Coexistent diabetes is associated with the presence of adverse phenotypic features in patients with hypertrophic cardiomyopathy

Background

Type 2 diabetes mellitus (DM) is associated with worsened clinical outcomes in hypertrophic cardiomyopathy (HCM) patients. The reasons for this adverse prognostic association are incompletely understood. Although distinct entities both HCM and DM share common features of impaired myocardial energetics and coronary microvascular function.

Purpose

We sought to test the hypothesis that co-existent diabetes is associated with greater reductions in myocardial energetics and perfusion, and higher scar burden in HCM.

Research design and methods

Seventy-five age- and sex-matched participants with concomitant HCM and DM (HCM-DM, n=20), isolated HCM (n=20), isolated DM (n=20) and healthy volunteers (HV, n=15) underwent 31phosphorus magnetic resonance spectroscopy and cardiovascular magnetic resonance imaging. The HCM groups were matched for HCM phenotype. The DM groups were matched for diabetes treatment, duration, HbA1c, body mass index and hypertension comorbidity.

Results

ESC sudden cardiac death risk scores were comparable between the HCM groups (HCM: 2.2±1.5%, HCM-DM: 1.9±1.2%; p=NS) and sarcomeric mutations were equally common. HCM-DM had the highest NT-proBNP levels (HV: 42 ng/L [IQR: 35–66], DM: 118 ng/L [IQR: 53–187], HCM: 298 ng/L [IQR: 157–837], HCM-DM: 726 ng/L [IQR: 213–8695]; p<0.0001). Left-ventricular ejection fraction, mass and wall thickness were similar between the HCM groups. HCM-DM displayed a greater degree of fibrosis burden with higher scar percentage, and lower global longitudinal strain compared to the isolated HCM. PCr/ATP was similarly decreased in the HCM-DM and DM (HV: 2.17±0.49, DM: 1.61±0.23, HCM: 1.93±0.38, HCM-DM: 1.54±0.27; p=0.0003). HCM-DM had the lowest stress myocardial blood flow (HV: 2.06±0.42 ml/min/g, DM: 1.78±0.45 ml/min/g, HCM: 1.74±0.44 ml/min/g, HCM-DM: 1.39±0.42 ml/min/g; p=0.004).

Conclusions

We show for the first time that HCM patients with DM comorbidity display greater reductions in myocardial energetics, perfusion, contractile function and higher myocardial scar burden and serum NT-proBNP levels compared to patients with isolated HCM despite similar LV mass and wall thickness and presence of sarcomeric mutations. These adverse phenotypic features may be important components of the adverse clinical manifestation attributable to a combined presence of HCM and DM.

Funding Acknowledgement

Type of funding sources: Foundation. Main funding source(s): Diabetes UK

Female sex and systolic blood pressure are independently associated with coronary microvascular dysfunction in asymptomatic adults with type 2 diabetes

Introduction

Coronary microvascular dysfunction is frequently reported in people with type 2 diabetes (T2D), is associated with reduced exercise capacity, and is a prognostic marker. Identifying modifiable risk factors associated with microvascular dysfunction may facilitate early intervention to improve outcomes in these patients.

Purpose

To identify independent determinants of myocardial perfusion reserve (MPR) in asymptomatic adults with T2D and no prevalent cardiovascular disease.

Methods

Prospective cross-sectional study. People with and without T2D and no signs, symptoms or evidence of cardiovascular disease underwent comprehensive phenotyping with echocardiography, coronary artery calcium scoring, and multiparametric cardiac MRI including adenosine stress and rest perfusion with automated pixel-wise myocardial blood flow (MBF) mapping. Participants with regional perfusion defects indicating obstructive coronary disease or silent myocardial infarct on late-gadolinium enhancement were excluded from analysis. Univariable and multivariable linear regression was performed to identify independent determinants of MPR.

Results

Two-hundred people with T2D (diabetes duration 11±8 years) were compared with 39 sex- and ethnicity-matched non-diabetic controls (Table 1). People with T2D had higher body mass index (BMI) and ambulatory 24-hour systolic blood pressure (SBP). There was evidence of concentric left ventricular (LV) remodelling (higher LV mass/volume), extracellular matrix expansion (higher ECV fraction), and both systolic and diastolic dysfunction (lower global longitudinal systolic strain and E/A ratio, respectively) in those with T2D. Resting MBF was similar between groups, but stress MBF tended to be lower in T2D compared to controls with significantly reduced MPR in T2Ds (2.87±0.86 vs 3.18±0.82, p=0.043). In univariable analysis, MPR correlated with sex, 24-hour SBP, and E/e' ratio. In a multivariable model adjusting for clinical (age, sex, smoking status, BMI, ambulatory SBP, diabetes duration, HbA1c, low-density lipoprotein, albuminuria) and imaging variables (E/e' ratio, LV mass/volume, global longitudinal strain, myocardial ECV, coronary calcium score) known to affect coronary perfusion, female sex (β=−0.227, p=0.013) and 24-hour SBP (β=−0.275, p=0.001) were the only variables independently associated with MPR.

Conclusion

Female sex is associated with coronary microvascular dysfunction in asymptomatic people with T2D but not LV mass or myocardial extracellular volume. Systolic BP is the only modifiable independent determinant of MPR and may be an early target for intervention to prevent heart failure development in these patients.

Funding Acknowledgement

Type of funding sources: Public Institution(s). Main funding source(s): National Institute for Health Research (NIHR) United Kingdom through a Research Professorship award (RP-2017-08-ST2-007).British Heart Foundation through a Clinical Research Training Fellowship award (FS/16/47/32190).

Clinical outcomes and myocardial recovery in energetics, perfusion and contractile function after valve replacement surgery in severe aortic stenosis patients with diabetes comorbidity

Background

Aortic stenosis (AS) and type 2 diabetes mellitus (DM) are increasingly frequent comorbidities in aging populations, and diabetes is associated with increased morbidity and mortality after aortic valve replacement (AVR). Although distinct pathological entities, AS and DM share common features of impaired myocardial energetics and coronary microvascular dysfunction (CMD). The mechanisms for the adverse prognostic association between AS and DM are incompletely understood.

Purpose

Utilising 31phosphorus magnetic resonance spectroscopy (31P-MRS) and CMR, we tested the hypotheses that the collective impact of severe AS and DM on the myocardium aggravates the impairment in energetics, function and perfusion.

