Protocol for volumetric quantification of intrahepatic triglyceride content in adults with obesity Authors: Alba Camacho-Cardenosa, Antonio Clavero-Jimeno, Alejandro López-Vázquez, Mara Concepción, José L Martín-Rodríguez, Idoia Labayen, Jonatan R Ruiz, Rafael Cabeza Published: 2026 ABSTRACT SummaryThe study describes a protocol for the semi-automatic volumetric quantification of intrahepatic triglyceride (IHTG) levels in individuals with obesity based on the Dixon technique. Here, we describe a step-by-step procedure for image acquisition and exportation, followed by segmentation of liver regions and subsequent quantification of IHTG. This protocol is designed for application in studies with a specific focus on adults with obesity.For complete details on the use and execution of this protocol, please refer to Camacho-Cardenosa et al.1Subject areas: Health Sciences, Clinical Protocol, Metabolism, Systems biology Before you begin This protocol outlines the steps required to perform magnetic resonance imaging proton density fat fraction (MRI-PDFF) using a multi-echo Dixon technique for the volumetric quantification of IHTG content in adults with obesity. It may be more challenging in individuals with obesity compared to subjects with normal-weight, due to limitations in patient positioning and the need to carefully optimize image acquisition to avoid incomplete coverage or truncation of liver segments. Metabolic dysfunction-associated steatosis liver disease (MASLD), characterized by excessive accumulation of triglycerides in hepatocytes, is the most common chronic liver disease worldwide, affecting an estimated 30–40% of the population2 It is strongly associated with metabolic comorbidities, particularly type 2 diabetes and cardiovascular disease, placing a significant contributor to the global healthcare burden3 IHTG content is typically assessed using MRI, which is well-established, non-invasive gold standards for quantifying hepatic steatosis4 Among the various MRI techniques, image acquisition in this protocol is performed using the multi-echo Dixon sequence (6 echoes), which captures data at two specific time points: in-phase, when the signal of fat and water protons is aligned, and out-of-phase, when the signals are opposed. Each echo generates two datasets: one representing the water (W) signal and the other the fat (F) signal. These datasets are subsequently processed using a fitting algorithm that estimates and corrects signal effects, models the fat signal, and determines the proton density of both fat and water, allowing calculation of the fat fraction (FF), defined as F/(F + W), representing the percentage of fat within a specific tissue region5 IHTG content quantification is often performed using the mean FF from multiple regions of interest (ROIs) to simplify the process6,7 However, this approach may be limited by the heterogeneous distribution of lipids within the liver. We propose that a complete three-dimensional analysis helps avoid biases in the quantification of intrahepatic fat derived from the use of specific two-dimensional ROIs, or even from the use of a complete single slice. The inhomogeneity of ectopic fat distribution in the hepatic parenchyma is well-documented, especially in patients with metabolic diseases.8 Innovation This protocol lies in the integration of multi-echo Dixon MRI-PDFF with a standardized whole-liver volumetric analysis workflow. Unlike conventional approaches based on selected ROIs, this method enables comprehensive assessment of IHTG distribution, reducing sampling bias associated with hepatic fat heterogeneity. In addition, the protocol incorporates optimized image acquisition and segmentation procedures specifically adapted for individuals with obesity, addressing common challenges such as positioning constraints and incomplete liver coverage. By combining acquisition, segmentation, and volumetric quantification into a single, reproducible pipeline, this approach improves accuracy, consistency, and applicability across studies. This represents a significant advancement over existing ROI-based methods by providing a more robust and representative estimation of liver fat content. Institutional permissions This protocol is based on a previously conducted study. All participants provided written informed consent, and the study was approved by the Granada Provincial Research Ethics Committee (ref. CEI Granada—0365-N-23). The study was registered at ClinicalTrials.gov under the identifier NCT05897073. Researchers intending to apply this protocol in human studies must obtain ethical approval from their corresponding institutional review board. Subject screening Timing: 2 weeks Before enrollment, potential participants underwent clinical and physical examinations to verify their eligibility for inclusion.1 Participants completed the MRI assessments under fasting conditions (minimum 4–5 hours). 