Energy Savings Potential for MRI Scanners in Routine Clinical Practice

Abstract

Purpose

We investigated the energy savings in our radiology department by changing the manner of operation of MRI scanners.

Materials and Methods

Since October 2022, two of our MRIs were consistently shut down overnight and on weekends instead of being left in prepared-to-scan mode. Also, an energy-saving mode was activated for one of the scanners. Previously, the scanners were only shut down on some days, and no energy-saving mode was active. We determined the energy savings by measuring the power consumption in the section of the building where the two MRI scanners are housed and comparing it with previous values.

Results

By shutting down both MRIs at night, the building section’s power consumption could be reduced by 7.04 kW, and by activating the energy-saving mode by an additional 2.15 kW. Through these measures, annual energy savings of up to 25000 kWh were achieved. This corresponds to a cost reduction of approx. EUR 4200, as well as a reduction in CO₂ emissions of about 10t. According to our measurements, a hospital that has previously left its MRIs ready for scanning at all times would save up to 20000 kWh per year per scanner, which corresponds to approx. EUR 3300 in cost savings and a reduction in CO₂ emissions of approx. 8t. In addition, there was no noticeable impact on the quality of patient care.

Conclusion

Energy-saving measures in radiology departments can be implemented effectively and with little effort by changing the manner of operation of MRI scanners.

Key Points

• Shutting down MRIs outside of routine operating hours reduces power consumption

• Activating an energy-saving mode further reduces consumption

• Implementing these measures is simple and has no identifiable disadvantages

Citation Format

• Thurner J, Fellner C, Stroszczynski C et al. Energy Savings Potential for MRI Scanners in Routine Clinical Practice. Rofo 2026; 198: 77–83

Keywords

MR-imaging - cost-effectiveness - carbon dioxide

Introduction

The topic of sustainability has become more and more important in recent years, and the challenge of using available resources responsibly has now also come to the field of medicine, where it is being discussed widely. Radiology is one of the biggest consumers of energy in healthcare, and this is due primarily to the energy-intensive imaging scans performed on a daily basis in radiology departments. Heye et al. determined that 4% of total energy consumption in their hospital resulted from MRIs and CTs [1]; the exact amount will vary from hospital to hospital, depending on the size of the radiology department. In any case, measures to save energy are particularly compelling in radiology, because the high power consumption of the scanners used means that changes in energy consumption can have a greater impact compared to other disciplines.

Our university hospital (839 beds as of 2022) has an annual energy consumption of approx. 30 million kWh [2]. In our radiology department, the main consumers of energy include: three MRIs (including one 3 T MRI and two 1.5 T MRIs), three CTs, three angiography systems, one mammography system, several X-ray systems (DR systems and digital mobile systems), three ultrasound scanners, several interventional medical devices (microwave ablation, thrombectomy, irreversible electroporation, etc.); in addition, there is equipment for room cooling, diagnostic workstations (PCs + monitors), video projectors in demo rooms, room heating, lighting, and other smaller electronic devices.

Like in most radiology departments, MRI and CT scanners are responsible for the highest amount of energy consumed. MRIs, in particular, are energy-intensive, more so than all other established imaging procedures in radiology [3]. However, these imaging devices are not regularly used at night or on weekends. Esmaeili et al. investigated the CO₂ balance of MRIs and identified a total consumption of 22.4 kg per patient [4]. Woolen et al. measured the energy consumption of MRIs for different modes of operation, and calculated that shutting down an MRI that was previously kept prepared-to-scan reduced power consumption by 25% to 33%; additionally, activating an energy-saving mode, if available, reduced power consumption by a further 22% to 28% [5].

In winter 2022/23 in light of the energy crisis at the time, we changed the way we operated our MRIs and also activated an energy-saving mode on one scanner. We measured the energy savings that resulted from this change. For the first time, we did not carry out individual measurements on the MRIs for the various operating modes. Rather we changed the set schedules for shutting down the scanners, and measured the energy consumption for the part of the building that houses the two 1.5 T MRI scanners. This allowed us to capture the real impact of these changes, including indirect variations in total energy consumption that resulted from changing the way we operated our MRIs.

