One-click co-registered T2 mapping and dual black- and bright-blood late gadolinium enhancement MRI for comprehensive assessment of myocardial injury after acute STEMI

Abstract

Aims

In acute ST-segment elevation myocardial infarction, ischaemia and reperfusion lead to a cascade of myocardial injury that can be characterized by cardiac magnetic resonance (CMR) imaging, including coagulation necrosis, oedema, papillary muscle damage, microvascular obstruction, and intramyocardial haemorrhage. Conventional CMR protocols require multiple sequences to be performed and complicated analysis. This study evaluates SPOT-MAPPING, a sequence that acquires co-registered T2 maps and dual bright- and black-blood late gadolinium enhancement (LGE) images in a single scan.

Methods and results

SPOT-MAPPING employs a single-shot, ECG-triggered 2D sequence alternating between bright- and black-blood LGE imaging with varying T2 weightings. We prospectively enrolled 20 STEMI patients undergoing CMR at 1.5 T within 4–7 days post-emergent coronary intervention. SPOT-MAPPING’s segmentation times and reproducibility of myocardial injury markers (oedema, scar size, transmurality, papillary muscle damage) were assessed against conventional T2 mapping and phase-sensitive inversion recovery (PSIR). SPOT-MAPPING halved left ventricular wall segmentation time (∼3 min) while maintaining high reproducibility for oedema, scar size, and transmurality (ICC > 0.8). It improved papillary muscle damage detection over PSIR (8 vs. 3 patients) and showed comparable T2 values with conventional T2 mapping (remote: 45.0 ± 3.6 ms vs. 45.9 ± 3.7 ms, P = 0.746; oedema: 67.6 ± 10.3 ms vs. 71.8 ± 8.6 ms, P = 0.373). Agreement with PSIR for scar quantification was strong (mean bias: volume +1.5 mL, size +2.9%, transmurality +2.8%). SPOT-MAPPING demonstrated higher inter- and intraobserver reproducibility for scar size as a percentage of oedema volume compared with PSIR combined with conventional T2 mapping (ICC = 0.98 vs. 0.89 and 0.93 vs. 0.85).

Conclusion

SPOT-MAPPING offers a time-efficient and reproducible CMR method for myocardial injury assessment post-STEMI.

Graphical Abstract

Graphical Abstract.

Introduction

Accurately predicting the risk of future cardiovascular events after acute ST-segment elevation myocardial infarction (STEMI) remains a critical goal in cardiology.¹ Identifying high-risk patients is essential for optimizing prognosis and guiding therapeutic strategies. Risk stratification after STEMI typically integrates clinical risk scores, biomarkers, and multimodal imaging, including echocardiography and cardiovascular magnetic resonance (CMR) imaging.^1–3

CMR has emerged as a powerful tool for comprehensive risk assessment, capable of evaluating key prognostic markers of myocardial and microvascular damage after STEMI.^4–6 These include regional and global left ventricular (LV) function, myocardial oedema, infarct size, infarct transmurality, microvascular obstruction (MVO), intramyocardial haemorrhage (IMH), and papillary muscle infarction.^7–9 Numerous studies have demonstrated that these CMR-derived markers are strong predictors of major adverse cardiac events, including heart failure and mortality.^2,8 Beyond prognostic value, these imaging biomarkers also serve as surrogate endpoints in cardioprotection studies and clinical trials.¹⁰

However, conventional CMR-based risk assessment has significant limitations. Comprehensive evaluation requires multiple distinct imaging sequences—cine imaging, T2 mapping, and late gadolinium enhancement (LGE)—leading to prolonged scan times, repeated breath-holds, and increased complexity for both technicians and medical professionals alike. These challenges may contribute to suboptimal risk stratification and increased variability in clinical decision-making. Additionally, bright-blood LGE¹¹ suffers from poor contrast at the blood-scar interface, making it challenging to accurately assess subendocardial scars^12,13 or to visualize papillary muscle damage,¹⁴ both of which are highly relevant in the acute STEMI population. Recent advances in black-blood LGE imaging have demonstrated superior scar demarcation, offering new opportunities for more precise evaluation of ischaemic injury.¹²

Another key challenge remains the accurate assessment of the myocardial salvage index or the proportion of the oedema that was infarcted (POI). POI often serves as a secondary endpoint in clinical trials, reflecting injury and cardioprotective therapy effectiveness.^10,15 This marker is derived from both oedema-sensitive CMR sequences (pre-contrast T2-weighted imaging or T2 mapping) and post-contrast LGE.² However, since these two sequences are acquired at different time points with varying spatial resolution and slice coverage, spatial misalignment introduces uncertainty in POI measurements. A fully co-registered imaging approach integrating T2 mapping and advanced LGE with matched scanning parameters could enhance prognostic precision, simplify data acquisition, improve patient comfort, and facilitate more consistent and clinically actionable risk stratification in STEMI patients.

This study aimed to develop, optimize, and evaluate the preliminary clinical performance of SPOT-MAPPING—a novel time-efficient CMR sequence that simultaneously acquires co-registered T2 maps and dual bright- and black-blood LGE images in a single scan. The technology was validated through numerical simulations and an in vivo animal model of radiofrequency ablation lesion. Its technical feasibility, diagnostic accuracy, and potential for clinical adoption were tested in a prospective cohort of 20 patients with acute STEMI.

Methods

SPOT-MAPPING sequence

Image acquisition: We propose a novel 2D imaging technique, SPOT-MAPPING (Scar-specific imaging with Preserved myOcardial visualizaTion and T2 MAPPING), designed to simultaneously acquire co-registered T2 maps and dual bright- and black-blood LGE images in an interleaved fashion (Figure 1). SPOT-MAPPING utilizes a single-shot electrocardiogram-triggered, 2D balanced steady-state free-precession sequence (bSSFP), alternating between black- and bright-blood contrasts on consecutive heartbeats. During odd heartbeats, a non-selective 180° inversion pulse (duration = 10 ms) is directly followed by an adiabatic T2 preparation module to generate black-blood contrast.¹⁶ The adiabatic T2-prep module included two non-selective 90° hard pulse (total duration = 1 ms) and two adiabatic 180° refocusing pulses (total duration = 26 ms).¹⁷ During even heartbeats, only the T2 preparation module was applied, followed directly by readout, to generate bright-blood contrast. The black-blood images suppressed signals from both healthy myocardium and blood, enhancing scar detection, while the bright-blood images preserved blood-myocardium contrast, aiding in scar localization, and LV wall segmentation (Figure 1).

Figure 1.