Methods

Eighty-eight severe AS patients with (AS-DM) and without DM (Iso-AS) undergoing AVR and 15 healthy volunteers were recruited. Patients with coronary artery disease were excluded. Participants with AS underwent 31P-MRS and comprehensive CMR imaging 1 month prior to and 6 months after AVR.

Results

Demographic, biochemical and CMR/31P-MRS data are shown in Table-1. All groups were matched for age and sex distribution, with AS groups matched for surgical scores and frailty scores. NTproBNP levels were similarly elevated in AS groups. Left ventricular (LV) volumes and ejection fraction (EF) were similar between the groups, with no significant difference in LV mass or wall thickness between the AS groups. The baseline differences in myocardial energetics, stress myocardial blood flow (MBF) and global longitudinal strain (GLS) are shown in the Figure. AS-DM patients showed greater reductions in myocardial energetics (p<0.0001), global stress MBF (p<0.0001) and more significant reductions in GLS (p=0.001) than the Iso-AS patients. At 6 month post AVR both AS groups showed significant improvements in stress MBF and GLS. However, only the Iso-AS patients showed significant improvement in myocardial energetics.

AS patients were followed up for a median of 12 months. Cumulative incidence of the clinical events post AVR (composite of cardiovascular death, stroke, heart failure admission, infective endocarditis) were significantly higher in the AS-DM group than the Iso-AS group (Hazard Ratio: 3.35; 95% CI: 0.97–11.6; p=0.02).

Conclusion

Diabetes was associated with increased morbidity and mortality after AVR. We showed for the first time that the collective impact of T2DM and AS on the myocardium aggravates energetic impairment, CMD and contractile dysfunction. While myocardial recovery following AVR was associated with similar improvements in perfusion and contractile function in severe AS patients with and without T2DM, improvements in energetics were only detected in isolated AS patients. However, despite the significant improvements in contractile function and perfusion following AVR in diabetes patients, these parameters remained lower in the group with diabetes comorbidity compared to isolated AS patients.

Funding Acknowledgement

Type of funding sources: Foundation. Main funding source(s): Wellcome Trust

Improved diagnostic accuracy for apical hypertrophic cardiomyopathy

Introduction

The diagnosis of apical hypertrophic cardiomyopathy (ApHCM) is contingent on demonstrating apical maximum wall thickness (MWT) of ≥15mm; the same threshold as other HCM subtypes. However, the myocardium naturally tapers towards the apex in healthy individuals, so ≥15mm MWT is proportionately higher in the apex than in naturally thicker basal segments. Using cardiac magnetic resonance (CMR), relative ApHCM has been described (typical ECG features, loss of apical tapering, cavity obliteration but hypertrophy <15mm). Wall thickness measurement using machine learning now exceeds human performance.

Purpose

We aimed to redefine the optimal diagnostic threshold for ApHCM using segment-specific criteria based on a large cohort of healthy control subjects.

Methods

Segmental wall thickness was measured using healthy subjects from the UK Biobank using a clinically validated machine learning algorithm1,2. A normative reference range was established for all 16 segments, conditioned to body surface area (BSA), sex and age. Derived segment-specific wall thickness thresholds were used to define optimal disease thresholds for patients clinically managed with overt (MWT ≥15mm) and relative ApHCM (MWT <15mm, but typical ECG and imaging findings).

Results

4118 UK biobank subjects were used to define normal segmental thicknesses and reference ranges. These were applied to ApHCM (73 overt, 31 relative). There were no apical wall thickness age related differences. The upper limit of the 95% confidence interval corresponded to a combined maximum apical MWT for both males and females of 10.4mm using non-indexed measurement, or 5.6mm/m2 when indexed to BSA. Non-indexed segmental threshold identified 100% of ApHCM patients (true positives), 81% (25 of 31) relative ApHCM and 3% (115 of 4118) of healthy UK biobank subjects (false positives). Indexed segmental thresholds improved the diagnostic potential in relative ApHCM without an increase in false positives (100% of ApHCM patients, 84% (26 of 31) of relative ApHCM patients, and 3% healthy UK biobank (127 of 4118).

Conclusion

We propose new diagnostic criteria for ApHCM using segmental indexed apical wall thickness of >5.6 mm/m2 to better identify inappropriate apical hypertrophy in those whose wall thickness does not meet current criteria for diagnosis.

Funding Acknowledgement

Type of funding sources: Foundation. Main funding source(s): British Heart Foundation

Body mass index associated differences in cardiac stress energetics in type 2 diabetes

Background

Patients with T2D and heart disease have normal body mass index (BMI), suggesting that diabetes and obesity mediate cardiovascular change by different mechanisms. Changes in cardiac energy metabolism in lean diabetic patients during exercise stress have not been previously reported.

Objectives

We aimed to assess if there are BMI-associated differences in cardiac stress metabolism in patients with T2D.

Methods

Twenty-five overweight T2D patients (O-T2D) and eleven lean T2D patients (LnT2D), age- and ethnicity-matched and with no other comorbidities were studied. Patients were on oral hypoglycaemics only and were free of diabetes complications. Participants underwent rest and dobutamine stress phosphorus magnetic resonance spectroscopy (31P-MRS) and cardiovascular magnetic resonance (CMR) at 3T for the assessment of myocardial phosphocreatine to ATP ratio (PCr/ATP) as a measure of myocardial energetics, biventricular volumes, rest and stress left ventricular (LV) ejection fraction, global longitudinal shortening, and mitral in-flow E/A ratio for assessment of diastolic function and perfusion.

Intravenous Dobutamine was administered at a dose of 10μg/kg/min, increasing at 90 second intervals up to a maximum of 40 μg/kg/min to achieve a target heart rate of 65% of the age-predicted maximal heart rate. Mean rate pressure product (RPP) was recorded at rest and stress. Heart rate was maintained at target for the duration of the 31P-MRS and stress CMR cine, mitral in-flow and perfusion acquisitions.