1.Pre-screening:a.Online survey.b.Telephone interview.i.Assessment of eligibility.ii.Evaluation of medical history. Pre-screening:a.Online survey.b.Telephone interview.i.Assessment of eligibility.ii.Evaluation of medical history. a.Online survey. Online survey. b.Telephone interview.i.Assessment of eligibility.ii.Evaluation of medical history. Telephone interview.i.Assessment of eligibility.ii.Evaluation of medical history. i.Assessment of eligibility. Assessment of eligibility. ii.Evaluation of medical history. Evaluation of medical history. 2.Screening: first in-person visit.a.Obtain oral and written informed consent.b.Final verification of inclusion and exclusion criteria.i.Record body weight and height measurements.ii.Assess aptitude for exercise: conduct incremental treadmill exercise test. Screening: first in-person visit.a.Obtain oral and written informed consent.b.Final verification of inclusion and exclusion criteria.i.Record body weight and height measurements.ii.Assess aptitude for exercise: conduct incremental treadmill exercise test. a.Obtain oral and written informed consent. Obtain oral and written informed consent. b.Final verification of inclusion and exclusion criteria.i.Record body weight and height measurements.ii.Assess aptitude for exercise: conduct incremental treadmill exercise test. Final verification of inclusion and exclusion criteria.i.Record body weight and height measurements.ii.Assess aptitude for exercise: conduct incremental treadmill exercise test. i.Record body weight and height measurements. Record body weight and height measurements. ii.Assess aptitude for exercise: conduct incremental treadmill exercise test. Assess aptitude for exercise: conduct incremental treadmill exercise test. Key resources table REAGENT or RESOURCESOURCEIDENTIFIERSoftware and algorithmsDixon acquisition sequenceSiemensMDixonLiverLab: (for PDFF analysis) (v. 1.0)SiemensN/A3DSlicer software (v. 5.8.1)3D slicer (Kitware)–R softwareThe R Projectwww.r-project.orgPyRadiomics library (v. 3.0.1)AIM-Harvardhttps://github.com/AIM-Harvard/pyradiomics/Elastix software (v. 5.2.0)https://elastix.dev/Other3 T MRI scannerSiemensMagnetom VidaPhased array flexible torso coilSiemensFlexible Body Coil (18-channel) Step-by-step method details Scanning procedure Timing: 30 min Here, we describe a standardized MRI-PDFF protocol for quantifying IHTG level in individuals with obesity. The procedure is reproducible and supports sequential application for longitudinal volumetric assessment of IHTG over time. 1.Pre-scanning:CRITICAL: This initial step must be performed by a trained technician to ensure participant safety and proper image acquisition.a.Enter the participant’s identification number, body weight, height, and date of birth into the system.b.Ensure compliance with standard MRI safety procedures.Note: Verify that the participant does not possess any ferromagnetic metal objects on or within their body. Pre-scanning:CRITICAL: This initial step must be performed by a trained technician to ensure participant safety and proper image acquisition.a.Enter the participant’s identification number, body weight, height, and date of birth into the system.b.Ensure compliance with standard MRI safety procedures.Note: Verify that the participant does not possess any ferromagnetic metal objects on or within their body. CRITICAL: This initial step must be performed by a trained technician to ensure participant safety and proper image acquisition. a.Enter the participant’s identification number, body weight, height, and date of birth into the system. Enter the participant’s identification number, body weight, height, and date of birth into the system. b.Ensure compliance with standard MRI safety procedures.Note: Verify that the participant does not possess any ferromagnetic metal objects on or within their body. Ensure compliance with standard MRI safety procedures.Note: Verify that the participant does not possess any ferromagnetic metal objects on or within their body. Note: Verify that the participant does not possess any ferromagnetic metal objects on or within their body. 2.Participant positioning:CRITICAL: This step is essential to ensure reproducibility of serial MRI examinations in longitudinal studies. Accordingly, each participant should be positioned in an identical manner for every imaging session.a.To minimize lower-body motion during acquisition, place cushions under the back and/or legs.b.Supply the participant with hearing protection and an emergency call button to ensure safety and allow communication in case of distress (troubleshooting 1).c.Position the torso coil used for liver imaging over the abdomen, centered above the liver.d.Move the scanner table forward until the participant is positioned at the isocenter of the scanner bore.e.Ask the participant to remain as motionless as possible