Overview of the SPOT-MAPPING image acquisition. SPOT-MAPPING collects 2D co-registered bright-blood and black-blood late gadolinium enhancement images and T2 maps in a single sequence enabling the extraction of myocardial scar, myocardial oedema, papillary muscle enhancement, and microvascular obstruction/intramyocardial haemorrhage. Abbreviations: ECG, electrocardiogram; GRAPPA, generalized autocalibrating partially parallel acquisitions; IMH, intramyocardial haemorrhage; LV, left ventricular; MVO, microvascular obstruction; PMD, papillary muscle damage.

T2 mapping is achieved by acquiring five bright-blood images with varying T2 weightings, using progressively longer T2 preparation durations (no preparation, 27 ms, 35 ms, 40 ms, and 55 ms). For each slice position, 10 single-shot images (five black-blood and five bright-blood images) were acquired in mid-diastole over 12 heartbeats, with the first two heartbeats serving as dummy cycles. Full ventricular coverage was obtained across multiple breath-holds.

Detailed imaging parameters are provided in Table 1, including the following: 15 slices (range: 12–20), 15 × 1.5 mm in-plane resolution, 8 mm slice thickness, flip angle = 60°, GRAPPA ×2 (36 calibration lines), ∼160 ms acquisition window, echo time = 1.2 ms, repetition time = 2.9 ms, bandwidth = 849 Hz/pixel. The inversion time (TI) is determined from a prior scout acquisition.¹⁸

Table 1.

Imaging sequence parameters for pre-contrast T2 mapping and post-contrast phase-sensitive inversion recovery (PSIR) and SPOT-MAPPING

Parameter	T2 mapping	PSIR	SPOT-MAPPING
Field strength, Tesla		1.5
Readout	2D single-shot bSSFP
Cardiac control	Electrocardiogram triggering
Respiratory control	Breath-held
Slice orientation	Short-axis
Contrast agent	Yes
Reconstruction resolution, mm	1.5 × 1.5 × 8
Number of slices, median [min–max]	15 [12–20]
Phase partial Fourier	6/8
Phase resolution, %	72
Field of view phase, %	82.8
Scan acceleration	GRAPPA ×2
Ordering scheme	Linear
Repetition time, ms	2.5	2.9	2.9
Echo time, ms	1.1	1.2	1.2
Recovery heartbeats	2	0	0
Acquisition window, ms	145	160	160
Acquisition time, heartbeat/slice	7	12	12
Flip angle, odd heartbeats, degrees	70	50	60
Flip angle, even heartbeats, degrees	—	8	60
Inversion time, ms, mean ± SD	—	289 ± 26	116 ± 36
T2-prep duration, odd heartbeats, ms	[0, 27, 55]	—	27
T2-prep duration, even heartbeats, ms	—	—	[0, 27, 35, 40, 55]
Bandwidth, Hz/pixel	1184	781	849

Image reconstruction: The 10 single-shot images are reconstructed with a parallel imaging GRAPPA algorithm,¹⁹ followed by image averaging to enhance signal-to-noise ratio. Each single-shot image is reconstructed on-the-fly at the scanner and displayed in real-time for immediate assessment of contrast and artefacts. Four sets of co-registered images are generated as follows:

1. Averaged black-blood images for scar detection.

2. Averaged bright-blood images for anatomy information and LV wall segmentation.

3. A coloured-fusion of the bright- and black-blood images for optimal myocardial scar localization within the heart anatomy.

4. T2 maps derived from the bright-blood images using a two-parameter curve fitting.²⁰

Simulation study

Extended phase graph (EPG) simulations²¹ were conducted to analyse the signal evolution of the proposed SPOT-MAPPING sequence as a function of time and to help inform optimal parameter ranges for in vivo studies. The sequence was simulated for T1/T2 characteristics of post-contrast blood (460/140 ms), viable myocardium (670/42 ms) and scar (440/44 ms) at 1.5 T, and for a heart rate of 60 beats per minute. The influences of the timing parameters TI (ranging from 1 to 200 ms) and T2 preparation times (ranging from 1 to 100 ms) were examined for their effectiveness in suppressing both blood and myocardium signals and in providing sufficient scar contrast. Simulations were performed in MATLAB (version R2024a, The MathWorks, Natick, MA, USA).

Animal study

The optimized SPOT-MAPPING sequence was first validated in an animal model to assess its accuracy in characterizing damaged tissue, using gold standard histology and gross pathology as a reference standard. The study followed EU Directive 2010/63/EU on animal protection and was approved by the local ethical committee (CEEA50, APAFIS#39503). Experiments were conducted at IHU LIRYC with regulatory approvals.

Radiofrequency ablations were performed in a 7-year-old sheep by an experienced interventional cardiologist (K.V.) to mimic human infarction. An 8.5F force-sensing catheter (INTELLANAV STABLEPOINT™) was inserted into the LV and guided via fluoroscopy and echocardiography (Vivid E95, GE Healthcare, Chicago, IL-US). Radiofrequency ablation was conducted using a MAESTRO 4000™ system in power-controlled mode (30 W, 60 s, 17 mL/min irrigation, >5 g contact force).

CMR was performed immediately after ablation (within 1 h after the last radiofrequency application). Images were acquired in mid-diastole at end-expiration using a 32-channel spin coil and an 18-channel body coil. Pre-contrast T2 mapping²⁰ and post-contrast whole-heart phase-sensitive inversion recovery (PSIR)²² and SPOT-MAPPING short-axis images were collected 12 min after intravenous injection of 0.2 mmol/kg gadoterate meglumine (Dotarem®, Guerbet).

After imaging, the animal was euthanized, and the heart explanted. The LV was sliced (5 mm sections) and stained for 15 min at 37°C with 1% triphenyltetrazolium chloride (TTC) to reveal ablation lesions and then fixed in 4% paraformaldehyde for histological analysis. Tissue was dehydrated (HistoCore PEARL, Leica Biosystems, Wetzlar, Germany), embedded in paraffin, and sectioned (6 μm). Masson’s trichrome staining identified coagulation necrosis (blue) and adjacent viable myocardium (purple). Slices were examined at 20× magnification (Axio Scan Z1, Zeiss, Oberkochen, Germany). SPOT-MAPPING images were visually assessed for quality and effectiveness in attenuating blood and healthy myocardium signals while enhancing lesion signals. A visual analysis evaluated the spatial congruence between CMR, Masson trichrome, and TTC staining.

