Department of Computer Science

The study of the brain is a compelling example of the power of convergent science. Over the last few decades, advances in neuroscience techniques and experimentation, as well as in data science tools to analyze the resulting data, have dramatically furthered our understanding of fundamental brain functions. Historically, it has been common for analytical approaches to have a considerable lag in development following the availability of new neuroscience techniques. However, this relationship has not simply been unidirectional, as there have been examples in which analytical developments have directly led to new scientific questions and experiments. Here we review how this interplay between neuroscience and data science advances has unfolded in the past and into the present, with a focus on electrophysiology and calcium imaging. Applying lessons learned from the past and present, we then discuss expected developments, challenges, and opportunities in the future. We end by providing recommendations on how to foster the necessary team science approach to continue the advancement of research at the intersection of neuroscience and data science, which we call neurodatascience, toward a sustainable future.

We study the problem of aggregating a set of polygons by covering them with disjoint representative regions, thereby inducing a clustering of the polygons. Equivalently, this can be seen as a fence enclosure problem, where the goal is to surround the polygons with a set of closed curves. Our objective is to minimize a weighted sum of the total area and the total perimeter of the regions, which naturally extends other fencing problems and has applications in geographical information systems. Previously, this objective was only studied in a restricted variant, in which the boundary curves of the regions must be selected from a fixed subdivision of the plane. It is natural to ask whether the problem is still tractable if this restriction is removed, allowing output regions to be bounded by arbitrary curves. We provide a positive answer in the form of an algorithm with runtime Õ(n⁴), where n is the number of input vertices. To achieve this, we fully characterize the optimal solutions by showing that their boundaries are composed of input edges and circular arcs of constant radius. Additionally, we consider the parametric problem, where for every weighting factor we seek a solution that is optimal for that trade-off of area and perimeter. We show that O(n²) combinatorial solutions suffice to describe all optimal solutions across all weighting factors, and provide both an exact algorithm and an approximation scheme. To make the algorithms scalable in practice, we develop engineering techniques that exploit structural properties of the solutions. Our experimental evaluation on real-world data shows linear runtime in practice, even for the parametric variant.

DUNE (Deep Underground Neutrino Experiment) is a long-baseline neutrino oscillation experiment currently under construction, whose far detectors will be the largest liquid argon time projection chambers ever built. This detector design calls for custom-built cryogenic front-end electronics to meet its performance requirements. This paper describes the charge readout electronics that will be used in the DUNE horizontal drift (HD) far detector and presents performance results using data from the ProtoDUNE-HD detector, a 770 ton liquid argon time projection chamber operated at the CERN Neutrino Platform in 2024 that served as the final prototype of the DUNE HD design.

Alterations of gut microbiota are common in chronic kidney disease (CKD) and contribute to increased uremic toxins including indoxyl sulfate (IS), p-cresyl sulfate (pCS) and trimethylamine N-oxide (TMAO), which are linked to cerebrovascular disease risk. This study examined the kidney-gut-brain axis in CKD mice and in dialysis patients. Male and female mice with adenine-induced CKD were fed a high-amino-acid (HAA) diet to increase precursors of gut-derived uremic toxins. A subgroup of mice received antibiotics in drinking water to suppress gut microbiota and evaluate its role in toxin generation. Behavior tests, gut microbiome composition and brain histology for cerebral microhemorrhages were analyzed. CKD mice had higher serum levels of creatinine, cystatin C and gut-derived toxins, a 2.5-fold increase in brain microhemorrhages, and decreased locomotor activity. The HAA diet significantly increased serum TMAO but not IS and pCS, and all three toxins were reduced by antibiotic therapy. Sex differences were observed; in male animals, higher TMAO was associated with increased brain microhemorrhages, whereas in female mice, pCS was associated with brain microhemorrhage burden. The suppression of toxins with antibiotics improved working memory in male animals. Gut microbiota analysis revealed the expansion of Lactobacillus and Ileibacterium in CKD mice. The HAA diet and antibiotics altered gut microbiota composition without changing alpha diversity. The human study utilized biobanked serum samples and a retrospective review of brain imaging scans in a hemodialysis patient cohort; TMAO levels were associated with increased lobar microbleeds. Our study supports a role for bacterial-derived uremic toxins in the kidney-gut-brain axis and cerebral microhemorrhage formation in CKD.