Results

The cardiac volumes, systolic or diastolic function and LV mass were similar between LnT2D and O-T2D. Although the O-T2D patients had a numerically lower rest and stress PCr/ATP ratio, this did not reach statistical significance. Resting PCr/ATP was reduced in LnT2D and O-T2D patients similarly. However, LnT2D showed a greater reduction in PCr/ATP (stress PCr/ATP LnT2D 1.51±0.2 vs O-T2D 1.41±0.25, p=0.02) despite similar increases in RPP. Stress myocardial blood flow (MBF) was also significantly lower in the O-T2D patients. There were significant correlations of BMI with LV mass (r=0.35, p=0.03); stress LVEF (r=−0.34, p=0.04); stress MBF stress (r=−0.53, p=0.001) and stress E/A (r=0.46, p=0.01) (figure 1).

Conclusions

Despite their better stress perfusion and similar glycaemic control, LnT2D show worse metabolic reserve characterised by more significant decrements in energetics in response to hemodynamic stress compared to overweight patients with T2D. Higher BMI correlates inversely with stress myocardial blood flow and with stress left ventricular ejection fraction. The presence of these subtle alterations in measures of stress metabolism and perfusion might signify a distinct metabolic phenotype of “lean diabetic cardiomyopathy”. Future studies are needed to further delineate alterations in cardiac energy metabolism in lean and overweight/obese type 2 diabetes patients, and their role in the development of cardiac dysfunction.

Funding Acknowledgement

Type of funding sources: Foundation. Main funding source(s): Wellcome TrustBHF

Gestational diabetes, preeclampsia and the maternal heart

Background

Gestational diabetes mellitus (GDM) and preeclampsia (pE) are both associated with an increased risk of cardiovascular mortality and morbidity, including an increased risk of developing heart failure in later life. Both conditions are increasing in prevalence; GDM affects up to 12% and pE affects 3–5% of pregnancies worldwide. Compromised cardiac energy production is an important contributor to most forms of heart disease. The changes in myocardial energetics in GDM and pE have not been characterised previously.

Purpose

We sought to assess if women with GDM and women with pE in the third trimester of pregnancy exhibit adverse cardiac alterations in myocardial energetics, function or tissue characteristics.

Methods

Thirty-eight healthy pregnant (HP) women, thirty women with GDM and fifteen women with pE were recruited, matched for age and ethnicity. Participants underwent phosphorus magnetic resonance spectroscopy and cardiovascular magnetic resonance for assessment of myocardial energetics (phosphocreatine to ATP ratio (PCr/ATP)), tissue characteristics, biventricular volumes and ejection fractions, left ventricular (LV) mass, global longitudinal strain (GLS) and mitral in-flow E/A ratio.

Results

The biochemical characteristics and multiparametric MR results are given in Table 1.

The women in the GDM and the pE groups had higher body-mass index. There was a stepwise increase in the systolic and diastolic BP from the HP to the GDM to the pE group. There was no difference in NTproBNP concentrations between the groups. The gestational weight gain was higher in women with GDM and pE compared to the HP group.

The women in the GDM and the pE groups showed similar reductions in myocardial PCr/ATP ratios compared to HP group (Figure 1a), accompanied by lower LV end-diastolic volumes and higher LV mass (Figure 1b) and enhanced LV concentricity in both groups (Figure 1c). While LV ejection fractions were similar across the groups, the GLS was reduced in women with GDM and in women with pE (Figure 1d).

Conclusions

We show here for the first time that despite no prior diagnosis of diabetes or hypertension, women with GDM or pE manifest impaired myocardial contractility and higher LV mass, associated with reductions in myocardial energetics. These findings may aid our understanding of the long-term cardiovascular risks associated with these conditions.

Funding Acknowledgement

Type of funding sources: Foundation. Main funding source(s): Wellcome Trust

Evaluation of cardiac involvement in patients with clinical post-COVID-19 syndrome

Introduction

The underlying pathophysiology of Post-COVID-19 syndrome remains unknown, but increased cardiometabolic demand and state of mitochondrial dysfunction have emerged as candidate mechanisms. Cardiovascular magnetic resonance (CMR) provides insight into pathophysiological mechanisms underlying cardiovascular disease and 31-phosphorus magnetic resonance spectroscopy (31P-MRS) allows non-invasive assessment of the myocardial energetic state.

Purpose

We sought to assess whether Post-COVID-19 syndrome is associated with abnormalities of myocardial structure, function, perfusion and tissue characteristics or energetic derangement.

Methods

Prospective case-control study. A total of 20 patients with a clinical diagnosis of Post-COVID-19 syndrome (seropositive) and no prior underlying cardiovascular disease (CVD) and ten matching controls underwent 31P-MRS and CMR at 3T at a single time point. (Figure 1) All patients had been symptomatic with acute COVID-19, but none required hospital admission.

Results

Between the Post-COVID-19 syndrome patients and matched contemporary controls there were no differences in myocardial energetics (phosphocreatine to ATP ratio), in cardiac structure (biventricular volumes, left ventricular mass), function (biventricular ejection fractions, global longitudinal strain), tissue characterization (T1 and extracellular volume [ECV] fraction mapping, late gadolinium enhancement) or perfusion (myocardial rest and stress blood flow, myocardial perfusion reserve). One patient with Post-COVID-19 syndrome showed subepicardial hyperenhancement on the late gadolinium enhancement imaging compatible with prior myocarditis, but no accompanying abnormality in cardiac size, function, perfusion, ECV, T1, T2 mapping or energetics. This patient was excluded from statistical analyses. (Table 1)

Conclusion

In this study, the overwhelming majority of patients with a clinical Post-COVID-19 syndrome with no prior CVD did not exhibit any abnormalities in myocardial energetics, structure, function, blood flow or tissue characteristics.

Funding Acknowledgement

Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Welcome Trust Clinical Career Development Fellowship (221690/Z/20/Z);NIHR-UKRI COVID-19 Rapid Response Rolling Call (COV0254)ESC Training Grant

Convolutional neural network transformer (CNNT) for free-breathing real-time cine imaging

Funding Acknowledgements

Type of funding sources: None.

Background

Real-time cine imaging does not require breath-holding and is a robust cine imaging technique in the presence of irregular heartbeats. It is a good alternative to the conventional breath-hold retro-gated cine for simplified acquisition and improved patient comfort. Real-time acquisition is achieved with the single-shot BSSFP readout without retro-gating. To maintain good temporal and spatial resolution, higher acceleration (e.g. >4x parallel imaging) is required. As a result, the real-time cine images experience reduced signal-to-noise ratio (SNR), which limits its clinical acceptance.

Purpose

We developed a novel deep learning model architecture, the Convolutional Neural Network Transformer (CNNT), to improve the quality of real-time cine, under 4x, 5x and 6x acceleration.