Materials and Methods

Since no patient data were used, there was no need to submit the study to the ethics committee for review.

Until October 11, our MRIs were operated on a fixed schedule (except for emergencies at night and on weekends): They were shut down on Tuesday night and over the weekend. On Monday, Wednesday, and Thursday nights they remained prepared-to-scan. This schedule is referred to as “Scenario 1” below. The term “weekend” describes the period of “Friday night + Saturday + Sunday”.

After the changeover on October 12, 2022, we began to follow a new schedule: The MRIs are shut down every night and on the weekend, and energy-saving mode is activated on those scanners for which it is available. This schedule is referred to as “Scenario 2” below. For a visual representation of the two scenarios, see [Fig. 1].

Fig. 1 Scenario 1 (= MRIs shut down on Tuesday night and on the weekend; in prepared-to-scan mode on Monday, Wednesday, and Thursday nights; energy-saving mode of MRI 1 deactivated) and scenario 2 (= MRIs shut down every night and on the weekend; energy-saving mode of MRI 1 activated) each shown schematically over a period of one week. The term “weekend” refers to the period of “Friday night + Saturday + Sunday”. Purple = clinical operation, blue = prepared-to-scan mode, yellow = shut down (energy-saving mode disabled), green = shut down (energy-saving mode enabled). Drawing not to scale.

The following three MRI scanners are running in our department: A 1.5T Siemens Magnetom Sola (hereafter “MRI 1”), a 1.5T Siemens Magnetom Avanto fit (“MRI 2”), and a 3T Siemens Magnetom Skyra, which was not included in our readings because it is installed in a separate wing of the building and we were therefore unable to selectively measure its energy consumption. For manufacturer’s information on the scanners and their respective power consumption, see [Table 1].

Table 1 Average power consumption of the MRIs.

	System off	System prepared-to-scan	Typical measurement
Average power consumption of our department’s MRIs according to manufacturer specifications. Both scanners use a gradient system with a maximum amplitude of 45 mT/m and a maximum slew rate of 200 T/m/s. * Energy-saving mode enabled
MRI 1 (Siemens Magnetom Sola)	4.3 kW*	8.7 kW	22.7 kW
MRI 2 (Siemens Magnetom Avanto fit)	6.1 kW	9.3 kW	19.0 kW

The changeover results in two periods in which savings are generated (assuming the number of patients outside of routine clinical practice remains unchanged): starting with Scenario 2, MRI 1’s energy-saving mode is activated on Tuesday nights and weekends, further reducing energy consumption when shut down. The energy-saving mode achieves this by automatically switching off the compressor for helium cooling when not needed. Also starting with Scenario 2, both scanners are shut down on Monday, Wednesday, and Thursday nights, instead of remaining prepared-to-scan as in Scenario 1. During this period, additional savings are further generated by MRI 1’s energy-saving mode. No change is expected for energy consumption during routine operating hours (7 a.m. to 7 p.m.).

Energy consumption was measured over two periods for each of the scenarios and then compared. Energy consumption was not determined for the scanners themselves, but for a separate part of the building in which MRI 1 and MRI 2 are located. This enabled us to capture not only the direct effects of the change (i.e. reduced power consumption by the MRIs) but also the indirect effects (e.g. reduced cooling for scanner and room). The separate part of the building only houses MRI 1 and MRI 2 imaging systems, so it is purely MRI-focused. During the measurement periods, all other controllable variables (room cooling, diagnostic workstations, lights, etc.) were left essentially unchanged. It can therefore be assumed that differences in energy consumption between the two periods are mainly due to our deliberate change in the manner of operating the MRIs (i.e. shutting down scanners, activating an energy-saving mode).