Patients study

Population and study design

From December 2023 to December 2024, we prospectively recruited 26 STEMI patients treated with primary percutaneous coronary intervention (PPCI) at our institution. STEMI diagnosis and patient management followed the ACCF/AHA and ESC guidelines.²³ CMR imaging was performed between 4 and 7 days post-PPCI. Exclusion criteria included age <18 years, previous myocardial infarction, history of severe renal failure, allergic reaction to gadolinium-based contrast agents, presence of a non-MR-conditional implantable device, inability to lie supine for 50 min, pregnancy, breastfeeding, claustrophobia, and inability to provide informed consent. The study was approved by the Biomedical Research Ethics Committee, and all participants provided informed consent.

Cardiac magnetic resonance protocol

All patients underwent CMR in the supine position using a 1.5 T clinical system (MAGNETOM Aera, Siemens Healthcare, Erlangen, Germany, software version VE11C), equipped with a dedicated 32-channel spine coil and an 18-channel body coil. The CMR protocol comprised standard cine imaging in two-, three-, and four-chamber views, along with a stack of contiguous short-axis slices encompassing the ventricles. Reference pre-contrast T2 mapping was performed using a T2-prepared bSSFP sequence in a stack of adjacent 8-mm-thick short-axis slices covering the whole LV.²⁰ LGE imaging was conducted 12 min post-administration of 0.2 mmol/kg gadoteric meglumine (Dotarem, Guerbet, France) using both the proposed SPOT-MAPPING sequence and a breath-held PSIR²² sequence (selected in a random order) in a short-axis stack of contiguous slices covering the ventricle. T2 mapping, PSIR, and SPOT-MAPPING images were collected at the same slice positions. TIs were meticulously adjusted to null viable myocardium for PSIR and to null both blood and myocardium for SPOT-MAPPING. Parameters for the sequences are detailed in Table 1.

CMR image analysis

Image analysis was performed in accordance with the 2020 recommendations of the Society for Cardiovascular Magnetic Resonance (SCMR),²⁴ using CVI42 software (Circle Cardiovascular Imaging, Calgary, Canada). Although all images were acquired during breath-hold, raw images were visually assessed for misalignment due to motion, mis-triggering, partial volume effects, or artefacts. Any slices exhibiting these issues were excluded from analysis. Left and right ventricular ejection fractions were analysed from end-diastolic and end-systolic short-axis cine views according to current guidelines.²⁵ Volumes were indexed to body surface area.

Two experienced cardiac radiologists (H.C. and S.S., with 22 and 6 years of CMR experience, respectively) independently delineated the endocardial and epicardial borders on PSIR, T2 mapping, and bright-blood SPOT-MAPPING images. Hyperenhanced scar regions were manually segmented on PSIR and black-blood SPOT-MAPPING images, while myocardial oedema was segmented on T2 mapping and SPOT-MAPPING images. Trabecular tissue and papillary muscles were carefully excluded from LV wall segmentation. Both observers were blinded to each other’s annotations. One of the experts (H.C.) performed a second, independent re-annotation of the dataset over a month later to assess intraobserver variability. Manual segmentations, along with the derived clinical measurements, served as the gold standard for comparison. Segmentation processing times, acquisition durations (including pauses between breath-holds), and reconstruction times were recorded.

The accuracy of LV wall and scar segmentations was assessed using the DICE similarity coefficient (0% indicating no overlap, 100% indicating perfect agreement).²⁶ Intra- and interobserver reproducibility were assessed. The distribution of scar across the 16 segments of the American Heart Association (AHA) model was reported.

Myocardial T2 values of oedema and remote myocardium were reported, as defined by the SCMR recommendations,²⁷ from the reference T2 mapping and SPOT-MAPPING sequences. A region of interest (ROI) was drawn in the central layer of remote myocardium, while oedema was quantified as myocardium with a T2 value ≥2 standard deviations above the remote myocardium mean (≥ upper limit of the 95% confidence interval). Contours were manually adjusted to minimize partial volume effects. When possible, the remote myocardial ROI was positioned 180° away from the infarcted segments, in regions without LGE and with normal wall motion on cine imaging. The effect of heart rate on remote T2 values was assessed for both reference T2 mapping and SPOT-MAPPING.

MVO/IMH was visually defined on a segmental basis as a hypointense core embedded within hyperenhanced myocardium on PSIR images. MVO/IMH was also assessed using SPOT-MAPPING, where it appeared as a dark signal within hyperenhanced myocardium on black-blood images and exhibited lower T2 values than the surrounding oedematous tissue on T2 maps. If present, MVO/IMH was included in hyperenhanced myocardium for scar size and oedema quantification and was also manually delineated on short-axis slices.

The presence or absence of papillary muscle enhancement was noted for both PSIR and SPOT-MAPPING. For PSIR, LGE images were compared with cine images to ensure accurate identification of papillary muscles. For SPOT-MAPPING, LGE within the LV blood pool on black-blood images was superimposed on corresponding anatomical bright-blood images to confirm papillary muscle localization.

The following quantitative measures were reported: LV wall volume (mL), oedema volume (mL) and size (% of LV mass), scar volume (mL) and size (% of LV mass and % of oedema volume), scar transmurality (%), and the presence of papillary muscle enhancement and MVO/IMH. Scar transmurality was assessed using the established centreline chord method,²⁸ which segments each myocardial slice into 100 evenly distributed chords that are crossing perpendicularly a centreline drawn at equidistance to endocardial and epicardial contours and calculates infarct extent along the length of each chord. Average transmurality was determined by considering only chords where the infarct size was at least 1%.

Statistical analysis

Statistical analysis was performed using SPSS Statistics v28 (IBM Corp., Armonk, NY). Descriptive statistics were reported. Normality of continuous variables was assessed using the Shapiro–Wilk test. Normally distributed continuous variables are presented as mean ± SD, while non-normally distributed variables are expressed as median [interquartile range, Q1–Q3]. Categorical variables are reported as proportions (%). For dependent continuous variables, paired-sample parametric (paired Student’s t-test) or nonparametric (Wilcoxon signed-rank test) tests were applied based on normality. Independent continuous variables were compared using independent-sample parametric (unpaired Student’s t-test) or nonparametric (Mann–Whitney U test) tests depending on data normality. Dependent categorical variables were compared using the paired-sample McNemar test (χ²).

Intra- and interobserver variability was assessed using the intraclass correlation coefficient (ICC) with 95% confidence intervals. Agreement was classified as poor (ICC < 0.50), moderate (0.50–0.75), good (0.75–0.90), or excellent (>0.90). Bland–Altman analysis was performed to assess systematic bias and the relationship between observed discrepancies. Linear regression analysis was used to examine the relation between T2 values and heart rate in patients. The regression equation, slope, intercept, and coefficient of determination (R²) were reported for each model. A P value <0.05 was considered statistically significant.