Method

Convolutional Neural Networks (CNN) are widely used in CMR research to process cardiac images. Cardiac images are often acquired as a time series with strong inter-phase correlation. We combined the CNN with the more recent transformer model to develop a novel CNNT architecture. It takes in the entire 2D+T time series as input and has advantages of CNN for efficient computation and spatial invariance. It further inherits the advantages of attention layer in the transformer and is able to efficiently utilize the temporal correlation within a time series.

A CNNT model is developed to improve the SNR of real-time cine imaging. N=10 patients were scanned at a heart center, with 4x, 5x and 6x acceleration. Typical imaging parameters are: FOV 360×270mm2, flip angle 50°, acquired matrix size 160×90 for R=4 acceleration, 192×108 for R=5 and 6, temporal resolution 40ms for R=4, 42ms for R=5 and 35ms for R=6. The real-time images went through a TGRAPPA reconstruction [1] and the CNNT model. The SNR of TGRAPPA was measured with SNR units [2]. The Monte-Carlo pseudo-replica test was used to measure SNR for the CNNT model. For every cine series, two phases were picked for the end-systole and end-diastole. For every image picked, two region-of-interests were drawn in the myocardium and in the LV blood pool. The CNNT model was deployed inline on the MR scanner using the Gadgetron InlineAI [3].

Results

Figure 1 gives real-time cine images for three accelerations, reconstructed with TGRAPPA and CNNT. The parallel imaging TGRAPPA reconstruction suffers significant SNR loss from elevated g-factor and less acquired data. The deep learning CNNT model recovered SNR even at the very high 6x acceleration, without observed loss of boundary sharpness.

Table 1 lists the SNR measurement results. The TGRAPPA SNR decreased ∼4x from R=4 to R=6 for both the blood and myocardium. For the blood, the CNNT increased the SNR by 170%, 335%, 371% at R=4, 5 and 6. For the myocardium, the SNR increases were 335%, 634% and 828%.

Conclusion

We developed a convolutional neural network transformer model to recover the SNR for real-time cine imaging at higher acceleration.

CNNT DB-LGE: free-breathing dark blood late enhancement imaging using the convolutional neural network transformer speeds acquisition by 50%

Funding Acknowledgements

Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Supported in part by the Division of Intramural Research of the National Heart, Lung, and Blood Institute, National Institutes of Health (grants Z1A-HL006214-05 and Z1A-HL006242-02).

Background

Dark blood late gadolinium enhancement (DB-LGE) imaging shows superior delineation of myocardial infarction (MI), especially at the sub-endocardial boundary. Our previous study [1] developed a free-breathing DB-LGE with the single shot SSFP readout, phase sensitive inversion recovery (PSIR) reconstruction, and respiratory motion corrected averaging. To compensate the potential signal-to-noise ratio loss, our previous DB-LGE doubled the measurements, thereby increasing the acquisition time.

Purpose

In this study, we developed a deep learning image enhancement model using a novel neural network architecture called the convolutional neural network transformer (CNNT) to improve the image quality of DB-LGE and to reduce the acquisition time by decreasing the number of measurements.

Methods

A novel image enhancement model was developed using a novel network architecture called the Convolutional Neural Network Transformer (CNNT) proposed by us. This architecture is suitable for the 2D+Time CMR acquisition, by exploiting the temporal correlation between images over multiple averages.

The evaluation was first retrospectively conducted on a cohort of 12 patients acquired with the original protocol [1] using the full 16 measurements. For every subject, a complete short-axis stack (typically 12 slices) was acquired to cover the entire left ventricular. The imaging data was reconstructed in three ways. Original: using all acquired 16 measurements. This is our base-line protocol. Original 50%: using only the first 8 measurements. CNNT 50%: using only the first 8 averages, but performing the CNNT deep learning image enhancement before MOCO PSIR reconstruction. Two experienced imaging researchers (PK and MF, >10 years of experience for both) scored all DB-LGE images for the overall quality, diagnostic confidence and delineation of MI/boundaries (5 = excellent, 4 = good, 3 = fair, 2 = poor, and 1 = non-diagnostic). The CNNT DB-LGE was deployed to the MR scanner using the Gadgetron InlineAI [2].

Results

Figure 1 gives examples of DB-LGE with three reconstruction methods. The CNNT image has higher SNR and well delineated MI. The Original images with the longest acquisition have good quality and the Original-50% acquired with 8 measurements are good quality but have reduced SNR. The mean scores for overall image quality, diagnostic confidence and MI delineation of two reviewers were 4.88±0.23, 4.88±0.23, 4.83±0.25 for CNNT and 4.96±0.14, 4.96±0.14, 4.67±0.39 for the original approach. No significant differences were found between the original and the CNNT (P>0.15 for all).

Figure 2 shows an acute MI patient prospectively acquired with the 50% scan time reduction, with and without the CNNT enhancement. The resulting PSIR images well delineate the MVO due to the acute MI, with improved SNR.

Conclusion

A novel CNNT model was proposed and evaluated to speed up the free-breathing MOCO DB LGE by 50% without sacrificing image quality.

Improving the diagnostic accuracy of apical hypertrophic cardiomyopathy using machine learning

Funding Acknowledgements

Type of funding sources: Private grant(s) and/or Sponsorship. Main funding source(s): Dr Hughes is supported by the British Heart Foundation (grant number FS/17/82/33222).

Introduction

The imaging criteria for diagnosis of apical hypertrophic cardiomyopathy (ApHCM) is a maximum wall thickness (MWT) ≥15mm. CMR enables detection of subtle phenotypic features (e.g. loss of apical tapering, cavity obliteration) and coupled with characteristic electrocardiogram changes, ApHCM can be diagnosed without overt hypertrophy. However, these patients are not captured by current diagnostic criteria.

Purpose

We sought to use a machine learning tool to quantify wall thickness and identify patients with ‘relative’ ApHCM that do not reach current diagnostic thresholds.

Methods

CMR images from 4118 healthy participants from the UK Biobank were segmented automatically with a clinically validated machine learning algorithm and wall thickness measured at all point in the myocardium by solving a solution to Laplace’s equation. MWT were pooled into 16 AHA segments and indexed to body surface area (BSA). The non-indexed and indexed segmental upper limit of normal was calculated as the mean + 3 standard deviations (the equivalent of 95% confidence interval after correcting for multiple [16] comparisons using the Bonferroni method).