BACKGROUND/AIMS: To apply retinal nerve fibre layer (RNFL) optical texture analysis (ROTA) to investigate (1) the patterns of RNFL bundle defects, and (2) the frequency of papillomacular and papillofoveal bundle involvement across early, moderate and advanced glaucoma. METHODS: All eyes underwent 24-2 visual field (VF) testing and optical coherence tomography (OCT) for ROTA. The borders of RNFL defects were delineated from ROTA, and the involvement of the arcuate, papillomacular and papillofoveal bundles was determined for each eye. 24-2 VF stimulus projections were mapped onto the corresponding topographic areas of ROTA images. Multilevel logistic regression analysis was applied to evaluate the structure-function association. RESULTS: Papillomacular bundle defects were highly prevalent in glaucoma, increasing from 87.7% in early to 95.35% in moderate and 100% in advanced glaucoma. Papillofoveal bundle defects were also common, increasing from 29.7% in early to 36.05% in moderate and 60.98% in advanced glaucoma. Central four 24-2 test locations that projected onto the trajectories of papillomacular or papillofoveal RNFL bundle defects demonstrated significantly increased likelihood of VF sensitivity abnormality (ORs of 22.42 at PDP<5% and 20.26 at TDP<5%, respectively, p<0.001 for both). CONCLUSION: ROTA uncovers a wide spectrum of RNFL bundle defects spanning the entire glaucoma continuum. It also provides visualisation of the preserved RNFL bundles in advanced glaucoma. Papillomacular and papillofoveal RNFL bundle defects are present in a considerable proportion of eyes with early, moderate and advanced glaucoma, and, when detected, they significantly increase the likelihood of abnormality in the corresponding central 24-2 test locations.

Family health informatics tools can help support well-being with shared data tracking. Prior work typically focused on shared data review, but often in specific moments, like bedtime, or centered on caregiving of children or elderly members. To investigate how tracking can support mutual health collaboration between family members pervasively across daily contexts, we designed and deployed FamilyBloom, a glanceable smartwatch and home display system for mood and goal tracking. Twelve families with both neurotypical and ADHD members used FamilyBloom for three months on average. Our findings reveal how family-centered tracking created collaboration opportunities and tensions across multiple ecological systems: individual self-regulation, collaborations within family dynamics, involvement of care networks with varying trust levels, institutional school constraints and cultural stigma, and temporality of regular routines and crisis periods. We discuss an ecosystem-aware approach to family informatics, wherein design can attend to how families navigate multiple contexts while sustaining family-level collaboration.

BACKGROUND: Patient selection for neoadjuvant therapy in colon cancer (CC) needs to be improved as utilizing computed tomography (CT) tumor (T) staging alone is associated with overstaging and overtreatment. Therefore, we sought to identify specific nodal imaging features that can be combined with radiologic T staging to improve patient selection. METHODS: Pre-operative CTs of stage I-III CC patients treated with upfront resection (2018–2023) were assessed by an expert abdominal radiologist blinded to the histopathologic staging. Radiologic T and node (N) staging based on five imaging features (single lymph node (LN) > 1 cm, ≥ 3 prominent LNs, irregular borders, heterogenous enhancement, and extramural venous invasion) were compared to the pathologic staging, grouped by proficient or deficient mismatch repair status (pMMR vs. dMMR). RESULTS: Of the 177 patients, 147 and 30 had pMMR and dMMR CC, respectively. The overstaging rate for pathologic T staging was 6.1% pMMR vs. 6.7% dMMR CC (p = 0.91). The overstaging rate for pathologic N disease by any imaging feature was 17.0% pMMR vs. 33.3% dMMR (p = 0.04) but improved to 4.8% pMMR vs. 6.7% dMMR (p = 0.67) with irregular borders and/or heterogenous enhancement alone. Compared to radiologic T staging only, the addition of N staging (any feature) to T staging resulted in decreased overstaging rates of stage I/low-risk stage II to high-risk stage II/III CC from 25.2% to 15.6% for pMMR CC and 60.0% to 33.3% for dMMR CC. Overstaging rates further decreased to 4.8% for pMMR and 16.7% for dMMR CC with the combination of radiologic T and N staging by irregular borders and/or heterogenous enhancement specifically. CONCLUSION: Combining radiologic T and N stage based on any of the five imaging features resulted in lower overstaging rates than radiologic T staging alone, but overstaging rates continued to improve when only irregular borders and heterogenous enhancement were used to assess nodal disease. These results can be used as an initial framework for combining CT T and N staging for CC and improve patient selection for neoadjuvant treatment.