during acquisition. Participant positioning:CRITICAL: This step is essential to ensure reproducibility of serial MRI examinations in longitudinal studies. Accordingly, each participant should be positioned in an identical manner for every imaging session.a.To minimize lower-body motion during acquisition, place cushions under the back and/or legs.b.Supply the participant with hearing protection and an emergency call button to ensure safety and allow communication in case of distress (troubleshooting 1).c.Position the torso coil used for liver imaging over the abdomen, centered above the liver.d.Move the scanner table forward until the participant is positioned at the isocenter of the scanner bore.e.Ask the participant to remain as motionless as possible during acquisition. CRITICAL: This step is essential to ensure reproducibility of serial MRI examinations in longitudinal studies. Accordingly, each participant should be positioned in an identical manner for every imaging session. a.To minimize lower-body motion during acquisition, place cushions under the back and/or legs. To minimize lower-body motion during acquisition, place cushions under the back and/or legs. b.Supply the participant with hearing protection and an emergency call button to ensure safety and allow communication in case of distress (troubleshooting 1). Supply the participant with hearing protection and an emergency call button to ensure safety and allow communication in case of distress (troubleshooting 1). c.Position the torso coil used for liver imaging over the abdomen, centered above the liver. Position the torso coil used for liver imaging over the abdomen, centered above the liver. d.Move the scanner table forward until the participant is positioned at the isocenter of the scanner bore. Move the scanner table forward until the participant is positioned at the isocenter of the scanner bore. e.Ask the participant to remain as motionless as possible during acquisition. Ask the participant to remain as motionless as possible during acquisition. 3.Data acquisition:a.Verify that the subject and the coil are correctly positioned.b.Acquire T1-weighted VIBE Dixon images for fat–water separation of the liver.Note: T1 are segmentation series aimed to anatomical delineation. While additional sequences are acquired, t1 out-of-phase (named as t1_.opp) will be the sequence used. VIBE are quantification series aimed at tissue quantitative analysis, being the sequences used Echo 2 (named as vibe_.ECO2) and FF (named as vibe_.FF). Use these parameters settings: echo times (TE1 to TE6) of 1.05, 2.46, 3.69, 4.92, 6.15 and 7.38 ms; a repetition time (TR) of 9.0 ms; a flip angle of 4°; and a slice thickness of 3.0 mm.c.After the acquisition and before exporting the images, the technician should verify that the liver has been fully acquired, as its position may vary due to breathe-holding.Note: Repeat in case of respiratory artifact (Troubleshooting 2). Data acquisition:a.Verify that the subject and the coil are correctly positioned.b.Acquire T1-weighted VIBE Dixon images for fat–water separation of the liver.Note: T1 are segmentation series aimed to anatomical delineation. While additional sequences are acquired, t1 out-of-phase (named as t1_.opp) will be the sequence used. VIBE are quantification series aimed at tissue quantitative analysis, being the sequences used Echo 2 (named as vibe_.ECO2) and FF (named as vibe_.FF). Use these parameters settings: echo times (TE1 to TE6) of 1.05, 2.46, 3.69, 4.92, 6.15 and 7.38 ms; a repetition time (TR) of 9.0 ms; a flip angle of 4°; and a slice thickness of 3.0 mm.c.After the acquisition and before exporting the images, the technician should verify that the liver has been fully acquired, as its position may vary due to breathe-holding.Note: Repeat in case of respiratory artifact (Troubleshooting 2). a.Verify that the subject and the coil are correctly positioned. Verify that the subject and the coil are correctly positioned. b.Acquire T1-weighted VIBE Dixon images for fat–water separation of the liver.Note: T1 are segmentation series aimed to anatomical delineation. While additional sequences are acquired, t1 out-of-phase (named as t1_.opp) will be the sequence used. VIBE are quantification series aimed at tissue quantitative analysis, being the sequences used Echo 2 (named as vibe_.ECO2) and FF (named as vibe_.FF). Use these parameters settings: echo times (TE1 to TE6) of 1.05, 2.46, 3.69, 4.92, 6.15 and 7.38 ms; a repetition time (TR) of 9.0 ms; a flip angle of 4°; and a slice thickness of 3.0 mm. Acquire T1-weighted VIBE Dixon images for fat–water separation of the liver.Note: T1 are segmentation series aimed to anatomical delineation. While additional sequences are acquired, t1 out-of-phase (named as t1_.opp) will be the sequence used. VIBE are quantification series aimed at tissue quantitative analysis, being the