Results

Simulation study

Supplementary data online, Figure S1 shows the EPG simulation outcomes that were used to guide subsequent in vivo acquisitions. For black-blood imaging, the combination of an inversion pulse alongside a T2 preparation module validated the near-complete suppression of both remote myocardium and blood signals, where their zero planes intersect at a T2 preparation duration of 27 ms, which is the shortest achievable duration for this pulse. At this intersection point, viable myocardium and blood signals are effectively nulled while scar signal remains positive.

Animal study

Selected SPOT-MAPPING images from the sheep model are presented in Supplementary data online, Figure S2A. The images are juxtaposed with slice-matched Masson trichrome histology and TTC staining images (see Supplementary data online, Figure S2B-C). PSIR imaging demonstrated a conspicuous lack of contrast for lesion tissue adjacent to the blood pool, an issue effectively addressed by SPOT-MAPPING. There was a good spatial concordance between SPOT-MAPPING and ex vivo images regarding the localization of myocardial injuries, which was confirmed at post-mortem dissection on TTC staining images. Masson trichrome staining revealed extensive coagulation necrosis within the ablation lesions, thereby validating the observed hyperintensity on SPOT-MAPPING images. Myocardial oedema, surrounding lesions, was present on both reference T2 mapping (mean remote: 44 ms vs. oedema: 88 ms) and SPOT-MAPPING images (mean remote: 41 ms vs. oedema: 77 ms).

Patients study

Patient characteristics and sequence timings

The final study population consisted of 20 STEMI patients (15% female), with a mean age of 62 ± 13 years, all treated with PPCI (see Supplementary data online, Figure S3). Six patients were excluded due to incomplete CMR (n = 4) or poor image quality (n = 2). Residual motion was noted in one patient (1 slice), five patients (6 slices), and four patients (5 slices) on T2 mapping, PSIR, and SPOT-MAPPING sequences, respectively. Baseline demographics, angiographic characteristics, clinical presentation, and CMR parameters are summarized in Table 2. CMR imaging was performed at a mean of 5.6 ± 2.0 days post-PPCI. MVO/IMH was detected in nine patients (45%). Median infarct size by PSIR imaging was 17.5% [Q1–Q3: 9.6–24.5%] of LV myocardial mass. The average oedema size by T2 mapping was 22.4 ± 6.3% of LV myocardial mass. Acquisition times for T2 mapping, PSIR, and SPOT-MAPPING sequences were 4 min7 s ± 44 s, 5 min27 s ± 59 s, and 5 min23 s ± 54 s, respectively. The entire reconstruction process for SPOT-MAPPING, including image averaging, was completed in under 10 s in the scanner console.

Table 2.

Patient demographics and characteristics

Clinical characteristics
Data
Number of patients	20
Number of images for analysis	864
Demographics
Age, year	62 ± 13
Female, n (%)	3 (15)
Weight, kg	78 [70–88]
Height, cm	173 ± 6
BMI, kg/m2	27.5 ± 5.5
Symptom onset to reperfusion, min	287.9 ± 176.6
Delays
Time from onset to first call or contact, hours	1.8 [0.5–5.0]
Time from onset to ICU admission, hours	3.0 [2.0–5.5]
Time from onset to balloon, hours	4.0 [2.6–6.2]
Time from door to balloon, hours	1.3 [0.5–16.8]
Time from PPCI to CMR, days	5.6 ± 2.0
Comorbidity
CAD family history, n (%)	4 (20)
Hypertension, n (%)	6 (32)
Dyslipidaemia, n (%)	12 (80)
Diabetes mellitus, n (%)	5 (25)
Smoking history, n (%)	15 (75)
Previous myocardial infarction, n (%)	0 (0)
Clinical presentation
Admission Killip class
1	12 (60)
2	5 (25)
3	3 (15)
4	0 (0)
Pre-PPCI TIMI flow grade
0–1	17 (85)
2	3 (15)
3	0 (0)
GRACE score	124.2 ± 29.2
Systolic blood pressure, mm Hg	122.4 ± 35.0
Diastolic blood pressure, mm Hg	92.7 ± 35.6
Heart rate, beats/min	66.1 ± 14.0
Laboratory data
NT-proBNP, pg/mL	40.0 [18.5–111.5]
Peak troponin I, μ/L	1261.5 [496.3–2249.8]
LDL cholesterol, g/L	1.2 [1.1–1.4]
Invasive coronary angiography
Infarct artery
Left anterior descending, n (%)	9 (45)
Right coronary artery, n (%)	8 (40)
Left circumflex, n (%)	7 (35)
Multivessel disease, n (%)	7 (35)
Stent implanted, n (%)	19 (95)
Post-PPCI TIMI flow grade 3, n (%)	19 (95)
Secondary medical prevention discharge therapy
Beta-blockers, n (%)	20 (100)
ACE-I or ARB, n (%)	7 (35)
Statins, n (%)	16 (80)
Aspirin, n (%)	19 (95)
Sacubitril/valsartan, n (%)	5 (25)
Warfarin, n (%)	3 (15)
Prasugrel, n (%)	5 (25)
Spironolactone or eplerenone, n (%)	11 (55)
Ticagrelor, n (%)	10 (50)
CMR characteristics
Heart rate, beats/min	63.4 ± 10.0
Infarct size by PSIR, %LV mass	17.5 [9.6–24.5]
MVO/IMH presence by PSIR, N (%)	9 (45)
MVO/IMH size by PSIR, %LV mass	6.1 ± 6.8
MVO/IMH extent by PSIR, number of segments	4.8 ± 3.6
PM enhancement presence by PSIR, n (%)	3 (15)
Oedema by T2 mapping, %LV mass	22.4 ± 6.3
Remote T2 by T2 mapping, ms	45.0 ± 3.6
LV mass by PSIR, g	121.1 [104.4–142.6]
LV thrombus, n (%)	0 (0)
Cardiac function
LVEF, %	43.7 ± 10.2
LVEDVi, mL/m2	78.0 [67.5–90.8]
LVESVi, mL/m2	42.0 [37.0–53.0]
Severe LVEF impairment (LVEF < 35%), n (%)	2 (10)
RVEF, %	48.5 [42.5–53.0]
RVEDVi, mL/m2	74.0 [59.8–82.3]
RVESVi, mL/m2	36.0 [31.0–45.0]

Abbreviations: ACE-I, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; BMI, body mass index; CAD, coronary artery disease; CMR, cardiac magnetic resonance; CRP, C-reactive protein; hs-TnT, high-sensitivity cardiac troponin T; ICU, intensive care unit; IMH, intramyocardial haemorrhage; LDL, low-density lipoprotein; LV, left ventricle; LVEF, left ventricular ejection fraction; LVEDVi, left ventricular end-diastolic index; LVESVi, left ventricular end-systolic index; MVO, microvascular obstruction; PM, papillary muscle; PPCI, primary percutaneous coronary intervention; PSIR, phase-sensitive inversion recovery; RVEF, right ventricular ejection fraction; RVEDVi, right ventricular end-diastolic index; RVESVi, right ventricular end-systolic index; TIMI, thrombolysis in myocardial infarction.

