copy.jpg

Different individuals, when faced with the same information, often arrive at different interpretations. This is particularly true for information that is complex, ambiguous, and/or social in nature. Resolving ambiguity is a core ability of human perceptual and cognitive systems, yet we do not fully understand how this process occurs on a behavioral and neural level, and how it varies across individuals.

This project will develop a framework to algorithmically generate animated videos that are intentionally ambiguous yet fully parameterized in their low-level visual features, allowing us to model how properties of both the stimulus and the observer affect perception. We will use these stimuli to build computational models of ambiguity resolution that combine biological readouts (i.e., brain imaging) with behavioral ones (i.e., natural language processing [NLP] on elicited speech) to shed light on which features of a stimulus beget ambiguity, and how these features interact with individuals' neural processes to lead different people to draw different conclusions.

Elucidating the Structure and Function of Top-down and Bottom-up Connectivity in the Neocortex

rhill_copy.jpg

The neocortex is the brain area most associated with perception, cognition, and the generation of voluntary behavior. While much is known about the cellular structure of the neocortex, little is known about the neuron subtypes that contribute to long-distance connectivity among hierarchically organized cortical areas subserving distinct sensory and motor modalities. While the existence of these intracortical axonal projections is well established, and often classified as top-down or bottom-up connectivity, the specificity by which different neuron subtypes participate in each connection is less understood. Moreover, defined functional roles for this neural circuitry are not clear. Our project combines advanced optical imaging, cellular labeling, and optogenetic manipulations with electrophysiological recordings to elucidate the structure and specificity of these intracortical connections and reveal how they contribute to conscious states including wakefulness and different forms of sleep. Once we identify the cellular subtypes and projection specificity involved in top-down and bottom-up connections, future projects will use optogenetic and chemogenetic approaches to manipulate these connections in behaving animals to test their roles in perception and behavior. Thus, this project will uncover additional patterns of cellular connectivity in the cortex and the functional output of this neural circuitry at the organismal level. 

Deciding synaptic transmission with light

picture1_copy.png

The long slender signaling projection (axon) of a neuron, when seen under a microscope, is abundantly decorated with specialized swellings called presynaptic terminals. Presynaptic terminals are full of organelles, the majority of which are packets of chemical neurotransmitter called vesicles. The fusion of a vesicle and release of chemical neurotransmitter known as synaptic transmission is what underlies essential functions of our nervous system from sensing pain to forming a memory.  Furthermore, the dysfunction of synaptic transmission is causally linked to several neurological disorders including Alzheimer's disease. Amazingly, all of this happens in a structure less than 1/1000th the width of a human hair, making it very difficult to study in a live cell.

Our proposal seeks to develop the most sensitive microscopy approaches to date capable of accurately measuring the fusion of single vesicles at individual synapses in live neurons using novel fluorescent indicators. Second, we hope to deploy computational approaches to resolve the location of vesicle fusion at the nanoscale level of the synapse. By combining these two measurements we hope to uncover physiological processes that counteract age and disease-related neuronal perturbations, enabling brain resilience. Going forward, we plan to use this support from the Neukom to leverage these tools for studying the molecular underpinnings of this resiliency with an eye towards developing therapeutic targets.

Optimizing Schools? The Empirics and Ethics of Algorithmic Prioritization in K-12 Schooling

rjohnson_headshot_copy.jpeg

As governments turn to algorithms to help them decide whom to help and whom to punish, recent work has investigated implications for inequality. K-12 schools are beginning to use large-scale data and predictive models to allocate help to students, raising questions specific to this institutional domain. The present research uses a combination of methodologies---a survey experiment; computational text analysis; analysis of digital trace data---to investigate how algorithms in K-12 schooling intersect with longstanding racial and socioeconomic inequalities.

The first component involves investigating perceptions of algorithmic prioritization in K-12 schooling among three sets of stakeholders: members of the general public; K-12 teachers; and current parents of school-age children. The Neukom CompX award will be used to enhance the representativeness of the teacher and parent sample as well as to improve open-source tools for analyzing latent moral themes in respondents' free-text responses about the relative fairness of the methods. The second component involves using large-scale metadata and click data to investigate K-12 educators' use of a dashboard displaying both algorithmic assessments of student need and non-algorithmic assessments (e.g., raw test score histories). While the first two components focus on formal prioritization policies that districts adopt, the third component will be a virtual workshop focused on using new forms of large-scale text and "digital trace data" to study how educators implicitly prioritize between families and students. Overall, the Neukom CompX award will aid in both research on and tools for studying the fairness of pre-algorithmic and algorithmic prioritization in education. 

Towards a Digital Corpus of Corinthian Vase Inscriptions

oppen.jpg

Although classics and related fields have long relied on print publications of inscriptions, the cost of acquiring such publications often means that scholars at all but the most well-off institutions do not have access to such data around which they might base further research. EpiDoc, a consortium of guidelines and tools for the digital publication of inscriptions (and ancient texts more generally), is the main computational method this project will use to address this problem. Specifically, with the support of a Neukom CompX Grant and the Dartmouth Department of Classics, this project will begin digitally publishing the corpus of Corinthian vase inscriptions while also creating funded research opportunities for Dartmouth students.

Beginning in Fall 2021, the project will extend existing efforts to contemplate ancient writing beyond the world of elite literary production in Latin and ancient Greek courses at Dartmouth by assigning students in such courses Corinthian vase inscriptions in either language to transcribe. In Winter 2022, leading experts in digital epigraphy will lead a five-day training in EpiDoc open to all Dartmouth students and faculty. Building on abilities and interest generated from both Latin and ancient Greek courses and this training, we will digitally publish Corinthian vase inscriptions using EpiDoc in Spring 2022, culminating in both a webpage and a collaborative print article for publication in a peer-reviewed journal.

Snap, Crackle, Pop: Quantifying Frost Cracking in the Arctic

20180202_palucis_eb_015_copy.jpg

meyer_photo_copy.jpg

                                                                                                                                                                                                                                                                    We hypothesize that rates of sediment generation from bedrock erosion are increasing with warming in periglacial watersheds. The rationale is that warming has resulted in larger and more frequent bedrock landslides and rockfall events due to 1) increased physical weathering (i.e., warmer winters leading to sustained temperature conditions in the frost weathering 'window') and 2) liquid water availability (due to warmer and more extensive summers), leading to increased sediment production.

We are testing whether rates of sediment production have increased since the last glacial maximum (LGM) within two small watersheds along a climate gradient within the Aklavik Range (NWT, Canada). To determine current rates of bedrock erosion, however, requires annual repeat high-precision bedrock mapping. Neukom funding will enable us to purchase a research-grade terrestrial Light Detection and Ranging (LiDAR) scanner (i.e., TLS) with sub-cm accuracy and imaging capabilities. Using the two-way travel time and speed of light, LiDAR determines the travel distance and ultimately a three-dimensional model (DEM) of an area can be created.

Total Body Visualization of Radiation Dose with Computer Animation

brian2019b_copy.jpg

This multi-disciplinary project will advance the tools for computational visualization of radiation dose in radiation therapy treatment of lymphoma.  The project will use body surface position scripting software for computer animation of the radiation dose given to patients in different body positions.  The tools for analysis of the display will be developed, to quantify the accuracy & of treatments, allowing the clinical team an intuitive way to interpret the whole body radiation dose to each part of the skin.