sequences used Echo 2 (named as vibe_.ECO2) and FF (named as vibe_.FF). Use these parameters settings: echo times (TE1 to TE6) of 1.05, 2.46, 3.69, 4.92, 6.15 and 7.38 ms; a repetition time (TR) of 9.0 ms; a flip angle of 4°; and a slice thickness of 3.0 mm. Note: T1 are segmentation series aimed to anatomical delineation. While additional sequences are acquired, t1 out-of-phase (named as t1_.opp) will be the sequence used. VIBE are quantification series aimed at tissue quantitative analysis, being the sequences used Echo 2 (named as vibe_.ECO2) and FF (named as vibe_.FF). Use these parameters settings: echo times (TE1 to TE6) of 1.05, 2.46, 3.69, 4.92, 6.15 and 7.38 ms; a repetition time (TR) of 9.0 ms; a flip angle of 4°; and a slice thickness of 3.0 mm. c.After the acquisition and before exporting the images, the technician should verify that the liver has been fully acquired, as its position may vary due to breathe-holding.Note: Repeat in case of respiratory artifact (Troubleshooting 2). After the acquisition and before exporting the images, the technician should verify that the liver has been fully acquired, as its position may vary due to breathe-holding.Note: Repeat in case of respiratory artifact (Troubleshooting 2). Note: Repeat in case of respiratory artifact (Troubleshooting 2). Data analysis Timing: 50 min This process fundamentally consists of two phases: liver segmentation and FF quantification. Note: This procedure will be performed using a freely available open-source software application for medical image computing known as 3D Slicer (v. 5.8.1), along with its available modules.9 4.Perform liver segmentation slice by slice using the Segmentation Editor in 3DSlicer:CRITICAL: Edition and corrections are applied across the axial image stack for each patient and adjustments are propagated to the coronal and sagittal planes to ensure consistency across all anatomical views (see Figure 1).a.Manually refine the initial semi-automated liver segmentation, which may miss parts of the liver or include nearby structures (e.g., the inferior vena cava) (see Methods Video S1).Note: Liver MRI images acquired with Siemens equipment include a preliminary automatic segmentation, shown as a white dotted contour in the “t1_.W_SEG” series.b.Perform image registration using the SlicerElastix module within 3DSlicer (see Methods Video S2).Note: This additional image registration step is performed to correct for potential patient motion between acquisitions, ensuring accurate alignment between the edited segmentation and the quantification data. This is an essential step in medical image processing, establishing precise point-to-point correspondence between a reference segmentation series (t1_.opp) and a target quantification series (vibe._FF) within a common coordinate system. Based on the fixed (or reference) serie (named as t1_._opp) and a moving serie (named as vibe_._Eco2), the registration yields a new volume (named as Eco2_registered) and a transformation file (named as Transform_registered).Methods Video S1. Edit liver segmentation with 3DSlicer, related to Data analysisMethods Video S2. Image registration process with 3DSlicer, related to Data analysisc.Apply the transformation file, that contains the transformation vectors, to the target serie using the Transforms module within 3DSlicer (see Methods Video S2).Note: The transformation file generated in the previous phase is applied to the target serie (named as vibe_._FF). The result is a newly registered version of the target image serie (named as FF_registered).Optional: A manual segmentation could be delineated directly on the quantification series (vibe_.FF), using Segmentation Editor in 3DSlicer, and avoiding the need for image registration. It is performed slice by slice across the axial image stack and the resulting contours are propagated to the coronal and sagittal planes to ensure consistency across anatomical views. Ideally, this process should be carried out independently by two observers, and inter-observer reliability should be assessed to ensure reproducibility of the results. Perform liver segmentation slice by slice using the Segmentation Editor in 3DSlicer:CRITICAL: Edition and corrections are applied across the axial image stack for each patient and adjustments are propagated to the coronal and sagittal planes to ensure consistency across all anatomical views (see Figure 1). CRITICAL: Edition and corrections are applied across the axial image stack for each patient and adjustments are propagated to the coronal and sagittal planes to ensure consistency across all anatomical views (see Figure 1). a.Manually refine the initial semi-automated liver segmentation, which may miss parts of the liver or include nearby structures (e.g., the inferior vena cava) (see Methods Video S1).Note: Liver MRI images acquired with Siemens equipment include a preliminary