Examples of whole-heart short-axis images acquired in two STEMI patients using reference pre-contrast T2 mapping, reference PSIR, and the proposed SPOT-MAPPING sequences are shown in Figure 2 and Supplementary data online, Figure S4.

Figure 2.

Whole-heart short-axis images acquired using reference pre-contrast T2 mapping, reference PSIR, and the proposed SPOT-MAPPING sequences in a 58-year-old male patient with acute inferior ST-STEMI in the right coronary artery (RCA). Cardiac MRI was performed 5 days post-revascularization. T2 maps show elevated T2 values (reference T2 mapping: 43.9 ms vs. 64.4 ms; SPOT-MAPPING: 47.1 ms vs. 67.6 ms), indicating transmural myocardial oedema. LGE sequences (PSIR and SPOT-MAPPING) reveal subendocardial enhancement in the RCA territory, with no sign of microvascular obstruction/intramyocardial haemorrhage. SPOT-MAPPING demonstrates strong visual concordance with reference T2 mapping for oedema detection and with PSIR for infarct visualization. Additionally, its co-registered nature allows for quantification of scar size as a percentage of myocardial oedema, yielding a value of 27.6%. Abbreviations: PSIR, phase-sensitive inversion recovery.

LV wall segmentation time and reproducibility

Segmentation times

LV wall segmentation required approximately the same duration for T2 mapping, PSIR, and SPOT-MAPPING, averaging 3 min per sequence. Consequently, the proposed SPOT-MAPPING, which combines T2 mapping and LGE, achieved a two-fold reduction in LV wall segmentation time.

Intra- and interobserver reproducibility

Dice scores for the LV wall are summarized in Table 3. No significant differences were observed between PSIR and SPOT-MAPPING for intraobserver (89.3 ± 2.1% vs. 88.9 ± 1.5%, P = 0.517) or interobserver (84.7 ± 3.4% vs. 83.6 ± 2.5%, P = 0.055) agreement. SPOT-MAPPING demonstrated higher reproducibility in LV wall volume measurements compared with PSIR (Table 4). Intraobserver reproducibility showed a bias of −5.2 ± 5.0 mL (ICC [95% CI]: 0.97 [0.94–0.99], P = 0.596) for SPOT-MAPPING vs. −6.9 ± 10.2 mL (ICC [95% CI]: 0.93 [0.83–0.97], P = 0.493) for PSIR. Similarly, interobserver reproducibility was higher with SPOT-MAPPING, with a bias of −19.0 ± 11.6 mL (ICC [95% CI]: 0.74 [0.46–0.89], P = 0.046), compared with PSIR, which had a bias of −20.0 ± 15.8 mL (ICC [95% CI]: 0.70 [0.39–0.87], P = 0.048). Examples of LV wall segmentations on SPOT-MAPPING images from four patients with acute STEMI are shown in Figure 3.

Table 3.

Dice scores for intraobserver and interobserver for left ventricular wall and left ventricular scar segmentations

Parameter	Dice score (%)
	Intraobserver	Interobserver
Left ventricular wall
T2 mapping	85.2 ± 3.4	81.6 ± 5.2
PSIR	89.3 ± 2.1	84.7 ± 3.4
SPOT-MAPPING	88.9 ± 1.5	83.6 ± 2.5
Left ventricular scar
PSIR	83.2 ± 6.6	76.2 ± 7.7
SPOT-MAPPING	85.6 ± 4.2	77.8 ± 9.4

Table 4.

Intra- and interobserver reproducibility of left ventricular (LV) volume, oedema volume and size, and scar size, volume and transmurality, and scar size as a percentage of oedema volume

	Intraobserver			Interobserver
	ICC [95% CI]	Bias (SD)	P value	ICC [95% CI]	Bias (SD)	P value
LV wall volume, mL
T2 mapping	0.84 [0.64–0.93]	−16.0 (11.2)	0.134	0.77 [0.52–0.90]	−22.0 (10.0)	0.047
PSIR	0.93 [0.83–0.97]	−6.9 (10.2)	0.493	0.70 [0.39–0.87]	−20.0 (15.8)	0.048
SPOT-MAPPING	0.97 [0.94–0.99]	−5.2 (5.0)	0.596	0.74 [0.46–0.89]	−19.0 (11.6)	0.046
Oedema volume, mL
T2 mapping	0.95 [0.88–0.98]	−1.0 (1.7)	0.615	0.85 [0.66–0.94]	−1.9 (2.8)	0.314
SPOT-MAPPING	0.98 [0.94–0.99]	−1.0 (1.6)	0.711	0.82 [0.61–0.93]	−3.2 (3.1)	0.167
Oedema size, % of LV mass
T2 mapping	0.89 [0.74–0.95]	+2.5 (1.9)	0.226	0.82 [0.61–0.93]	+2.6 (3.0)	0.218
SPOT-MAPPING	0.92 [0.82–0.97]	+0.2 (2.5)	0.937	0.74 [0.46–0.89]	+2.9 (4.1)	0.180
Scar volume, mL
PSIR	0.99 [0.98–1.00]	−0.6 (2.1)	0.912	0.96 [0.90–0.98]	−2.5 (3.7)	0.607
SPOT-MAPPING	0.97 [0.92–0.99]	−2.1 (2.7)	0.970	0.88 [0.72–0.95]	−3.6 (5.5)	0.380
Scar size, % of LV mass
PSIR	0.98 [0.95–0.99]	+1.0 (2.7)	0.820	0.87 [0.72–0.95]	+2.5 (7.4)	0.605
SPOT-MAPPING	0.95 [0.88–0.98]	−2.0 (3.6)	0.630	0.92 [0.82–0.97]	−1.5 (4.8)	0.704
Scar transmurality, %
PSIR	0.91 [0.79–0.96]	+3.0 (6.1)	0.540	0.75 [0.48–0.89]	+5.0 (9.7)	0.303
SPOT-MAPPING	0.94 [0.87–0.98]	−2.4 (4.2)	0.598	0.91 [0.79–0.96]	−1.6 (5.5)	0.696
Scar size, % of oedema volume
PSIR + T2 MAPPING	0.89 [0.75–0.96]	−14.0 (31.1)	0.546	0.85 [0.66–0.94]	−13.0 (39.8)	0.589
SPOT-MAPPING	0.98 [0.95–0.99]	−5.8 (6.9)	0.679	0.93 [0.85–0.97]	−1.3 (17.5)	0.932

Figure 3.

