Reducing Patient Movement During Magnetic Resonance Imaging: A Case Study

Abstract

Magnetic resonance imaging (MRI) is a highly informative diagnostic method; however, its quality heavily depends on the patient’s immobility. Even minimal movements, such as breathing, can cause artifacts that complicate image interpretation, not to mention more significant movements, such as twitching or shivering. Given the high cost of the procedure, repeated scanning is undesirable. The aim of this study was to prepare patients for MRI procedures using specialized training software designed to minimize involuntary movements and improve diagnostic quality. The software tracked participants’ movements in an MRI simulator and reproduced characteristic scanning sounds. The Farnebäck optical flow algorithm detected even the slightest movements captured by the camera, allowing for the evaluation of movements during training sessions and improving patient readiness for actual scanning. A pilot study conducted on a group of 10 students aged 21–27 years demonstrated a significant reduction in the average number of movements during testing—from 27.7 in the first test to 8.3 in the second, corresponding to an average decrease of 19.4 movements. Additionally, two participants showed a noticeable reduction in anxiety levels after the first test, which likely contributed to the decrease in movements, emphasizing the importance of psychological preparation in enhancing training effectiveness. The study results suggest potential improvements in the quality of diagnostic images, which can increase their diagnostic value and enhance patient comfort during actual scanning, reducing the likelihood of repeated procedures.

1. Introduction

Magnetic resonance imaging (MRI) is one of the most effective and informative diagnostic methods in modern medicine [1]. By harnessing the power of a strong magnetic field and radio-frequency pulses, MRI provides detailed images of internal structures without using ionizing radiation. This makes the method safe for a wide range of patients, including children and those for whom X-ray exposure is contraindicated.

The primary advantages of MRI include its high accuracy in soft tissue visualization, which makes it indispensable for diagnosing diseases of the brain and spinal cord, heart, and other vital organs. In addition, unlike X-rays, which are effective for visualizing bones, MRI is particularly valuable for imaging soft tissue structures in joints, such as ligaments and fibrocartilage. Due to its high resolution and the ability to detect pathologies at early stages, MRI enables doctors to make more precise diagnoses and effectively plan treatments. Additionally, the technology allows for the creation of three-dimensional models of organs, which is especially useful for preparing for complex surgeries.

The quality of MRI images directly depends on the patient’s ability to remain still [2]. Even minor movements can blur the images, reducing diagnostic accuracy. Therefore, when patients are unable to remain still, repeat scanning or sedation may be required, which increases risks. Sedation during MRI can lead to complications such as respiratory depression and airway obstruction, necessitating careful monitoring of the patient’s condition. Minimizing sedation levels, especially for those at risk of respiratory issues, helps reduce the likelihood of these complications and allows for a safer procedure [3].

Because MRI equipment demands substantial financial and staffing resources, it is not available in many medical facilities, especially in remote areas. This scarcity makes it extremely important to stay completely still during the MRI scan to avoid the need for additional scanning, which would result in further delays due to long waiting times.

Presently, patient movement during MRI procedures remains a challenge. To address this gap, this study investigated preparing patients for a successful MRI procedure through the use of an early version of specialized training software. This software, in conjunction with an MRI simulator, not only gradually familiarizes patients with diagnostic conditions, but also provides specialists with objective insights into the patient’s readiness for the procedure.

2. Previous Work

Currently, the topic of preparing patients for magnetic resonance imaging (MRI) procedures is of particular interest in the scientific and medical communities. Existing methods aim to reduce patient anxiety, improve their perception of the procedure, and enhance image quality, which helps minimize the need for anesthesia and sedative drugs, especially in the case of children and younger patients.

Among non-pharmacological preparation methods, training patients with MRI simulator models stands out, as it creates more realistic expectations and facilitates adaptation to the actual procedure. Ashmore et al. [4] demonstrated that using a mock scanner effectively reduced patient anxiety and minimized the need for sedatives. Hallowell et al. [5] highlighted the importance of familiarizing patients with the sounds and process of the scanner, allowing them to mentally prepare for lying still in an enclosed space.