For scenario 1, we chose the period of February 7–20, 2022, and for scenario 2, the period of February 6–19, 2023. The time periods were set in such a way that there would be as little fluctuation in energy consumption as possible due to different seasons. Energy consumption was measured by recording the meter reading in kWh every 15 minutes. For our analysis, the quarter-hourly readings were summed up to hourly ones, resulting in more clearly arranged graphs. We also decided to manually define the start and end points for the period of routine clinical operation in the analysis, because the daily shutdown times of the MRIs varied. This helped us to prevent supposed differences in night-time operation, because a scanner might have run longer or shorter than usual on one day. Consumption peaks due to unscheduled MRI operation (e.g. during night or weekend shifts), as well as periods when the shutdown requirements were not followed exactly, were excluded from calculations.

To evaluate the impact on routine clinical practice, we compiled and compared the entries in our digital appointment planner and the additional handwritten error log from two years prior to and two years after the changeover.

Results

The results of the readings are presented as a graph in [Fig. 2]. The savings achieved by MRI 1’s energy-saving mode result from the differences between Tuesday night and the weekend (blue areas). The readings showed an average difference of 2.15 kW. The savings from the combination of energy-saving mode on MRI 1 and the shutdown of both scanners (purple areas) totaled 9.19 kW, and consequently the pure effect of shutting down two MRI scanners results in savings of approx. 7.04 kW.

Fig. 2 Average hourly power consumption of the MRI building section during week 2. The grey graph represents the values from 2022 (= scenario 1), the blue graph those from 2023 (= scenario 2). The vertical blue lines indicate the manually defined start and end points of clinical and unplanned operation. Blue areas describe savings due to the activation of MRI 1’s energy-saving mode. Purple areas describe savings due to the combination of MRI 1’s energy-saving mode and shutting down both scanners. In both years, on Monday morning, only one of the two scanners was shut down, which is why the graphs are higher during this time. In 2022, only one of the two scanners was shut down on Friday night to Saturday morning, which is why the graph is also higher during this time. In 2023, an unscheduled examination was performed on Saturday afternoon.

We then compared our results to manufacturer’s specifications for power consumption of both MRI scanners, in order to obtain an estimate of what portion of the total savings is due to indirect effects. According to the manufacturer’s information ([Table 1]), a reduction in power consumption of 7.6 kW is expected (difference between “system prepared-to-scan” with 8.7 kW + 9.3 kW and “system off” with 4.3 kW + 6.1 kW: 18 kW − 10.4 kW = 7.6 kW). The practical savings in our readings was 9.19 kW, i.e. 21% more than expected based on the manufacturer’s specifications. This could be due mainly to indirect energy savings, such as reduced cooling requirements for scanners and rooms.

For an annual projection of the total savings, it was assumed that the operating hours of the MRIs are from 7 a.m. to 7 p.m. during the week and that operation at night and on weekends only takes place on an unscheduled basis. Extrapolated over the year, our department was able to achieve annual savings of up to 25000 kWh for the two MRIs measured by switching from Scenario 1 to Scenario 2 (not taking account of unscheduled operation). For more detailed information on the calculation, see [Table 2]. Depending on the price of electricity (which can vary greatly in the industry, the average in 2024 was 16.65 ct/kWh [6]), this currently corresponds to annual savings of approx. EUR 4200. The annual reduction in CO₂ emissions is approx. 10t, based on an emission factor of 380g CO₂/kWh (electricity mix in Germany) [7]. See also [Table 3] for a more detailed calculation.

Table 2 Measured power consumptions for both scenarios.