SPOT-MAPPING images from four patients with acute ST-STEMI. Endocardial and epicardial contours are delineated on the bright-blood images and propagated onto the co-registered black-blood and T2 mapping images. This enables comprehensive assessment of myocardial scar, myocardial oedema, papillary muscle damage, and microvascular obstruction/intramyocardial haemorrhage.

T2 mapping accuracy

T2 values of remote myocardium and myocardial oedema

There was no difference in T2 values between pre-contrast T2 mapping and post-contrast SPOT-MAPPING for remote myocardium (45.0 ± 3.6 ms vs. 45.9 ± 3.7 ms, P = 0.746) and myocardial oedema (67.6 ± 10.3 ms vs. 71.8 ± 8.6 ms, P = 0.373, Supplementary data online, Figure S5A). Bland–Altman analysis demonstrated good agreement between T2 mapping and SPOT-MAPPING for remote myocardium (mean bias +0.89 ms; 95% CI: −0.14 ms to 1.92 ms, P = 0.446, Supplementary data online, Figure S5B) and for myocardial oedema (mean bias +4.2 ms; 95% CI: 0.95 ms to 7.37 ms, P = 0.175, Supplementary data online, Figure S5C).

Reproducibility of myocardial oedema volume and size

Both T2 mapping and SPOT-MAPPING demonstrated excellent intraobserver agreement for oedema volume (ICC > 0.95) and good agreement for oedema size (ICC > 0.89), with minimal bias and non-significant differences (Table 4). Interobserver agreement remained good for oedema volume (ICC > 0.80), while oedema size showed moderate-to-good agreement, with T2 mapping (ICC = 0.82) slightly outperforming SPOT-MAPPING (ICC = 0.74).

When comparing the two sequences, SPOT-MAPPING showed a slight negative bias compared with reference T2 mapping, with oedema size as a percentage of LV myocardial mass being lower by −2.1% (±3.4%, 95% CI: −3.7 mL to −0.5 mL, P = 0.011) (Figure 4A) and oedema volume showing a bias of −1.4 mL (±4.2 mL, 95% CI: −3.4 mL to 0.6 mL, P = 0.146) (Figure 4B).

Figure 4.

Bland–Altman plots for oedema size as a percentage of left ventricular myocardial mass (A), oedema volume (B), scar size as a percentage of left ventricular myocardial mass (C), scar volume (D), scar size as a percentage of oedema volume (E), and scar transmurality (F).

Influence of heart rate on T2 values

Heart rate did not significantly influence remote T2 values in reference T2 mapping, showing a weak and non-significant relationship (R² = 0.16, P = 0.078). In contrast, SPOT-MAPPING demonstrated greater sensitivity to heart rate variations, with a stronger and statistically significant association (R² = 0.55, P < 0.001, Supplementary data online, Figure S6).

LGE accuracy

Reproducibility of scar volume, scar size, and scar transmurality

Scar volume, scar size, and scar transmurality were highly reproducible for both SPOT-MAPPING and PSIR sequences (Table 4).

Intraobserver reproducibility for scar volume showed an ICC of 0.99 [0.98–1.00] for PSIR and 0.97 [0.92–0.99] for SPOT-MAPPING, with a bias of −0.6 ± 2.1 mL and −2.1 ± 2.7 mL, respectively. Interobserver reproducibility was similarly high (ICC: 0.96 [0.90–0.98] for PSIR and 0.88 [0.72–0.95] for SPOT-MAPPING), with biases of −2.5 ± 3.7 mL and −3.6 ± 5.5 mL, respectively.

Scar size as a percentage of LV mass also showed strong intraobserver agreement (ICC: 0.98 [0.95–0.99] for PSIR vs. 0.95 [0.88–0.98] for SPOT-MAPPING) with minimal bias (+1.0 ± 2.7% vs. −2.0 ± 3.6%, respectively). Interobserver reproducibility remained high (ICC: 0.87 [0.72–0.95] for PSIR vs. 0.92 [0.82–0.97] for SPOT-MAPPING), with biases of +2.5 ± 7.4% and −1.5 ± 4.8%, respectively.

For scar transmurality, SPOT-MAPPING demonstrated slightly higher intraobserver reproducibility (ICC: 0.94 [0.87–0.98]) compared with PSIR (ICC: 0.91 [0.79–0.96]), with biases of −2.4 ± 4.2% and +3.0 ± 6.1%, respectively. Interobserver agreement was also slightly higher with SPOT-MAPPING (ICC: 0.91 [0.79–0.96] vs. 0.75 [0.48–0.96] for PSIR), with biases of −1.6 ± 5.5% and +5.0 ± 9.7%, respectively.

Scar volume, scar size, and scar transmurality showed strong agreement between SPOT-MAPPING and the reference PSIR sequence (Figure 4). Bland–Altman analysis demonstrated a mean bias of +1.5 mL (±4.3 mL, 95% CI: −0.7 mL to 3.6 mL, P = 0.168) for scar volume, +2.9% (±4.1%, 95% CI: 1.0% to 4.8%, P = 0.749) for scar size, and +2.8% (±7.8%, 95% CI: −0.8% to 6.4%, P = 0.541) for scar transmurality, indicating no statistically significant differences between the two methods. The AHA distribution of scar obtained with PSIR and SPOT-MAPPING is shown in Supplementary data online, Figure S7.

MVO/IMH detection

MVO/IMH was identified in eight patients using PSIR and seven patients using SPOT-MAPPING. SPOT-MAPPING detected one MVO/IMH case that was missed by PSIR, while PSIR identified two cases that SPOT-MAPPING did not detect (χ² = 0.00, P = 1.00), indicating no significant difference in MVO/IMH detection between the two methods. MVO extent by PSIR was 4.8 ± 3.6 segments. T2 values in regions with MVO/IMH were not significantly different from those in remote myocardium for SPOT-MAPPING (46.7 ± 7.6 ms vs. 45.4 ± 3.4 ms, P = 0.717) or for the reference T2 mapping sequence (46.7 ± 9.0 ms vs. 46.6 ± 2.9 ms, P = 0.983). Examples of PSIR and SPOT-MAPPING images from four patients exhibiting MVO/IMH are presented in Figure 5.

Figure 5.