Results

73 overt ApHCM subjects (MWT>15mm) and 31 relative ApHCM subjects underwent CMR scanning. In healthy controls, the non-indexed (and indexed) upper limits were calculated for the apical-anterior 10.2mm (5.2 mm/m2), apical-septal 11.1mm (5.6 mm/m2), apical-inferior 10.5mm (5.3 mm/m2) and apical-lateral 10.1mm (5.2 mm/m2) segments. With a non-indexed cut-off, all (73 of 73) overt ApHCM and 84% (26 of 31) relative ApHCM were classified as having an abnormally thick apex. 3% (127 of 4118) of the healthy UK Biobank cohort were classified as abnormal, as expected. Using an indexed cut-off, all overt ApHCM and 87% (27/31) relative ApHCM were classified as abnormal, and 3% (123 of 4118) of the healthy UK Biobank cohort were misclassified.

Conclusion

We can successfully classify 87% of relative ApHCM patients from a normative reference range derived from a large cohort of healthy patients – a significant improvement on existing methods. We show that the specificity and sensitivity is increased when MWT is indexed to BSA. For practical clinical application, we recommend a cut-off of 10mm or an indexed cut-off of 5mm/m2 in any apical segment to diagnose apical LVH. Overt and relative apical HCM examplesHealthy controls AHA maps (non-indexed)

Speeding up cardiac MR segmentation with semi-supervision: applications in cine imaging

Funding Acknowledgements

Type of funding sources: Private company. Main funding source(s): This material is based upon work supported by the Google Cloud Research Credits program with the award GCP19980904.

Background

Segmentation is an important postprocessing step in cardiac magnetic resonance (CMR) imaging that enables quantitative assessment of functional parameters. Deep learning can automate the segmentation process, producing accurate contours of cardiac structures while reducing the time required to analyze images and the interobserver variation compared to manual analysis. However, common approaches to training neural networks (NNs) require large amounts of labeled data, which is costly to generate and slows down the development of CMR segmentation NNs for new applications. Semi-supervision is an approach to alleviate this labeling burden by relying on abundant unlabeled data and a smaller amount of labeled data to train NNs.

Purpose

We propose a novel semi-supervised method to train CMR segmentation NNs. We use the proposed method to train NNs to segment the left ventricle in CMR cine images. Ultimately, we aim to show that semi-supervision can drastically reduce the amount of labeled data required to develop machine learning segmentation applications for CMR while maintaining high performance.

Methods

Our dataset consists of 1,208 short-axis cine CMR images and 1,244 long-axis cine CMR images. An expert annotator manually segmented the endocardium on the end-diastolic and end-systolic short-axis and long-axis images and the epicardium on the end-diastolic short-axis images. We split the dataset randomly by patient into 60% training, 20% validation, and 20% testing data. We train semi-supervised segmentation networks using a supervised cross-entropy loss to learn from the labeled training data and a cosine embedding loss in addition to a pseudo-labeling step to learn from the unlabeled training data. To evaluate how performance changes with different amounts of labeled training data, we vary the percent of training data that has labels from <1%-100%. We evaluate the predicted segmentation masks using the Dice coefficient.

Results

Using only 100 labeled image slices, the semi-supervised segmentation NNs achieve a mean Dice coefficient within 1.10% of networks trained with fully labeled training sets, corresponding to >85% reduction in required labeled training data (Table 1). The proposed semi-supervised method improves performance over naïve training by 6.21% for the most limited labeled data setting (i.e., 10 labeled image slices; Figure 1).

Conclusion

We have shown that NNs trained with limited labeled data achieve high performance on left ventricle segmentation in short-axis and long-axis CMR cines. The proposed approach is flexible and broadly applicable to different CMR segmentation tasks, enabling rapid development of segmentation networks for many cardiac structures and applications. Table 1Figure 1

Pharmacological inhibition of eIF2alpha phosphorylation by integrated stress response inhibitor (ISRIB) ameliorates vascular calcification in rats

Vascular calcification (VC) is an independent risk factor for cardiovascular events and all-cause mortality with the absence of current treatment. This study aimed to investigate whether eIF2alpha phosphorylation inhibition could ameliorate VC. VC in rats was induced by administration of vitamin D3 (3×10(5) IU/kg, intramuscularly) plus nicotine (25 mg/kg, intragastrically). ISRIB (0.25 mg/kg·week), an inhibitor of eIF2alpha phosphorylation, ameliorated the elevation of calcium deposition and ALP activity in calcified rat aortas, accompanied by amelioration of increased SBP, PP, and PWV. The decreased protein levels of calponin and SM22alpha, and the increased levels of RUNX2 and BMP2 in calcified aorta were all rescued by ISRIB, while the increased levels of the GRP78, GRP94, and C/EBP homologous proteins in rats with VC were also attenuated. Moreover, ISRIB could prevent the elevation of eIF2alpha phosphorylation and ATF4, and partially inhibit PERK phosphorylation in the calcified aorta. These results suggested that an eIF2alpha phosphorylation inhibitor could ameliorate VC pathogenesis by blocking eIF2alpha/ATF4 signaling, which may provide a new target for VC prevention and treatment.

[Effect of tourniquet on morphology and stiffness of quadriceps in patients receiving total knee arthroplasty: a randomized controlled trial]

Objective: To evaluate tourniquet effect on the morphology and stiffness of quadriceps in patients receiving total knee arthroplasty (TKA). Methods: Total of 80 patients with primary knee osteoarthritis receiving unilateral primary TKA from August 2020 to June 2021 in Peking University Third Hospital were enrolled in this randomized controlled trial. The patients were randomly divided into tourniquet group and non-tourniquet group (40 cases in each group). The study measured and compared postoperative thickness and stiffness of quadriceps, as well as circumference of the affected thigh in two groups. Results: There were 11 males and 29 females in tourniquet group, 9 males and 31 females in non-tourniquet group, with mean age of (69.1±5.1) years and (67.4±5.3) years, respectively. There was no significant difference in demographic information such as gender, age and BMI (all P<0.05). Postoperative thickness of quadriceps, stiffness of quadriceps and circumference［x¯±s or M （Q₁，Q₃）］of the affected thigh in all patients were (2.76±0.69) cm, 25.20 (17.83, 32.90) m/s, 54.00 (51.13, 56.00) cm. These outcomes in tourniquet and non-tourniquet group［x¯±s or M （Q₁，Q₃）］were (2.78±0.76) cm and (2.73±0.61) cm, 24.00 (17.15, 33.13) m/s and 25.20 (18.68, 32.85) m/s, 54.00 (49.75, 55.00) cm and 54.50 (52.13, 57.53) cm, respectively; no significant deference was found in these indexes between the two groups (all P>0.05). Conclusion: Tourniquet has no significant effect on morphology and stiffness of quadriceps in patients undergoing TKA.