automatic segmentation, shown as a white dotted contour in the “t1_.W_SEG” series.b.Perform image registration using the SlicerElastix module within 3DSlicer (see Methods Video S2).Note: This additional image registration step is performed to correct for potential patient motion between acquisitions, ensuring accurate alignment between the edited segmentation and the quantification data. This is an essential step in medical image processing, establishing precise point-to-point correspondence between a reference segmentation series (t1_.opp) and a target quantification series (vibe._FF) within a common coordinate system. Based on the fixed (or reference) serie (named as t1_._opp) and a moving serie (named as vibe_._Eco2), the registration yields a new volume (named as Eco2_registered) and a transformation file (named as Transform_registered).Methods Video S1. Edit liver segmentation with 3DSlicer, related to Data analysisMethods Video S2. Image registration process with 3DSlicer, related to Data analysisc.Apply the transformation file, that contains the transformation vectors, to the target serie using the Transforms module within 3DSlicer (see Methods Video S2).Note: The transformation file generated in the previous phase is applied to the target serie (named as vibe_._FF). The result is a newly registered version of the target image serie (named as FF_registered).Optional: A manual segmentation could be delineated directly on the quantification series (vibe_.FF), using Segmentation Editor in 3DSlicer, and avoiding the need for image registration. It is performed slice by slice across the axial image stack and the resulting contours are propagated to the coronal and sagittal planes to ensure consistency across anatomical views. Ideally, this process should be carried out independently by two observers, and inter-observer reliability should be assessed to ensure reproducibility of the results. a.Manually refine the initial semi-automated liver segmentation, which may miss parts of the liver or include nearby structures (e.g., the inferior vena cava) (see Methods Video S1).Note: Liver MRI images acquired with Siemens equipment include a preliminary automatic segmentation, shown as a white dotted contour in the “t1_.W_SEG” series. Manually refine the initial semi-automated liver segmentation, which may miss parts of the liver or include nearby structures (e.g., the inferior vena cava) (see Methods Video S1).Note: Liver MRI images acquired with Siemens equipment include a preliminary automatic segmentation, shown as a white dotted contour in the “t1_.W_SEG” series. Note: Liver MRI images acquired with Siemens equipment include a preliminary automatic segmentation, shown as a white dotted contour in the “t1_.W_SEG” series. b.Perform image registration using the SlicerElastix module within 3DSlicer (see Methods Video S2).Note: This additional image registration step is performed to correct for potential patient motion between acquisitions, ensuring accurate alignment between the edited segmentation and the quantification data. This is an essential step in medical image processing, establishing precise point-to-point correspondence between a reference segmentation series (t1_.opp) and a target quantification series (vibe._FF) within a common coordinate system. Based on the fixed (or reference) serie (named as t1_._opp) and a moving serie (named as vibe_._Eco2), the registration yields a new volume (named as Eco2_registered) and a transformation file (named as Transform_registered).Methods Video S1. Edit liver segmentation with 3DSlicer, related to Data analysisMethods Video S2. Image registration process with 3DSlicer, related to Data analysis Perform image registration using the SlicerElastix module within 3DSlicer (see Methods Video S2). Note: This additional image registration step is performed to correct for potential patient motion between acquisitions, ensuring accurate alignment between the edited segmentation and the quantification data. This is an essential step in medical image processing, establishing precise point-to-point correspondence between a reference segmentation series (t1_.opp) and a target quantification series (vibe._FF) within a common coordinate system. Based on the fixed (or reference) serie (named as t1_._opp) and a moving serie (named as vibe_._Eco2), the registration yields a new volume (named as Eco2_registered) and a transformation file (named as Transform_registered). Note: This additional image registration step is performed to correct for potential patient motion between acquisitions, ensuring accurate alignment between the edited segmentation and the quantification data. This is an essential step in medical image processing, establishing precise point-to-point correspondence between a reference segmentation series (t1_.opp) and a target quantification series (vibe._FF) within a