	Scenario 1	Scenario 2	Difference
The average total power consumption of the building section in which the two MRIs are accommodated during the different time periods for both scenarios as well as an extrapolation of the annual energy consumption during nights and the weekend. CI: Confidence interval “WKND”/“Weekend” refers to the period of “Friday night + Saturday + Sunday”. * Assuming 12h of routine clinical operations on weekdays. The period “Tuesday night + weekend” consequently corresponds to 72h/w, the period „Monday, Wednesday and Thursday night“ corresponds to 36h/w. Sample calculation: (72 h/w · 16.83 kW + 36 h/w · 23.96 kW) · 52 w/a = 107865 kWh/a.
Power consumption Tu night + WKND	16.83 kW (95%-CI: 16.71–16.95)	14.68 kW (95%-CI: 14.52–14.83)	2.15 kW
Power consumption Mo, We, Th night	23.96 kW (95%-CI: 23.87–24.05)	14.77 kW (95%-CI: 14.65–14.89)	9.19 kW
Energy consumption during nights and WKND per year extrapolated*	107865 kWh	82611 kWh	25254 kWh

Table 3 Energy savings, cost savings, and CO₂ reductions.

	Annual energy savings	Annual cost savings*	Annual CO2 reductions**
* at a price of 16.65 ct/kWh ** at an emission factor of 380 g CO2/kWh Energy, cost, and CO2 savings for the switch from scenario 1 to scenario 2 (measured values for two MRI scanners) as well as an extrapolation for the scenario in which an MRI was previously always left in prepared-to-scan mode and now is shut down consistently outside of routine clinical operations (= scenario 2). All energy prices including taxes. Average industrial price in Germany in 2024: 16.65 ct/kWh. Source: Statista [6] CO2 emission factor for German electricity mix in 2024: 380 g CO2/kWh. Source: Statista [7]
Measurement: Scenario 1 → Scenario 2 (for two MRI scanners)	25254 kWh	EUR 4205	9.60 t
Extrapolation: “Always prepared-to-scan” → Scenario 2 (for one MRI scanner)	19768 kWh	EUR 3291	7.51 t

Our analysis of the error logs revealed the following number of faulty scanner startups: 14 events in the two years before the change, when restarts were still occurring twice a week, and 16 events in the two years since then, when scanners were restarted five times a week.

Scenario 1 represents an already partially optimized strategy, because we were shutting down our MRIs on Tuesday nights and weekends before we changed their manner of operation. Our results can thus be extrapolated further to estimate the potential savings from consistently shutting down an MRI scanner that has been previously left prepared-to-scan every night and on weekends (not taking account of energy-saving mode). To calculate the reduction in total power consumption, we assumed half of the value saved by only shutting down both of our measured MRIs (without the changes from turning on energy-saving mode), i.e. 7.04 kW/2 = 3.52 kW in savings. The period when these savings are generated is Monday to Thursday nights and on weekends (= 108 h/w). This results in annual savings of up to 108 h/w · 52 w · 3.52 kW ≈ 20000 kWh per MRI scanner, which corresponds to financial savings of approx. EUR 3300 and CO₂ reduction of approx. 8 t ([Table 3]).

Discussion

Improving sustainability in radiology also supports the objective of providing a patient-centered practice, as described by Schreyer et al. [8], because 94% of all EU citizens state that environmental protection is important to them [9]. The topic of “sustainability in radiology” has also garnered attention in the literature; in recent years, several articles have been published that show the various ways in which radiology in general can be made more sustainable [10] [11] [12] [13]. According to Palm et al., sustainability in radiology (and beyond) stands on three pillars: ecology, economics, and social component [14]. The fact that the “ecology” pillar – and, as of late, the "economy" pillar as well, since the cost of electricity has increased significantly in recent years – is receiving so much attention likely stems from the many resources used by CTs and MRIs (e.g. power, cooling, helium).

As Woolen et al. have already shown, MRIs offer a high potential for savings, because energy consumption varies greatly for the different modes of operation (especially prepared-to-scan vs. shutdown) [5]. We measured the savings achieved through such a change in practice and were able to demonstrate for the first time how big such a change would be in reality, including all the indirect impacts.

Please note that the MRIs in our department were already shut down on Tuesday night and over the weekend before the changeover. The calculated annual saving of about 25000 kWh can therefore be seen as a further optimization of a strategy that was already partially optimized from the outset. In radiology departments that previously did not or only rarely shut down their MRIs, switching off scanners at night offers even greater potential for savings. Heye et al. estimate that about 50% of all imaging systems are not shut down overnight [15]. It is therefore particularly important that this issue receives even more attention in the future.