[Status of HVPG clinical application in China in 2021]

Objective: The investigation and research on the application status of Hepatic Venous Pressure Gradient (HVPG) is very important to understand the real situation and future development of this technology in China. Methods: This study comprehensively investigated the basic situation of HVPG technology in China, including hospital distribution, hospital level, annual number of cases, catheters used, average cost, indications and existing problems. Results: According to the survey, there were 70 hospitals in China carrying out HVPG technology in 2021, distributed in 28 provinces (autonomous regions and municipalities directly under the central Government). A total of 4 398 cases of HVPG were performed in all the surveyed hospitals in 2021, of which 2 291 cases (52.1%) were tested by HVPG alone. The average cost of HVPG detection was (5 617.2±2 079.4) yuan. 96.3% of the teams completed HVPG detection with balloon method, and most of the teams used thrombectomy balloon catheter (80.3%). Conclusion: Through this investigation, the status of domestic clinical application of HVPG has been clarified, and it has been confirmed that many domestic medical institutions have mastered this technology, but it still needs to continue to promote and popularize HVPG technology in the future.

POS0549 IMBALANCED Tfr/Tfh IN PERIPHERAL BLOOD OF NEW-ONSET RHEUMATOID ARTHRITIS

Background

Rheumatoid arthritis (RA) is a kind of autoimmune disease characterized with chronic aggressive arthritis, presence of abnormal antibodies and persistent synovitis[1]. However, the pathogenesis of RA remained unclear by now. Several observations have showed that the breakdown of immune tolerance was involved in the development of RA. T follicular regulatory (Tfr) cells and T follicular helper (Tfh) cells, as a new subset of CD4+T cell, can exert an opposite effect in the regulation of humoral immunity[2]. Intensive researches have showed that the imbalance of Tfr/Tfh cell is related to the pathogenesis and development of autoimmune disease. There is still a lack of understanding of the relationship between Tfr/Tfh and RA, which needs further exploration.

Objectives

To detect the expression of Tfh and Tfr cells in thr peripheral blood of patients with new-onset RA and healthy controls, and to explore the role of Tfh and Tfr cells in the pathogenesis and development of RA.

Methods

We enrolled 26 patients with new-onset RA who hospitalized at the Second Hospital of Shanxi Medical University from the June 2021 to the November 2021. And 17 age and gender-matched healthy adults were anticipated as controls. The absolute number of Tfh and Tfr cells in peripheral blood was detected by flow cytometry. Disease activity indicators were collected including erythrocyte sedimentation rate (ESR, mm/h) and Disease Activity Score in 28 joints (DAS28). Then we compared the expression of Tfh and Tfr cells between the patients and healthy controls and conducted the correlation analysis with disease activity.

Results

There was significant decreased level of Tfr cells in the patients with new-onset RA compared with healthy controls (P<0.001) and a lower ratio of Tfr/Tfh in the patients (P<0.01). The reduced Tfr cells and Tfr/Tfh were significant negative correlation with the disease activity indicators including ESR and DAS28 (r=-0.305, P=0.033). There was no statistically significant in the absolute number of Tfh cells between patients and healthy controls, but the level of Tfh cell showed an increasing trend in new-onset RA.

Conclusion

The results we investigated here showed that new-onset RA exhibited an imbalance of Tfr/Tfh, specifically reduced Tfr cells, compared with healthy controls, which were negatively correlated with higher disease activity in RA. It was likely that the imbalance of Tfr/Tfh in peripheral blood played an important role in the development of RA, which may be a target to treat RA.

Table 1.

A summary of data of all enrolled patients with RA and healthy controls

HC(n=17)New-onset RA(n=26)P valueAge(years)51.94±13.0355.88±13.56P=0.35Sex(male/female)4/137/19P=0.81ESR(mm/h) a-54.85±32.71-DAS28 a-5.09±1.56-Tfh cell count(cell/UL)b43.156(23.277,106.638)83.914(38.133,119.662)0.214Tfr cell count(cell/UL)b1.422(0.882,1.893)0.441(0.116,2.888)0.025*Tfr/Tfhb0.030(0.014,0.049)0.011(0.001,0.024)0.001**

a Results are expressed as the mean ± standard error. b Results are expressed as the median(Q1,Q3).

Normally distributed continuous variables were analyzed by the independent-samples Student’s t-test. And nonparametric variables were analyzed by Mann–Whitney U test

Figure 1.

The differences of Tfr and Tfh cells in peripheral blood between the healthy controls and patients with RA. Tfr cells were higher in new-onset RA leading to an imbalance of Tfr/Tfh. Statistical analyses were performed by the Mann-Whitney U test. (*P<0.05, **P<0.01)

Figure 2.

The correlation of disease activity with the level of Tfr cells and Tfr/Tfh. Tfr cells and Tfr/Tfh were negative associated with ESR and DAS28. Statistical analyses were performed by the Spearman correlation analysis.

References

[1]Sparks, J.A. Rheumatoid Arthritis [J]. Ann Intern Med, 2019, 170(1).DOI: 10.7326/AITC201901010.

[2]Deng, J., Y. Wei, V.R. Fonseca, L. Graca, and D. Yu. T follicular helper cells and T follicular regulatory cells in rheumatic diseases [J]. Nat Rev Rheumatol, 2019, 15(8): 475-490.DOI: 10.1038/s41584-019-0254-2.

Disclosure of Interests

None declared.