Reference PSIR images and proposed SPOT-MAPPING images from four patients with acute ST-elevation myocardial infarction. Arrows indicate areas of microvascular obstruction/intramyocardial haemorrhage. Abbreviations: PSIR, phase-sensitive inversion recovery.

Papillary muscle enhancement detection

Papillary muscle enhancement was observed in three patients (15%) with PSIR and in eight patients (40%) with SPOT-MAPPING. Out of 20 patients, SPOT-MAPPING classified five patients with positive PM that PSIR did not, whereas PSIR did not identify any cases missed by SPOT-MAPPING (χ² = 3.20, P = 0.074). Examples of images from six patients demonstrating papillary muscle damage are presented in Supplementary data online, Figure S8.

Myocardial scar extent as a ratio to oedema

SPOT-MAPPING demonstrated higher intraobserver and interobserver reproducibility for scar size as a percentage of oedema volume compared with PSIR combined with pre-contrast T2 mapping (Table 4). Intraobserver agreement was excellent for SPOT-MAPPING (ICC [95% CI]: 0.98 [0.95–0.99], bias = −5.8 ± 6.9%, P = 0.679) and moderate for PSIR + T2 mapping (ICC [95% CI]: 0.89 [0.75–0.96], bias = −14.0 ± 31.1%, P = 0.546). Similarly, interobserver reproducibility was higher with SPOT-MAPPING (ICC [95% CI]: 0.93 [0.85–0.97], bias = −1.3 ± 17.5%, P = 0.932) compared with PSIR + T2 mapping (ICC [95% CI]: 0.85 [0.66–0.96], bias = −13.0 ± 39.8%, P = 0.589). Bland–Altman analysis revealed a significant mean difference of −34.0% (±34.2%, 95% CI: −50.3% to −18.4%, P < 0.001) between SPOT-MAPPING and PSIR + pre-contrast T2 mapping with limits of agreement ranging from −100% to +33%, indicating variability between methods (Figure 4E). Figure 6 illustrates T2 maps, PSIR images, and SPOT-MAPPING images from four patients with acute STEMI, highlighting key clinical parameters such as LV wall volume and mass, myocardial oedema, and infarct characteristics, MVO/IMH extent, and the presence of papillary muscle enhancement.

Figure 6.

T2 maps, PSIR images, and SPOT-MAPPING images obtained from four patients with acute STEMI. Reported clinical parameters from the SPOT-MAPPING images include (i) left ventricular (LV) volume and mass; (ii) oedema volume, mass, and size as a percentage of LV mass; (iii) infarct volume, mass, and size as percentage of LV mass and oedema mass; (iv) microvascular obstruction (MVO)/intramyocardial haemorrhage (IMH) size as a percentage of LV mass; and (v) the presence of papillary muscle (PM) enhancement. Abbreviations: IMH, intramyocardial haemorrhage; LV, left ventricular; MVO, microvascular obstruction; PM, papillary muscle; PSIR, phase-sensitive inversion recovery.

Discussion

Our main findings are that SPOT-MAPPING:

1. Successfully integrated co-registered T2 mapping and dual bright- and black-blood LGE into a single scan, reducing overall imaging acquisition and LV wall segmentation time by two-fold compared with separate reference T2 mapping and PSIR sequences.

2. Demonstrated high intraobserver and interobserver reproducibility for LV wall volume, oedema, and scar measurements, with comparable or superior reliability to T2 mapping and PSIR.

3. Detected papillary muscle enhancement with improved sensitivity compared with PSIR.

4. Enabled a more robust measurement of myocardial scar extent as a ratio to oedema.

Weaknesses and strengths of the proposed SPOT-MAPPING sequence are in shown Table 5.

Table 5.

Performance comparison of cardiac magnetic resonance imaging techniques in acute STEMI evaluation

Sequence	Strengths	Weaknesses
T2 Mapping	Detects myocardial oedema Quantitative	Acquired separately from LGE (spatial misalignment) Manual segmentation
LGE (Bright-blood PSIR)	Standard for infarct visualization Quantifies infarct size and transmurality	Poor scar-blood contrast (especially subendocardial) Misses papillary muscle infarction Not co-registered with T2 mapping Manual segmentation
SPOT-MAPPING	Simultaneous, co-registered T2 mapping and dual bright- and black-blood LGE Shorter total scan time Fewer breath-holds Simplified scan planning (single-sequence setup) Higher sensitivity for papillary muscle infarct Enables direct scar-to-oedema ratio measurement Reduces segmentation time by ∼2× Ideal for automation	Heart rate dependent (needs future optimization) No fat suppression (confounding possible) Manual segmentation (automation possible)

Post-contrast myocardial T2 values

Our findings demonstrate a strong agreement between conventional pre-contrast T2 mapping and post-contrast SPOT-MAPPING for both remote myocardium (bias +0.89 ms) and myocardial oedema (bias +4.2 ms). SPOT-MAPPING exhibited slightly higher intraobserver consistency compared with T2 mapping. However, a heart rate dependency was observed with SPOT-MAPPING, likely due to the absence of recovery heartbeats necessary for full magnetization recovery, as used in standard T2 mapping. Future work will focus on incorporating EPG-based dictionary matching to improve accuracy and mitigate heart rate dependency effects.²⁹

Advantages of dual bright- and black-blood imaging

SPOT-MAPPING demonstrated reproducible myocardial quantification comparable to the reference PSIR sequence, with a slightly higher agreement for transmurality measurements (ICC = 0.94 vs. 0.91). Additionally, in line with previous dark-blood imaging studies,^14,30 we observed a higher sensitivity for detecting papillary muscle damage with black-blood SPOT-MAPPING compared with PSIR (40% vs. 15%). This is consistent with findings by Van De Heyning et al.¹⁴ who reported a similar trend (35% vs. 15%). However, given the absence of a definitive gold standard for papillary muscle damage detection, further validation with histological correlation in more animal models will be essential.

While SPOT-MAPPING enabled a two-fold reduction in LV wall segmentation time compared with standard techniques, segmentation remains manual. However, the bright-blood bSSFP-like contrast of SPOT-MAPPING is ideally suited for automated segmentation, which will further enhance workflow efficiency. Future work will focus on implementing AI-driven segmentation for both scar and myocardial oedema quantification.³¹

Combining T2 mapping and LGE imaging: more robust measurement of myocardial scar extent as a ratio to oedema

For the first time, SPOT-MAPPING enables the direct quantification of scar size as a percentage of oedema volume. This parameter has been proposed as a potential biomarker for risk stratification,² though its clinical utility remains debated.¹⁰ The true measurement of this metric could serve as a valuable endpoint for evaluating cardioprotective strategies, as post-reperfusion myocardial oedema, assessed by T2 mapping, is modifiable with intervention¹⁰ and may thus serve as a surrogate marker for myocardial injury and therapeutic efficacy.