One of the popular approaches to preparation is the use of virtual reality (VR) technologies, which allow patients to familiarize themselves with the procedure in advance and reduce fear of the unknown. Studies show that VR helps patients better adapt to the MRI environment and reduces pre-scan anxiety [6]. These findings are supported by other research where VR is considered an effective tool for lowering stress levels in patients before scanning [7].

Another innovation has been the use of multimedia materials, such as animations and educational videos, to reduce patient anxiety. Animated videos and interactive stories help patients better understand the procedure, alleviating their anxiety and fear [8]. Recent data indicate that these methods significantly help patients perceive the procedure as more familiar and predictable [9].

Traditional methods, such as explaining the procedure to parents or caregivers and providing pamphlets, are still commonly used. However, comparative studies show that more innovative approaches, such as VR and multimedia methods, prove to be significantly more effective in preparing patients for MRI [10]. At the same time, involving parents or close family members in the preparation process positively influences the patient, as emotional support helps reduce anxiety, as confirmed by studies examining the role of support systems in psychological preparation for MRI [11].

To summarize, studies demonstrate that using VR, multimedia technologies, and MRI simulators significantly reduces anxiety and improves the patient experience during MRI procedures. This underscores the relevance of this topic in medical practice, where advancing methods of psychological preparation remain a crucial area. Our work aimed to contribute to this field by proposing new approaches for further reducing patient stress before MRI.

3. Materials and Methods

For the purpose of the project, we developed a specialized application to monitor patient movements and simulate sounds characteristic of the MRI procedure.

3.1. Functionalities and Interface Design of the System

The application is designed with an intuitive user interface optimized to enhance usability and operational efficiency. Figure 1 presents the main screen of the application, showcasing its key control elements. The interface structure provides streamlined access to core functional modules, including features for searching and adding new patients, as well as analyzing statistical data from previously conducted scans.

The application is equipped with camera integration, enabling real-time display of the patient’s image on the MRI simulator screen (see Figure 2). The interface provides the operator with access to key parameters, including volume adjustment and motion detector data. This window serves as the operator’s primary tool during patient testing in the MRI simulator, ensuring continuous monitoring of the patient’s condition and testing parameters to enhance the accuracy of preparation for the actual procedure.

Figure 3 illustrates an example of the statistical data displayed by the application. In the figure, we can see a summary of the participants who took part in the simulation, including the total number of participants, gender distribution, average age, and time metrics (maximum, minimum, and average time). The graph illustrates the gender distribution of participants, facilitating demographic analysis. It is important to note that the statistics do not contain any personal data, ensuring confidentiality. The data can be downloaded in two formats (i.e., Excel and PDF) for further analysis or use.

The developed application is a comprehensive multifunctional system designed to enhance the efficiency and reliability of patient preparation processes. Its integrated functionality includes data management, training test administration, result analysis, and automated report generation, which contribute to optimizing the administrator’s workflow.

Table 1. Functionalities of the developed application.

Functionality	Description	Objective	Technical Details
Authorization	The operator accesses the system using a unique login and password.	To ensure data security and prevent unauthorized access.	Verifies credentials with encryption and creates secure sessions.
Patient registration and data storage	Registers new patients and stores their data in encrypted format.	To organize secure storage and processing of patient data.	Uses the SHA-256 algorithm for encryption and assigns a unique identifier to each patient.
Patient search and profile viewing	Searches for a patient using a unique identifier and displays their profile with test history.	To provide quick access to patient data to track progress.	Retrieves data from the database by patient ID.
Test initiation and sound simulation	Initiates or stops the test and adjusts MRI noise simulations.	To help patients acclimate to MRI conditions.	Synchronizes sound simulation with tests, allowing control of audio intensity.
Movement detection and value generation	Tracks patient movements and generates values representing movement intensity.	To record movements and minimize MRI simulation artifacts.	Uses the optical flow algorithm to filter minor fluctuations and focus on significant movements.
Automated result generation	Generates a report of test results upon completion.	To automate documentation and eliminate manual review.	Records movements and generates reports stored in the database.
Test history and statistics viewing	Displays test history and analyzes statistical data.	To track patient progress and personalize preparation.	Presents historical data in graphs and tables for easy assessment.
Data history storage and security	Stores test history with encrypted identifiers for protection.	To ensure long-term secure storage of data.	Uses SHA-256 encryption to protect data from unauthorized access.

presents the key functional modules of the system, their objectives, descriptions, and technical implementation details.