[The influence of subjective comfort of working environment on occupational stress of railway station workers]

Objective: To explore the influential factors of job stress suffered by workers in railway stations, the level of job stress of were measured and subjective comfort of employees targeting to working environment were reported. Methods: In March 2019, a cluster sampling study was designed to collect the personal characteristics, job characteristics and subjective comfort degree of working environment of 432 employees in Chongqing railway stations. Meanwhile, job stress was assessed using the effort-reward imbalance scale. Chi-square test was used to compare the difference of occupational stress detection rate among different stratified factors such as occupational characteristics. Logistic regression was applied to analyze the influential factors of occupational stress. Results: The detection rate of job stress of workers in the railway stations was 31.02% (134/432) . The detection rate of job stress was higher among the divorced workers in railway stations, those earning less than 5, 000 yuan per month, those with 10-20 years' length of service, those who worked as a conductor and other workers including baggageman, station master on duty and assistant engineer (χ(2)=9.61, 14.76, 23.28, 11.06, P=0.008, 0.002, 0.000, 0.011) . The detection rate of job stress was higher among those whose working environment subjective feelings were uncomfortable, the differences were statistically significant (P<0.001) . The results showed that the occupational stress of the staff in the railway stations was influenced by their subjective feeling of air quality, noise and Space Layout (P<0.05) . The risk factors of occupational stress were air quality, noise and uncomfortable space layout (OR=0.571, 0.068, 0.441, P=0.051, 0.054, 0.007) . Conductor, other (Bellboy, Duty Station Master, assistant engineer) were the risk factors of occupational stress (OR=1.884, 2.703, P=0.065, 0.019) . The employees of station A and station B were the risk factors of occupational stress (OR=4.681, 1.811, P=0.002, 0.067) . Conclusion: The higher detection rate of job stress of workers in the railway stations is correlated with the subjective comfort degree of the working environment of the workers.

LncRNA TTN-AS1 confers tamoxifen resistance in breast cancer via sponging miR-107 to modulate PI3K/AKT signaling pathway

OBJECTIVE

Tamoxifen resistance of breast cancer (BC) is a significant hindrance in clinical therapy. The long-noncoding RNA (lncRNA) TTN-AS1 has been reported as a crucial tumor promoting factor in various cancers. In this study, we set out to discover the specific pathologic regulatory mechanisms of tamoxifen-resistance in breast cancer.

METHODS

MTT assay was conducted to evaluate the cell viability of the breast cancer cell lines MCF-7 and MCF-7/TAM. QRT-PCR and western blot assay were performed to estimate the expression of TTN-AS1, miR-107 and related proteins. Flow cytometry was conducted to identify degree of apoptosis and cell cycle. The invasive ability was estimated by transwell chamber assay.

RESULTS

Our findings revealed that TTN-AS1 can enhance tamoxifen-resistance in BC cells and augment the invasive ability of tamoxifen-resistant breast cancer cells by down-regulating miR-107, and thereby encourage the development of drug-resistant BC. Further investigation indicates that lncRNA TTN-AS1 worsens the course of tamoxifen-resistant BC by regulating zinc and ring finger 2 (ZNRF2) via miR-107 and activating the PI3K/AKT pathway.

CONCLUSION

Our findings suggest that the lncRNA TTN-AS1 can encourage tamoxifen-resistance in BC by modulating the miR-107/ZNRF2 axis and stimulating the PI3K/AKT pathway.

Up-regulation of long noncoding RNA MBNL1-AS1 suppresses breast cancer progression by modulating miR-423-5p/CREBZF axis

Breast cancer is the leading cause of cancer-related death among females, which is required to be solved urgently. Recent studies have found significant changes in a large number of genes and their transcriptional levels during breast cancer development, which are often closely related to the abnormal expression of long noncoding RNAs (lncRNAs). Herein, our study found that MBNL1-AS1 was down-regulated both in breast cancer tissues and cell lines, and it functioned as a tumor suppressor to inhibit cancer cell proliferation, migration, and invasion. MiR-423-5p was found to be a target of MBNL1-AS1 with an inverse relationship: an increase in miR-423-5p could counteract the inhibitory effect induced by MBNL1-AS1 on cancer cell promotion. Further, CREBZF was negatively regulated by miR-423-5p. Accordingly, CREBZF knockdown could impair the hindrance of cancer cell growth mediated by low miR-423-5p expression. Also, MBNL1-AS1 influenced the PI3K/AKT pathway, which was associated with cell proliferation and apoptosis, by regulating CREBZF. As a result, our work illustrated the tumor suppressor role of MBNL1-AS1 in breast cancer via upregulating miR-423-5p-targeted CREBZF. Thereby, the evidence indicates the complete understanding of the role of MBNL1-AS1/miR-423-5p/CREBZF axis in the regulation of breast cancer development, which could be used as a biomarker for predicating survival among breast cancer patients.

A Transfer Learning Radiomics Nomogram for Preoperative Prediction of Borrmann Type IV Gastric Cancer From Primary Gastric Lymphoma

Objective

This study aims to differentiate preoperative Borrmann type IV gastric cancer (GC) from primary gastric lymphoma (PGL) by transfer learning radiomics nomogram (TLRN) with whole slide images of GC as source domain data.

Materials and Methods

This study retrospectively enrolled 438 patients with histopathologic diagnoses of Borrmann type IV GC and PGL. They received CT examinations from three hospitals. Quantitative transfer learning features were extracted by the proposed transfer learning radiopathomic network and used to construct transfer learning radiomics signatures (TLRS). A TLRN, which integrates TLRS, clinical factors, and CT subjective findings, was developed by multivariate logistic regression. The diagnostic TLRN performance was assessed by clinical usefulness in the independent validation set.

Results

The TLRN was built by TLRS and a high enhanced serosa sign, which showed good agreement by the calibration curve. The TLRN performance was superior to the clinical model and TLRS. Its areas under the curve (AUC) were 0.958 (95% confidence interval [CI], 0.883–0.991), 0.867 (95% CI, 0.794–0.922), and 0.921 (95% CI, 0.860–0.960) in the internal and two external validation cohorts, respectively. Decision curve analysis (DCA) showed that the TLRN was better than any other model. TLRN has potential generalization ability, as shown in the stratification analysis.

Conclusions

The proposed TLRN based on gastric WSIs may help preoperatively differentiate PGL from Borrmann type IV GC.

Borrmann type IV gastric cancer, primary gastric lymphoma, transfer learning, whole slide image, deep learning.

Coronary microvascular dysfunction is only detectable in type 2 diabetes in the presence of obesity

Background

Heart failure (HF) is a leading cardiovascular complication of type 2 diabetes (T2D). Coronary microvascular dysfunction (CMD) precedes HF in diabetes and carries important prognostic information. CMD is also evident in metabolically healthy obese individuals without diabetes or hypertension. Whether diabetes causes CMD in the absence of obesity is uncertain. The interrelation among visceral adiposity and CMD has not been assessed previously.