common coordinate system. Based on the fixed (or reference) serie (named as t1_._opp) and a moving serie (named as vibe_._Eco2), the registration yields a new volume (named as Eco2_registered) and a transformation file (named as Transform_registered). Methods Video S1. Edit liver segmentation with 3DSlicer, related to Data analysis Methods Video S1. Edit liver segmentation with 3DSlicer, related to Data analysis Methods Video S2. Image registration process with 3DSlicer, related to Data analysis Methods Video S2. Image registration process with 3DSlicer, related to Data analysis c.Apply the transformation file, that contains the transformation vectors, to the target serie using the Transforms module within 3DSlicer (see Methods Video S2).Note: The transformation file generated in the previous phase is applied to the target serie (named as vibe_._FF). The result is a newly registered version of the target image serie (named as FF_registered).Optional: A manual segmentation could be delineated directly on the quantification series (vibe_.FF), using Segmentation Editor in 3DSlicer, and avoiding the need for image registration. It is performed slice by slice across the axial image stack and the resulting contours are propagated to the coronal and sagittal planes to ensure consistency across anatomical views. Ideally, this process should be carried out independently by two observers, and inter-observer reliability should be assessed to ensure reproducibility of the results. Apply the transformation file, that contains the transformation vectors, to the target serie using the Transforms module within 3DSlicer (see Methods Video S2).Note: The transformation file generated in the previous phase is applied to the target serie (named as vibe_._FF). The result is a newly registered version of the target image serie (named as FF_registered).Optional: A manual segmentation could be delineated directly on the quantification series (vibe_.FF), using Segmentation Editor in 3DSlicer, and avoiding the need for image registration. It is performed slice by slice across the axial image stack and the resulting contours are propagated to the coronal and sagittal planes to ensure consistency across anatomical views. Ideally, this process should be carried out independently by two observers, and inter-observer reliability should be assessed to ensure reproducibility of the results. Note: The transformation file generated in the previous phase is applied to the target serie (named as vibe_._FF). The result is a newly registered version of the target image serie (named as FF_registered). Optional: A manual segmentation could be delineated directly on the quantification series (vibe_.FF), using Segmentation Editor in 3DSlicer, and avoiding the need for image registration. It is performed slice by slice across the axial image stack and the resulting contours are propagated to the coronal and sagittal planes to ensure consistency across anatomical views. Ideally, this process should be carried out independently by two observers, and inter-observer reliability should be assessed to ensure reproducibility of the results. 5.Perform volumetric quantification of FF with PyRadiomics module (see Methods Video S3). Perform volumetric quantification of FF with PyRadiomics module (see Methods Video S3). Hepatic fat fraction (FF: 20%) measurement using semiautomatic whole-liver segmentation in a 39-year-old man Note: Select as Input Image Volume vibe_.FF_registered generated in the phase “Image registration” or vibe_.FF if the segmentation was performed manually. By applying these parameters, a table containing the quantitative data of the liver can be obtained: mean FF (original, first-order, mean), median FF (original, first-order, median), and standard deviation of liver FF (original, first-order, variance). Notably, these values are typically calculated on a scale from 0 to 1000; therefore, they should be divided by 10 to express the FF as a percentage. Methods Video S3. Quantification of fat fraction with PyRadiomics, related to Data analysis To verify the quality of the segmentation and registration, seven images per case have been generated in PNG format at different heights of the liver morphology, overlaying the fat fraction image with the segmentation. This allows for a very quick verification of the segmentation quality along with the subsequent registration. Additionally, a subset of 10% of the total sample is randomly selected, in which quantification is performed manually and compared with the result of the automatic process. Optional: Quantification can alternatively be performed using a semi-automatic workflow in which the software-generated segmentation is employed, with user intervention restricted to editing and refining the contours when necessary. All subsequent steps of the analysis are carried out through a dedicated Python (version 3.11.4) script, which will be shared publicly on GitHub to facilitate accessibility (see Data and Code Availability section). 