The significance of our data is limited by the fact that only a period of two weeks was considered. It is possible that the savings (especially for indirect impacts such as room cooling) may vary slightly at other times of the year, because a constant temperature must be maintained in the examination room. In our readings, the outdoor temperature during Scenario 1 averaged 4.6 °C [16] and during Scenario 2 it averaged 0.9 °C [17]. We assume that these fluctuations are small in reality (and level out over the full year) and that the extrapolation of the data to one year is therefore realistic. However, future studies on the topic could examine this effect in more detail.

Furthermore, when extrapolating the savings over a full year, it should be noted that unscheduled MRI examinations (e.g. emergencies at night or on weekends) cannot be taken into account, and we therefore based our calculation on “ideal” operation (i.e. without unscheduled examinations outside of clinical operations). In reality, the actual savings are likely to be somewhat lower. In our two study periods of two weeks each, four examinations took place outside of routine operating hours in 2022 and three in 2023, all of them on weekends.

Another limitation of our data is that no individual readings were performed on the MRIs. This means that small changes in baseline power consumption may have gone unnoticed. Future studies on the topic could examine the deviations from manufacturer’s specifications and the exact extent of indirect impacts, and should record variations in baseline consumption by measuring the electricity consumption of both the section of the building and the MRIs (individually).

The analysis of the error logs is subject to certain limitations, as the errors may be independent of the power-on/power-off process or may occasionally, and unintentionally, not be recorded. Nevertheless, our logs do not indicate that the change resulted in any restrictions on routine clinical practice. Although restarts have increased 2.5 times since then, the number of errors logged has remained almost the same.

Shutting down MRIs outside of routine operating hours is just one way to avoid unnecessary energy costs. For CTs, shutting them down outside of routine operating hours is not generally recommended, because they are often needed for emergency patient care. However, especially when several CT scanners are available in a department, it is possible to shut down at least some of them and thus prevent unnecessary energy consumption without impacting patient care. Brown et al. determined a savings potential of 14000 kWh per year per scanner [18]. The problem of emergency patient care does not usually affect MRIs, because the arrival of a patient registered as an emergency usually takes significantly longer than an MRI scanner needs for startup (in our case, depending on the model, approx. 6–7 minutes).

Another approach to saving energy is to switch off diagnostic workstations at night, as reported by Prasanna et al. and McCarthy et al. [19] [20]. However, Büttner et al. found that this had the effect of increasing personnel costs more than it reduced energy costs [21]. A fully automated shutdown based on a schedule, as described by Hainc et al., therefore seems to be a more suitable approach [22]. Heye et al. found that there is also potential for energy savings with less energy-intensive electronic devices (computers, smart monitors, printers) that are left on unnecessarily overnight [15]. Klein et al. [23] showed that improved sustainability can also be achieved when constructing a radiology practice.

In recent years, MRI manufacturers have also increasingly equipped their scanners with energy-saving modes. As a result, new MRIs provide lower power consumption straight from the factory. In our department, we were able to retrofit two scanners with an energy-saving mode simply by installing a software update, without having to make any changes to the hardware. If an MRI scanner does not have an energy-saving mode, it may be worthwhile to consult the manufacturer whether retrofitting is possible. In its “Sustainability Initiative”, the German Radiological Society also recommends checking for an energy-saving mode when purchasing new CT and MRI scanners [24]. We hope that our findings will encourage adoption of shutting down MRIs outside of routine operating hours as a general recommendation for clinical practice in the future. Recommendations by the German Society for Medical Physics already include not leaving MRIs unnecessarily in a prepared-to-scan state [25].