Precise MVO/IMH assessment was facilitated by the co-registration of bright- and black-blood imaging with T2 mapping. Our results show that T2 values within MVO/IMH regions were similar to remote myocardium (46.7 ± 7.6 ms vs. 45.4 ± 3.4 ms), consistent with prior findings.³² This supports the notion that T2 mapping alone may not reliably differentiate MVO/IMH from non-infarcted myocardium, highlighting the importance of multi-contrast imaging.

Clinical implications

SPOT-MAPPING provides a single-sequence, one-click solution for the comprehensive analysis of myocardial scar, oedema, papillary muscle damage, and MVO/IMH. This streamlined approach has direct clinical implications, reducing patient breath-hold requirements, facilitating faster examinations, and improving workflow efficiency for MR technologists and interpreting medical professionals. Notably, this approach is well-suited for clinical trials, where fast, reproducible, and standardized CMR protocols are essential.

A potential future protocol for acute STEMI patients could involve administering contrast in the waiting room, followed by cine imaging for functional assessment, and SPOT-MAPPING for the evaluation of scar, oedema, papillary muscle, and MVO/IMH, reducing total scan time to under 20 min. This streamlined protocol would improve accessibility, particularly for patients unable to tolerate prolonged exams. Additionally, by reducing scan duration and complexity, SPOT-MAPPING may facilitate broader adoption of CMR in acute myocardial infarction management across centres.

Importantly, although our current cohort did not specifically target patients with significant arrhythmias or limited breath-hold capacity, we acknowledge that these clinical conditions represent potential challenges. SPOT-MAPPING employs a single-shot readout that is inherently more robust to arrhythmias compared with conventional segmented techniques. Furthermore, the sequence is compatible with in-plane motion-corrected averaging or motion-compensated reconstruction strategies, which have demonstrated efficacy in free-breathing LGE imaging.³³ We have previously integrated such a framework into our black-blood imaging pipeline¹³ and are currently investigating its application to SPOT-MAPPING. This would enable reliable performance in patients with arrhythmias or limited breath-hold capacity.

Beyond acute STEMI, SPOT-MAPPING also has potential applications in various cardiomyopathies, including cases with overlapping acute and chronic myocardial injury, such as myocarditis with prior infarction, amyloidosis, or Takotsubo syndrome coexisting with spontaneous coronary artery dissection³⁴ or chronic infarction.

Study limitations

This study has limitations. The single-centre design with a relatively small sample size introduces the possibility of centre-specific biases. Additionally, histological validation was limited to a single animal experiment, primarily used to visually compare contrast between LGE and T2 mapping sequences and to corroborate numerical simulation and in vivo human findings. A more comprehensive preclinical study comparing SPOT-MAPPING-derived scar and oedema quantification against histology and gross pathology is currently underway.

Fat suppression was not addressed in the sequence, which could lead to confounding effects from epicardial fat (see Figure 5). Future studies will explore Dixon-based fat suppression solutions.³⁵ Additionally, while most patients successfully performed breath-holds, residual respiratory motion artefacts were observed in five patients (affecting a total of six short-axis slices). A free-breathing version of SPOT-MAPPING, as previously explored for black-blood imaging,¹³ will be investigated in future work.

Perspectives and future work

SPOT-MAPPING currently acquires five T2-weighted images to generate T2 maps while preserving an adequate signal-to-noise ratio. Future efforts will focus on accelerating the sequence by reducing acquisitions to three images, similar to conventional T2 mapping, while balancing potential SNR trade-offs.

On the post-processing side, the co-registered nature of SPOT-MAPPING and its bSSFP-like contrast make it well-suited for fully automated segmentation using artificial intelligence. The integration of AI-based contour detection and scar quantification will further enhance efficiency and reproducibility. Moreover, while SPOT-MAPPING has thus far been implemented and validated exclusively on one vendor’s platform, broader clinical translation will require vendor-agnostic solutions. To facilitate this, we are actively working to adapt SPOT-MAPPING using the open-source Pulseq framework.³⁶

IMH is a key prognostic maker in acute STEMI,^9,37 with affected patients exhibiting higher major adverse cardiovascular event rates (16.4% vs. 7.0%).³⁸ Incorporating T2*-weighted IMH-sensitive imaging within SPOT-MAPPING could improve risk stratification. Furthermore, the addition of cine imaging in SPOT-MAPPING through free-running acquisition³⁹ could establish the technology as a true one-stop shop for post-infarction risk stratification. These advancements will be the focus of future work.

Lastly, as our patient cohort undergoes longitudinal follow-up, we will be able to correlate SPOT-MAPPING-derived biomarkers with clinical outcomes. This will help refine the prognostic relevance of the technique and identify the most clinically meaningful imaging biomarkers.

Conclusion

SPOT-MAPPING introduces a novel approach to CMR-based risk stratification after STEMI, offering a time-efficient and reproducible method for myocardial injury assessment. By seamlessly integrating co-registered T2 mapping with dual bright- and black-blood LGE into a single acquisition, SPOT-MAPPING may have the potential to enhance both clinical workflow and the consistency of imaging biomarkers in research trials. Future studies should explore its prognostic value in larger, multicentre cohorts and assess its role in guiding personalized therapeutic strategies for high-risk STEMI patients.

Supplementary data

Supplementary data are available at European Heart Journal – Imaging Methods and Practice online.

Consent

All patients provided informed consent prior to participation.

Funding

This work was supported by funding from the French National Research Agency under grant agreements Equipex MUSIC ANR-11-EQPX-0030, Programme d’Investissements d’Avenir ANR-10-IAHU04-LIRYC, ANR-22-CPJ2-0009-01, and from the European Research Council (ERC) under the European Union’s Horizon Europe research and innovation programme (Grant agreement No. 101076351).

Data availability

The data underlying this article will be shared on reasonable request to the corresponding author.

Lead author biography

Prof. Aurelien Bustin is a researcher in clinical MRI, cardiac imaging, and artificial intelligence applied to healthcare. Based at the IHU LIRYC, University of Bordeaux, France, he is also a visiting guest at the CHUV, Lausanne, Switzerland. He leads a research group dedicated to developing and translating advanced mathematical, imaging, and reconstruction techniques to improve cardiovascular diagnostics. As the principal investigator of the ERC Starting Grant SMHEART, his team focuses on enhancing the visualization and characterization of cardiac diseases, providing deeper insights into the heart’s anatomy, function, and tissue properties.