Abstract The Energy Exascale Earth System Model version 3 (E3SMv3) represents the latest advancement in Earth system modeling developed by the U.S. Department of Energy (DOE). Building upon previous versions, E3SMv3 introduces significant updates across its coupled components to enhance capability and improve fidelity. The atmosphere component incorporates advancements in chemistry, aerosol‐cloud interactions, convection, and microphysics. The ocean features a new time‐stepping scheme and a higher‐resolution unstructured mesh with sub‐ice‐shelf cavities, while the sea ice model integrates advanced snow and ice physics for more realistic cryospheric simulations. The land model introduces prognostic vegetation dynamics and a new sub‐grid topographic treatment of solar radiation. A new tri‐grid configuration harmonizes the horizontal grids of the land and river components for improved process coupling. It is enabled by a new non‐linear remapping between the atmosphere and land. E3SMv3 underwent extensive testing through a comprehensive simulation campaign, including pre‐industrial control, idealized experiments, and historical simulations spanning 1850–2024. The model demonstrates significant improvements in simulating the evolution of the historical surface temperature, particularly addressing the “pothole cooling” bias in earlier versions. Reduced aerosol‐related forcing contributes to more realistic radiative forcing and better alignment with the observational record. Ocean heat content (OHC) and sea ice trends are also improved as a result. Plain Language Summary The Energy Exascale Earth System Model version 3 (E3SMv3), developed by the U.S. Department of Energy, is a state‐of‐the‐science tool designed to advance our understanding of the Earth and energy systems. This model simulates interactions between the atmosphere, land, rivers, oceans, and sea ice to predict Earth system changes and their impacts. E3SMv3 includes major improvements, such as enhanced representations of atmospheric chemistry, clouds, aerosols, sea ice, and vegetation dynamics. It also introduces refined marine and land grids to improve the accuracy of how these components interact. E3SMv3 was evaluated through simulations of past and historical conditions, totaling over 5,000 years. The model successfully resolved a major issue leading to unrealistic cooling trends during the mid‐20th century. E3SMv3 now provides more accurate predictions of surface temperatures, ocean heat content, and sea ice changes, closely matching observed data. Key Points E3SMv3 introduces major updates to the atmosphere, sea ice, and land components, enhancing capability and improving fidelity E3SMv3 increases resolution in the ocean‐sea ice mesh and unifies land and river grids for tighter coupling within and between components Reduced aerosol forcing in E3SMv3 resolves mid‐20th century cooling biases, aligning historical temperature trends with observations

Acupuncture is an ancient practice that was developed within the framework of traditional Chinese medicine. While acupuncture has been recently proposed as a therapy for Alzheimer’s disease (AD), acupuncture effects are not well understood in terms of neural mechanisms. Here, we review and examine the studies that used AD mouse models and analyze the experiments where researchers administered electroacupuncture (EA) to AD mice to assess the potential therapeutic impact of acupuncture on disease pathology and cognitive function in controlled laboratory settings. We analyzed 29 relevant PubMed articles published between January 2014 and July 2025. Our results reveal that EA significantly reduces both amyloid-beta (Aβ) and phosphorylated tau (p-tau) levels and neuroinflammatory biomarkers, including molecular signatures for activated microglia and astrocytes in the brain. EA also enhances cognitive functions. While no study directly compared acupoint strategies, the indirect comparisons in our network analysis suggest that GV20 has potential as a therapeutic target for AD. Our present meta-analysis and review of literature add to the evidence of integrative health practices for acupuncture-based Alzheimer’s disease treatment.