The exact text we saved

This is the version exactly as it was saved, character for character. The SHA-256 hash below is a 64-character code worked out from that text. Run the same text through SHA-256 and you get the same code; change a single letter and the code comes out completely different. Each version also stores the previous version’s hash, which chains them together so no version can be swapped out quietly.

TRUVACE RECORD VERSION
record: TRV-2026-2607
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-11-12T08:02:36.178547Z
status: published
lens: p_space
sector: climate
headline: Evaluating and Mitigating Carbon Dioxide Equivalent Emissions in Stroke Management: A Modeling Study
dek: Background Artificial intelligence (AI) can improve stroke imaging workflows, but its computational carbon footprint remains poorly quantified. We estimated carbon dioxide equivalent (CO 2 eq) emissions from AI use in US stroke management and evaluated carbon‐aware mitigation. Methods We combined TriNetX neuroimaging utilization data with Global Burden of Disease stroke burden estimates for 2018 to 2019. Imaging utilization rates were aligned by state, year, stroke type, sex, and age group. We modeled 2 AI scena…
gain_title: (none)
problem_title: A modeling study of US stroke care estimated that a full AI imaging workflow would emit about 16,375 metric tons of CO2 equivalent a year, more than double the 7,693 tons of a minimal setup.
trace_subject: (none)
gain_reading: (none)
gain_evidence: (none)
problem_reading: A modeling study of US stroke care estimated that a full AI imaging workflow would emit about 16,375 metric tons of CO2 equivalent a year, more than double the 7,693 tons of a minimal setup.
problem_evidence: Ideal AI generated 16 375.09 metric tons CO 2 eq/year | compared with 7692.70 metric tons CO 2 eq/year
quick_read: A modeling study estimated the carbon cost of running AI on stroke imaging across the United States in 2018 to 2019. A full-scale ('Ideal') AI workflow, with five models covering triage, segmentation, treatment planning and rehabilitation, generated 16,375.09 metric tons of CO2 equivalent a year. A minimal setup with two models generated 7,692.70 tons.

Emissions depended heavily on where the computing happens. Per-patient emissions in the ideal scenario ranged from 0.18 kg for ischemic stroke in Vermont to 6.66 kg in Wyoming, reflecting differences in state power grids. In the model, moving AI processing to the cleanest quarter of states cut ideal-scenario emissions by 54.75%, to 7,408.68 tons.

The figures are model estimates built from 82,121 TriNetX stroke records and 12,249,750 Global Burden of Disease-estimated cases, not measured emissions from hospitals. They cover only computing for computer-vision models, not the footprint of the scans themselves, and the authors note that wider use of CT angiography could raise the footprint of ischemic stroke care.
limitation: The estimates rest on 2018 to 2019 data and modeling assumptions, such as uniform application of AI, and exclude large language models, other greenhouse gases, water use, and cloud capacity and latency constraints for relocating processing.
tag: Evidence-backed problem
key_points: The ideal AI scenario (two 2D classification models, two 3D segmentation models, one 3D classification model) produced 16,375.09 tons CO2 equivalent a year; the minimal scenario produced 7,692.70 tons. | Per patient, ideal-scenario emissions were 2.96 kg for ischemic stroke and 3.42 kg for hemorrhagic stroke, versus 1.39 kg and 1.61 kg in the minimal scenario. | CT angiography contributed the most (7,240.06 tons) in the ideal scenario, followed by brain MRI (5,768.53 tons), brain CT (2,261.60 tons) and MR angiography (1,104.91 tons). | Younger hemorrhagic stroke patients had the highest per-patient emissions at 3.95 kg, versus 2.89 kg for older ones, because they receive more imaging. | Wyoming and West Virginia had the highest per-patient emissions; Vermont, California and New York had among the lowest. | Relocating processing to the cleanest 25% of states cut ideal-scenario emissions from 16,375.09 to 7,408.68 tons, and minimal-scenario emissions from 7,692.70 to 3,480.46 tons. | The study was funded by the START Program of the Faculty of Medicine, RWTH Aachen University.
rundown: Researchers combined imaging use rates from the TriNetX network with Global Burden of Disease stroke estimates, matched by state, year, stroke type, sex and age group, to estimate annual procedure volumes. They then measured the energy use of median-energy models from the MONAI library with the experiment-impact-tracker tool, including data-center overhead, and converted it to emissions using each state's grid carbon intensity from Electricity Maps.

Ischemic stroke accounted for 91.9% of the TriNetX records and 80.6% of the estimated cases, so it drove most of the total: 12,919.84 tons versus 3,455.26 tons for hemorrhagic stroke in the ideal scenario. Differences between men and women were negligible, while age differences were statistically significant.

For the mitigation test, states in the lowest quarter for emissions per megawatt-hour, including California, Connecticut, Idaho, Maine, New York, Oregon, Vermont and Washington, were treated as hosts, and processing tasks were assigned among them at random. Per-patient emissions in the ideal scenario fell from 2.96 kg to 1.34 kg, and in the minimal scenario from 1.39 kg to 0.63 kg.

The authors say the estimate sits between the modest energy needs reported for segmentation workflows and the much larger footprints modeled for digital pathology. It is well below the energy used to train the largest AI models. They suggest smaller or multi-task models, federated learning, routing non-urgent processing to low-carbon regions and carbon-cost displays in electronic health records.
sources:
- peer_reviewed | Journal of the American Heart Association | https://doi.org/10.1161/jaha.125.047556 | 2026-07-03
prev: 0000000000000000000000000000000000000000000000000000000000000000

SHA-256 hash

91378da1aab8f73a82692b3295f002d45a510043b4fab45a89f00be75607e459

Previous version’s hash

0000000000000000000000000000000000000000000000000000000000000000