3.2. Motion Analysis Using Optical Flow

In the “Mock MRI Software” project, Gunnar Farnebäck’s optical flow algorithm is used for motion analysis, playing a key role in accurately tracking patient movements and minimizing artifacts in MRI simulations. This method is based on analyzing pixel brightness changes between consecutive images, allowing precise determination of the direction and magnitude of displacements. Farnebäck’s algorithm utilizes brightness decomposition in the pixel neighborhood, accounting for both linear and nonlinear variations, which ensures high accuracy even for small displacements [12].

The use of an image pyramid structure enhances accuracy through multiscale processing, making the algorithm resistant to noise and slight lighting variations. This approach enables high precision in real-time applications [13]. Studies confirm the effectiveness of Farnebäck’s algorithm in medical computer vision, including MRI analysis tasks [14].

To further improve accuracy, the system establishes a threshold based on the average magnitude of optical flow vectors in the initial frames. The system calculates the mean value of the first 800 vectors and adds a small offset (0.01) to ensure the threshold slightly exceeds this value. Further information on Farnebäck’s optical flow algorithm can be found in [15].

3.3. Participants

As part of our study, ten healthy volunteers were selected: five women and five men aged between 21 and 27 years (mean 23.8, std 2.04). All participants were students in Shamoon College of Engineering (https://en.sce.ac.il/ accessed on 4 December 2024), who voluntarily agreed to take part in the research. The selection criteria excluded participants with pacemakers, metallic implants, or claustrophobia, as these factors are absolute contraindications and could affect the reliability of the study results. Additionally, psychological stability and willingness to undergo a procedure simulating MRI conditions were assessed.

Before inclusion in the study, each participant was thoroughly instructed about the objectives of the experiment, participation procedures, potential risks (e.g., discomfort or anxiety under simulation conditions), and benefits (familiarization with the MRI procedure). The experiment was approved by the college ethics committee, which ensured compliance with all ethical norms and standards. All participants provided their written consent, confirming their willingness to participate and understanding of the research procedures.

3.4. Data Analysis

Quantitative statistical methods were used to analyze the collected data. The data included unique participant identifiers, gender, anxiety levels measured using the standardized Generalized Anxiety Disorder 7 (GAD-7) scale [16], and the number of movements recorded during each simulation.

3.4.1. GAD-7 Scale

The GAD-7 scale is a standardized tool for assessing anxiety levels. It consists of seven questions, to which participants respond on a 4-point scale: 0—never; 1—several days; 2—more than half the days; 3—nearly every day.

The total score can range from 0 to 21, with 0 to 4 points indicating minimal anxiety, 5 to 9 points indicating mild anxiety, 10 to 14 points indicating moderate anxiety, and 15 to 21 points indicating severe anxiety.

3.4.2. Normality Testing

Before conducting statistical analysis, the normality of data distribution was assessed using the Shapiro–Wilk test [17]. This test was chosen due to the small sample size ($n=10$) and its high sensitivity to deviations from normality.

3.4.3. Statistical Methods

To identify the relationship between anxiety levels and the number of movements, Pearson [18,19] correlation analysis was used, which assesses the linear relationship between variables. Before applying the test, we verified that the data met assumptions, including that the data was randomly collected and normally distributed. The significance level was set at $\alpha =0.05$. The magnitude of the correlation coefficient was interpreted according to the established classifications presented in

Table 2. Interpretation of correlation coefficient magnitude.

Observed Correlation Coefficient	Interpretation
0.00–0.10	Negligible correlation
0.10–0.39	Weak correlation
0.40–0.69	Moderate correlation
0.70–0.89	Strong correlation
0.90–1.00	Very strong correlation

.