We expect the future to bring even more opportunities for optimization. For example, MRI scans supported by artificial intelligence (AI) could soon result in shorter scan times with comparable image quality [26] [27] [28], which would have the positive side effect of reducing the often long wait times for MRI appointments. Doo et al. note that AI can also have downsides in terms of sustainability due to the high energy consumption for development and, in particular, cloud storage. However, AI has many upsides, and the authors therefore propose ten best practices for leveraging its benefits in the most environmentally-friendly manner [29].

Conclusions

Introducing energy-saving measures in a radiology department can be worth it in many cases. Shutting down MRIs outside of normal operating hours and activating an energy-saving mode, if available, are easy and effective ways to make a radiology department more sustainable while also saving money. These steps require little effort to implement and have no noticeable negative impact on patient care.

Clinical relevance of study

• Sustainability is important to many people, and should therefore also be considered in the context of patient-centered radiology.

• In many hospitals, MRIs remain in a prepared-to-scan state even outside of routine operating hours, although there is no need to use the scanner at these times.

• We measured the energy savings that can be achieved in practice by changing the manner of operation for MRIs in routine clinical practice, and were able to show that this change is easy to implement and effectively reduces energy consumption.

Conflict of Interest

The authors declare that they have no conflict of interest.

Acknowledgement

Wir möchten uns herzlich bei Herrn Bernhard Horn (Technische Zentrale des Universitätsklinikums) bedanken, der uns die Auswertung der Stromzählerstände ermöglicht hat.