6.Review unclear or inconsistent cases in 3DSlicer to ensure data integrity and accuracy. Review unclear or inconsistent cases in 3DSlicer to ensure data integrity and accuracy. Expected outcomes This document provides researchers from any discipline with a comprehensive protocol for the acquisition and quantification of IHTG content in individuals with obesity (as patient positioning and proper field-of-view coverage may be challenging). One of the main strengths of this protocol is its implementation using free and open-source software tools developed for medical image informatics, image processing, and three-dimensional visualization.9 The protocol described herein, which relies exclusively on open-source tools, promotes methodological transparency, facilitates reproducibility, and enhances accessibility for the broader scientific community. Limitations The present protocol presents several challenges and limitations. It may not be suitable for all patients—for example, individuals who are claustrophobic, exceed the physical size limits of the MRI scanner, or have contraindications such as certain metallic implants. Consequently, it is important to inform patients that MRI-PDFF is specifically intended to quantify liver fat content (particularly intrahepatic triglycerides), and should not be considered a reliably tool for excluding unrelated hepatic abnormalities. Troubleshooting Problem 1 Participant reports feeling anxious prior to the scan. Potential solution Participants should be reassured that modern MRI scanners are designed to maximize comfort, featuring wider, ventilated, and well-lit bores. During the examination, continuous communication with the technician is maintained through the scanner’s intercom system, and participants are provided with a call button to report any discomfort. If concerns about tolerating the procedure persist, consultation with a physician may be recommended to consider mild oral medication to reduce anxiety. Problem 2 In some cases, participants may have difficulty holding their breath for the 20 seconds indicated by the system. It could lead to the PDFF measurements are obtained at low resolution, which may limit detailed evaluation of hepatic fat distribution. Potential solution Participants should be instructed and trained in breath-holding prior to the scan, and their breath-hold capacity may be assessed using a stopwatch or wall clock. If necessary, shorter sequences during inspiration can be performed. In the presence of respiratory artifacts, allow a brief rest period and repeat the acquisition. Resource availability Lead contact Requests for additional information, resources, or reagents should be directed to the lead contact, Alba Camacho-Cardenosa (acamachocardenos@ugr.es). Technical contact Technical questions on executing this protocol should be directed to and will be answered by the technical contact, Rafael Cabeza (rcabeza@unavarra.es). Materials availability This study generated new materials, including Methods Video S1 (https://figshare.com/s/c155aafd5d0fb16b34cc), Methods Video S2 (https://figshare.com/s/d052a0b1bffa3301c205), and Methods Video S3 (https://figshare.com/s/d1cef80a652d50e8f53d). Data and code availability This protocol reports original code hosted on GitHub (https://github.com/rcabezalaguna/liver_FF_quant). Acknowledgments This study is funded by Grant PID2022.141506OB.I00 funded by MCIU/AEI/10.13039/501100011033 and by ERDF, EU. A.C.-C. is supported by the the Junta de Andalucía, Consejería de Salud y Consumo (RHJ-0098-2024). A.C.-J. is supported by the Spanish Ministry of Universities (FPU21/01161). Project PI25/01657, funded by Instituto de Salud Carlos III (ISCIII) and co-funded by the European Union (A.C.-C.). Author contributions J.R.R. and I.L. designed the research. A.C.-C., M.C., A.L.-V., and R.C. conducted the image processing. A.C.-C. and A.C.-J. wrote the manuscript. J.L.M.-R., M.C., and R.C. provided technical support. All authors read and approved the final version of the manuscript. Declaration of interests The authors declare no competing interests. Footnotes Supplemental information can be found online at https://doi.org/10.1016/j.xpro.2026.104585. Contributor Information Alba Camacho-Cardenosa, Email: acamachocardenos@ugr.es. Jonatan R. Ruiz, Email: ruizj@ugr.es. References Associated Data Supplementary Materials Methods Video S1. Edit liver segmentation with 3DSlicer, related to Data analysis Methods Video S2. Image registration process with 3DSlicer, related to Data analysis Methods Video S3. Quantification of fat fraction with PyRadiomics, related to Data analysis Data Availability Statement This protocol reports original code hosted on GitHub (https://github.com/rcabezalaguna/liver_FF_quant).