3.4.4. Outlier Detection

Outliers in the data were identified using the interquartile range (IQR) method [20], which is calculated as the difference between the third ($Q3$) and first ($Q1$) quartiles. Observations falling outside the range were considered outliers and were either excluded from the analysis or processed depending on the context.

4. Experimental Setup and Test Layout

To test the motion recognition software, we created a mock-up of a magnetic resonance imaging scanner to simulate real scanning conditions, as illustrated in Figure 4. The mock-up was designed to provide the most realistic experience possible. It included a patient bed for participants to lie on, a built-in camera located on top of the mock-up to monitor movements and transmit real-time data to the developed software, and a sound simulation system that reproduced typical MRI noises, which further adapted the participants to real conditions.

The experiment consisted of three main phases, each aimed at preparing participants for magnetic resonance imaging to minimize anxiety and improve the quality of the acquired data. The pipeline of the experiment is presented in Figure 5. In the following, we describe each of the experimental phases.

• Phase 1: Participant Preparation.
• The participants’ preparation consisted of two steps:

1.1 

Guidance and explanation for participants: In this step, the participants were informed about the study, including its objectives, structure, and all phases. Special attention was given to explaining the MRI simulation stages. Furthermore, the participants were provided an opportunity to ask any questions, which were answered clearly to address any concerns.

1.2 

Signing of consent: The participants signed a consent form confirming their understanding of and agreement to all study phases.

• Phase 2: Simulation Test Without Preparation.
• The simulation test included the following steps:

2.1 

Completing the GAD-7 Questionnaire to assess anxiety level: Assessing participants’ anxiety levels prior to the simulation represents a critical step aimed at identifying potential psychological barriers and subsequently adapting the preparatory process to enhance participant comfort and ensure the successful completion of the procedure. Participants completed the GAD-7 questionnaire, which consists of seven questions aiming to assess their level of anxiety before the simulation [16]. The anxiety level is evaluated on a scale from 0 to 21, where a higher score indicates a greater level of anxiety.

2.2 

Conducting the MRI simulation: Participants underwent an MRI simulation under conditions resembling real ones (including sound and spatial constraints). The simulation lasted approximately 15 min, during which participant movements were monitored to assess their impact on the quality of MRI diagnostics.

2.3 

Data analysis: Collected data were analyzed to evaluate the anxiety level and the number of registered movements that may affect the quality of the MRI scan.

• Phase 3: Mock Test After Training
• This phase included three activities: filling out the questionnaire, conducting the second MRI simulation, and data analysis.

3.1 

Completing the GAD-7 Questionnaire again to assess anxiety level: After the first simulation, participants completed the GAD-7 questionnaire once more to evaluate changes in their anxiety levels.

3.2 

Conducting the MRI simulation: Participants underwent a second MRI simulation under the same conditions as in Phase 2. The simulation lasted no less than 15 min, ensuring consistent conditions for comparison.

3.3 

Data analysis: Results of the repeated simulation were analyzed to assess changes in the anxiety level and the number of registered movements that may have affect the quality of MRI diagnostics.

5. Results

The data collected included unique participant identifiers, gender, anxiety levels, and the number of movements recorded during each simulation. This approach allowed for a comprehensive evaluation of motor activity under simulation conditions and facilitated the identification of potential correlations between anxiety levels and simulation outcomes.

The analysis included only those movements that exceeded a predefined threshold value. This threshold was calculated as the average magnitude of the first 800 optical flow vectors, with a small corrective coefficient (0.01) added to exclude minor movements, such as breathing, from the final statistics.

Figure 6 presents a chart showing the number of movements of each participant during the two simulations. The analysis of participants’ motor activity within the MRI simulator during the simulations demonstrated a significant reduction in the number of movements, indicating the impact of training activities. The average number of movements during the first simulation was 27.7, with a standard deviation of 11.44; during the second simulation, it was 8.3, with a standard deviation of 4.32. The relative decrease in movements was 70.04%, highlighting the effectiveness of the simulation in reducing participants’ movements within the MRI simulator. These results emphasize the importance of training or adapting participants prior to conducting studies that require immobility.

