Towards Sustainable Magnetic Resonance Imaging: Insights from long-term, high-resolution energy recordings across an entire scanner fleet
Florian Leonhard Raab, Fiona Mankertz, Nour Maalouf, Josephine Berger, Andreas Lingg, Reza Dehdab, Sebastian Werner, Judith Herrmann, et al.
physics.med-ph · cs.AI
Quick verdict
7-scanner, 1.5-yr MRI energy baseline: idle+eco exceed active scanning; duration associates with energy more than power. Rich data, inconsistent reporting.
Summary
This retrospective observational study characterizes MRI energy consumption across a fleet of seven Siemens scanners (four 1.5T, three 3T) at University Hospital Tübingen, merging 1 Hz power-meter recordings with system logs over roughly 23 months (January 2024–November 2025) and some 600,000 individual measurements. On average, a single measurement consumed 0.43 kWh (median 0.04 kWh) and a complete examination 13.50 kWh (median 11.96 kWh); mean scanner occupancy was 19.38%. Inter-scanner variation was substantial (9.70–17.51 kWh per examination), and idle plus eco-power modes cumulatively consumed more than active scanning (fleet daily: 1181.60 kWh idle + 132.00 kWh eco vs 1041.27 kWh scan, of 2354.89 kWh total). Univariable regressions on sequence-level averages found energy more strongly associated with scan duration (R²=0.769) than average power (R²=0.426). An interactive dashboard with live hospital feeds supports exploration and monitoring. Key limitations: a single vendor and institution, undisclosed idle/eco classification thresholds, unquantified data outages, unreconciled fleet figures (13.50 vs 13.02 kWh; 343.42×7 ≠ 2354.89 kWh/day), no data or code availability statement, and regressions that are partly definitional given E=P·t.
Review complete
Assessment at a glance
5
risk flags identified
● high 0
● medium 3
● low 2
5 strengths · 5 weaknesses · 5 author questions
AI text signal = Binoculars zero-shot detector score over 20 sampled passages (Fast-DetectGPT co-reported) — absence does not prove human authorship, and a flag does not imply misconduct.
Levels reflect verification complexity — not a quality score. Evolve never rates your paper.
Key claims under review
- Across a seven-scanner fleet over ~23 months, a single MRI measurement averaged 0.43 kWh (median 0.04) and a complete examination 13.50 kWh (median 11.96).
- Idle and eco-power modes cumulatively consumed more energy than active scanning (fleet: 1181.60 + 132.00 vs 1041.27 kWh/day of 2354.89 total), making non-scanning periods the prime optimization target.
- Sequence-level energy consumption is more strongly associated with scan duration (R²=0.769) than with average power (R²=0.426).
- Energy shares across operating modes must be interpreted relative to scanner occupancy, since utilization patterns strongly shape the mode distribution.
Strengths
Unmatched dataset scale and duration
Seven scanners, ~23 months (2024-01 to 2025-11), 1 Hz PAC3200T metering, ~600,000 measurements from 26,471 files (68 GB) — well beyond prior single-scanner or short-window studies such as Heye 2020 (19.9 kWh/exam).
Multi-level data integration down to the sequence
§2.1 merges energy recordings with vendor system logs per examination and per measurement, enabling the sequence-level regressions of Fig 7 and body-region breakdowns (Fig 6) that most prior MRI energy studies cannot attempt.
Transparent aggregation self-awareness
Operationally actionable headline finding
Live monitoring deliverable
Weaknesses
Internal numerical inconsistencies
Univariable regression despite a definitional identity
Mode classification opacity
Data quality unquantified
No availability statement; narrow scope
Full Review Report
$1.99 · one-timeUnlock the complete preliminary analysis Evolve produced for this paper.
Report preview
Scope & Claims
The paper presents a retrospective observational energy audit of seven Siemens MRI scanners (1.5T: 2× Aera, AvantoFit, Sola; 3T: PrismaFit, Vida, VidaFit) at two sites of University Hospital Tübingen, covering January 2024 to November 2025 (~23 months) with 1 Hz PAC3200T power-meter recordings merged with system logs (26,471 files, 68 GB, ~600,000 measurements). Central claims: (1) per-measurement mean 0.43 kWh (median 0.04) and per-examination mean 13.50 kWh (median 11.96); (2) idle plus eco-power modes cumulatively exceed active scanning (1181.60 + 132.00 vs 1041.27 kWh/day fleet-wide); (3) sequence energy correlates more strongly with duration (R²=0.769) than power (R²=0.426); (4) an interactive dashboard with live monitoring operationalizes the dataset. Claims are appropriately hedged ('associated with', 'may be an important factor') and mostly supported descriptively.
The remaining sections cover Methodology Assessment, Statistical Evidence, Results & Reproducibility, Limitations & Risks, and Verdict & Recommendations.
- All 5 strengths & 5 weaknesses, analyzed in detail
- 5 risk flags with severity, section & reviewer questions
- Methods & statistical evidence assessment
- 5 questions a peer reviewer would ask
- Full six-dimension evaluation report & recommendations
One-time purchase · this paper only
One-time purchase, this paper only. AI-generated preliminary analysis — not peer review.
Frequently asked questions
How much energy does one MRI scan use?
In this 23-month study of seven Siemens scanners at a German university hospital, a single measurement (sequence) averaged 0.43 kWh (median 0.04 kWh) and a complete examination averaged 13.50 kWh (median 11.96 kWh), with scanner means ranging from about 9.7 to 17.5 kWh per examination.
Do MRI scanners use more energy when idle than when scanning?
Cumulatively, yes. Fleet-wide daily consumption was 2354.89 kWh, of which idle mode used 1181.60 kWh and eco-power 132.00 kWh — together more than the 1041.27 kWh consumed during active scanning — although the split depends heavily on how occupied each scanner is.
What drives MRI energy consumption more: scan time or scanner power?
Linear regression on sequence-level averages found energy tracks scan duration more closely (R² = 0.769) than average power (R² = 0.426), suggesting that shortening acquisition time is a key efficiency lever.
Is the MRI energy dataset or dashboard code publicly available?
The paper describes an interactive dashboard fed by live hospital data but contains no data or code availability statement; the underlying records come from one hospital and Siemens equipment, so independent reuse is not currently possible from the publication alone.
About this review
This is an AI-generated preliminary review by Evolve.Pub, produced from the openly available text of the paper. It is not peer review and does not represent the authors' views. The canonical record of the paper is at the publisher link above.
- Figure images are not analyzable from the extracted text; judgments rest on captions and in-text numbers only.
- The apparent inconsistencies (13.50 vs 13.02 kWh; 0.43 vs 0.50 kWh; 343.42×7 vs 2354.89 kWh/day) are quoted verbatim from text and captions; a benign aggregation-method explanation may exist but is not stated in the paper.
- Any supplementary material was not included in the extraction.
Go deeper on this paper
Wrote a paper? Stress-test it before reviewers do.
Upload your manuscript for the same structured scrutiny — reasoning, statistics, evidence, figures, and reproducibility.