• References

• 1 Heye T, Knoerl R, Wehrle T. et al. The Energy Consumption of Radiology: Energy- and Cost-saving Opportunities for CT and MRI Operation. Radiology 2020; 295: 593-605
• 2 Management-Krankenhaus. Energiemanagement im Uniklinikum Regensburg | Management-Krankenhaus. Accessed February 11, 2024 at: https://www.management-krankenhaus.de/topstories/bauen-einrichten/energiemanagement-im-uniklinikum-regensburg
• 3 Goetzler W, Guernsey M, Foley K. et al. Energy Savings Potential and RD&D Opportunities for Commercial Building Appliances (2015 Update). Accessed December 18, 2023 at: https://www.energy.gov/sites/prod/files/2016/06/f32/DOE-BTO%20Comml%20Appl%20Report%20-%20Full%20Report_0.pdf
• 4 Esmaeili A, McGuire C, Overcash M. et al. Environmental impact reduction as a new dimension for quality measurement of healthcare services. International Journal of Health Care Quality Assurance 2018; 31: 910-922
• 5 Woolen SA, Becker AE, Martin AJ. et al. Ecodesign and Operational Strategies to Reduce the Carbon Footprint of MRI for Energy Cost Savings. Radiology 2023; 307: e230441
• 6 Statista. Industriestrompreise inkl. Stromsteuer in Deutschland bis 2024 | Statista. Accessed November 23, 2024 at: https://de.statista.com/statistik/daten/studie/252029/umfrage/industriestrompreise-inkl-stromsteuer-in-deutschland/
• 7 Statista. CO₂-Emissionsfaktor für den Strommix in Deutschland bis 2023 | Statista. Accessed November 23, 2024 at: https://de.statista.com/statistik/daten/studie/38897/umfrage/co2-emissionsfaktor-fuer-den-strommix-in-deutschland-seit-1990/
• 8 Schreyer AG, Schneider K, Dendl LM. et al. Patientenzentrierte Radiologie – Eine Hinführung durch ein narratives Review. RoFo 2022; 194: 873-881
• 9 European Commission. Eurobarometer: Protecting the environment and climate (03.03.2020). https://ec.europa.eu/commission/presscorner/detail/en/ip_20_331
• 10 Chawla A, Chinchure D, Marchinkow LO. et al. Greening the Radiology Department: Not a Big Mountain to Climb. Can Assoc Radiol J 2017; 68: 234-236
• 11 Sumner C, Ikuta I, Garg T. et al. Approaches to Greening Radiology. Acad Radiol 2023; 30: 528-535
• 12 Woolen SA, Kim CJ, Hernandez AM. et al. Radiology Environmental Impact: What Is Known and How Can We Improve?. Acad Radiol 2023; 30: 625-630
• 13 Chaban YV, Vosshenrich J, McKee H. et al. Environmental Sustainability and MRI: Challenges, Opportunities, and a Call for Action. J Magn Reson Imaging 2024; 59: 1149-1167
• 14 Palm V, Heye T, Molwitz I. et al. Nachhaltigkeit und Klimaschutz in der Radiologie – Ein Überblick. RoFo 2023; 195: 981-988
• 15 Heye T, Meyer MT, Merkle EM. et al. Turn It Off! A Simple Method to Save Energy and CO₂ Emissions in a Hospital Setting with Focus on Radiology by Monitoring Nonproductive Energy-consuming Devices. Radiology 2023; 307: e230162
• 16 Meteostat. Regensburg | Wetterrückblick & Klimadaten | Meteostat. Accessed November 23, 2024 at: https://meteostat.net/de/station/10776?t=2022–02–07/2022–02–20
• 17 Meteostat. Regensburg | Wetterrückblick & Klimadaten | Meteostat. Accessed November 23, 2024 at: https://meteostat.net/de/station/10776?t=2023–02–06/2023–02–19
• 18 Brown M, Snelling E, De Alba M. et al. Quantitative Assessment of Computed Tomography Energy Use and Cost Savings Through Overnight and Weekend Power Down in a Radiology Department. Can Assoc Radiol J 2023; 74: 298-304
• 19 Prasanna PM, Siegel E, Kunce A. Greening radiology. Journal of the American College of Radiology 2011; 8: 780-784
• 20 McCarthy CJ, Gerstenmaier JF, O’ Neill AC. et al. EcoRadiology--pulling the plug on wasted energy in the radiology department. Acad Radiol 2014; 21: 1563-1566
• 21 Büttner L, Posch H, Auer TA. et al. Switching off for future-Cost estimate and a simple approach to improving the ecological footprint of radiological departments. Eur J Radiol Open 2021; 8: 100320
• 22 Hainc N, Brantner P, Zaehringer C. et al. Green Fingerprint Project: Evaluation of the Power Consumption of Reporting Stations in a Radiology Department. Acad Radiol 2020; 27: 1594-1600
• 23 Klein HM. Ein neuer Ansatz zur Verbesserung der Energieeffizienz in radiologischen Versorgungseinheiten. RoFo 2023; 195: 416-425
• 24 DRG. Kleiner Aufwand, große Wirkung – sechs Energiesparstipps für die Radiologie. Accessed February 28, 2024 at: https://www.nachhaltigkeit.drg.de/de-DE/10078/energiesparstipps-fuer-die-radiologie/
• 25 DGMP. Handlungsempfehlungen zur Energieeinsparung in radiologischen, strahlentherapeutischen und nuklearmedizinischen Einrichtungen. Accessed February 28, 2024 at: https://www.dgmp.de/de-DE/1580/handlungsempfehlungen-zur-energieeinsparung-in-radiologischen-strahlentherapeutischen-und-nuklearmedizinischen-einrichtungen/
• 26 Johnson PM, Lin DJ, Zbontar J. et al. Deep Learning Reconstruction Enables Prospectively Accelerated Clinical Knee MRI. Radiology 2023; 307: e220425
• 27 Recht MP, Zbontar J, Sodickson DK. et al. Using Deep Learning to Accelerate Knee MRI at 3 T: Results of an Interchangeability Study. AJR Am J Roentgenol 2020; 215: 1421-1429
• 28 Edalati M, Zheng Y, Watkins MP. et al. Implementation and prospective clinical validation of AI-based planning and shimming techniques in cardiac MRI. Med Phys 2022; 49: 129-143
• 29 Doo FX, Vosshenrich J, Cook TS. et al. Environmental Sustainability and AI in Radiology: A Double-Edged Sword. Radiology 2024; 310: e232030

Publication History

Received: 02 September 2024

Accepted after revision: 05 February 2025

Article published online:
27 March 2025