The anxiety level of the participants was assessed before each simulation using the standardized GAD-7 scale. Analysis of the results showed that a decrease in anxiety before the second simulation compared to the first was observed in only 20% of the participants (2 out of 10) (see Figure 7). For the remaining 80% of participants, the anxiety level remained stable between the two stages of testing.

The normality test of the data on an interval scale using the Shapiro–Wilk test showed that the anxiety levels before the first ($W=0.957, p=0.748$) and second testing ($W=0.940, p=0.551$), as well as the number of movements in the first ($W=0.934, p=0.485$) and second tests ($W=0.940, p=0.555$), follow a normal distribution ($p>0.05$).

The correlation between anxiety level and the number of movements for the first and second tests was significantly moderate (correlation coefficient = 0.67, p < 0.01). However, when examining the data, we suspected that there might be outliers, i.e., participants №3 and №8. For the other participants, anxiety levels remained similar, but the number of movements during the MRI simulation still decreased. This may be attributed to the acquisition of experience and the adaptation to the experimental conditions. We used the interquartile range (IQR) [20] method to check outliers. Participant №3 was found as an outlier and eliminated from the dataset, and the correlation decreased and was found still be significant (correlation coefficient = 0.47, p < 0.05). These findings indicate that there was a positive correlation, but it is weaker.

6. Discussion and Conclusions

In this study, an assessment of participants’ motor activity in simulated MRI conditions was conducted, taking into account their anxiety levels before the procedure. Specialized software, described in previous sections, was used to detect and record movements during each simulation. The collected data included unique participant identifiers, gender, anxiety levels, and the number of recorded movements. This comprehensive approach allowed for a detailed analysis of motor activity in simulation conditions and the identification of possible correlations between anxiety levels and simulation outcomes.

The study results showed a significant 70% reduction in participants’ motor activity during the second MRI simulation. This highlights the importance of preliminary preparation, which helps increase patient awareness and understanding of the upcoming procedure. These findings are supported by studies such as Ashmore et al. [4], where the use of VR solutions to prepare children for MRI improved their perception of the procedure, reduced the need for anesthesia and sedation, and decreased the number of movements. Similarly, in the study by Szeszak et al. [10], it was noted that animated educational videos help reduce children’s anxiety and improve their ability to remain still during scanning, which, in turn, reduces the number of repeat scans and lowers the need for sedation.

Although the anxiety questionnaires completed before each simulation did not reveal significant changes in anxiety levels for most participants, the observed reduction in motor activity suggests that familiarity with the procedure is a key factor in reducing movements. These data are consistent with findings presented in the study of Hamd et al. [8], where preparation using video materials reduced anxiety by an average of 15–16%, which, in turn, contributed to a decrease in involuntary movements. The most pronounced reduction in anxiety was observed in patients undergoing the procedure for the first time, emphasizing the importance of preliminary preparation in improving patient behavior during scanning.

Nevertheless, the study has several limitations. The primary limitation is the small and relatively homogeneous sample size, which reduces reliability and limits the generalizability of the results to a broader population. The lack of demographic diversity among participants may introduce systematic biases and restrict the applicability of the conclusions. Additionally, the use of a simulator cannot fully replicate real MRI conditions, as differences in noise levels, lighting, and spatial configuration may affect participants’ responses and study outcomes.

To overcome these limitations, future plans include increasing the sample size and incorporating a more diverse group of participants, such as patients with various medical conditions and age groups, particularly children and elderly individuals. Further research will also focus on exploring the relationship between participants’ emotional states and their motor responses during the simulation, allowing for a deeper understanding of factors influencing patient behavior. Moreover, a more detailed analysis of anxiety levels and their correlation with movement data during MRI is required to provide a more comprehensive understanding of the influencing factors. To enhance the reliability of the study, improvements to the simulator are necessary, including better noise levels, lighting, and spatial parameters, which will allow for a more realistic MRI procedure model and improve the accuracy of the results.