Objectives

We sought to better understand the links between visceral and epicardial adipose tissue (VAT and EAT respectively) distribution, insulin resistance with myocardial perfusion, energetics and function in asymptomatic lean (LnT2D) and overweight/obese T2D patients (ObT2D) without cardiovascular disease.

Methods

62 participants [27 Ob-T2D, 15 Ln-T2D, and 20 overweight controls] were recruited. Subjects underwent cardiac and abdominal magnetic resonance imaging and 31P-magnetic resonance spectroscopy, for measurements of EAT and VAT areas, rest and adenosine stress myocardial blood flow (MBF), cardiac function and phosphocreatine to ATP ratio (PCr/ATP). Fasting blood samples were taken for plasma homeostasis model assessment of insulin resistance (HOMA-IR) index calculations.

Results

The biochemical characteristics and multiparametric MR results are given in Table 1 and results of Pearson's regression analysis in the entire study population are given in Table 2.

Stress MBF was lowest in ObT2D, while rest MBF was highest in LnT2D. Left ventricular ejection fraction (LVEF) and myocardial PCr/ATP were similarly reduced in diabetes groups. In the absence of obesity, there was no significant increase in VAT, EAT or HOMA-IR in T2D patients compared to controls. BMI and VAT, negatively correlated with LVEF, and strain parameters. PCr/ATP correlated with LVEF, but not HOMA-IR. BMI, EAT and VAT all correlated significantly with HOMA-IR, and HOMA-IR correlated with cardiac functional parameters. There was no association between HOMA-IR and myocardial perfusion.

Conclusions

In this study CMD was only evident in ObT2D patients, with normal rest and stress MBF in LnT2D patients. Despite normal perfusion and no significant increase in insulin resistance, LVEF and myocardial PCr/ATP were similarly reduced in LnT2D and ObT2D, and PCr/ATP correlated with LVEF. This suggests that alterations in cardiac energy metabolism are mechanistically more relevant for the pathophysiology of diabetic cardiomyopathy in LnT2D patients. In the absence of correlation between insulin resistance and myocardial perfusion, factors like inflammation and altered adipokine profile may play important roles for the pathophysiology of CMD in ObT2D patients. A better understanding of the underlying pathophysiological mechanisms of diabetic cardiomyopathy in LnT2D and ObT2D may help to develop contemporary tailored treatment and prevention strategies to tackle excess heart failure risk.

Funding Acknowledgement

Type of funding sources: Foundation. Main funding source(s): BHFWellcome trust Table 1Table 2

Mechanistic insights from a multiparametric magnetic resonance imaging study regarding the role of sodium glucose co-transporter 2 inhibitors

Background

Type 2 diabetes (T2D) is associated with an increased risk of heart failure (HF) and cardiovascular (CV) mortality. Sodium–glucose-co transporter-2 (SGLT2) inhibitors reduce the risk of major adverse CV events and hospitalisation for HF in T2D patients with high cardiovascular risk, despite only a modest improvement in glycemic control. Restoring cellular energy homeostasis and reversing adverse cardiac remodelling in diabetes have been speculated as a potential metabolic modulatory effect of SGLT2 inhibitors leading to their beneficial CV outcomes. Myocardial energy deficient states can be detected non-invasively by 31-phosphorus magnetic resonance spectroscopy (31P-MRS).

Objectives

Utilising cardiovascular magnetic resonance imaging (CMR) and 31P-MRS in a single centre longitudinal cohort study, we aimed to investigate the effects of the selective SGLT2 inhibitor empagliflozin on myocardial energetics, function, perfusion, and myocardial cellular volume in patients with T2D.

Methods

Eighteen consecutive T2D patients who were commenced on empagliflozin in cardiometabolic optimisation clinics underwent CMR and 31P-MRS scans before and after twelve-week empagliflozin treatment, and plasma N-terminal pro hormone B-type natriuretic peptide (NT-proBNP) levels were measured. Ten controls with no diabetes underwent an identical 31P-MRS and CMR protocol on a single visit.

Results

When compared to controls, patients with T2D showed: lower myocardial energetics (1.52±0.40 vs 2.20±0.5, p=0.0005), lower stress myocardial blood flow (1.60±0.50 vs 2.10±0.50, p=0.02) and lower left ventricular ejection fraction (52±13% vs 63±4%, p=0.01). Treatment with empagliflozin led to significant improvements in myocardial energetics (PCr/ATP: 1.52 to 1.76, p=0.009). This was accompanied by a relative 13% improvement in left ventricular ejection fraction (p=0.001), 3% improvement in global longitudinal strain (p=0.01), 61% reduction in NTproBNP (p=0.05), and 9% reduction in myocardial cell volume (p=0.04). No significant change in myocardial blood flow or diastolic strain was detected.

Conclusions

For the first time, we demonstrate that empagliflizon improves myocardial energetics and function, reduces myocardial cellular volume, and reduces NT-proBNP levels in patients with T2D.

Funding Acknowledgement

Type of funding sources: Foundation. Main funding source(s): British Heart Foundation PCr/ATPLVEF

Computed Tomography-Based Radiomics Nomogram: Potential to Predict Local Recurrence of Gastric Cancer After Radical Resection

Objective

Accurate prediction of postoperative recurrence risk of gastric cancer (GC) is critical for individualized precision therapy. We aimed to investigate whether a computed tomography (CT)-based radiomics nomogram can be used as a tool for predicting the local recurrence (LR) of GC after radical resection.

Materials and Methods

342 patients (194 in the training cohort, 78 in the internal validation cohort, and 70 in the external validation cohort) with pathologically proven GC from two centers were included. Radiomics features were extracted from the preoperative CT imaging. The clinical model, radiomics signature, and radiomics nomogram, which incorporated the radiomics signature and independent clinical risk factors, were developed and verified. Furthermore, the performance of these three models was assessed by using the area under the curve (AUC) of receiver operating characteristic (ROC) curve analysis and decision curve analysis (DCA).

Results

The radiomics signature, which was comprised of two selected radiomics features, namely, contrast_GLCM and dissimilarity_GLCM, showed better performance than the clinical model in predicting the LR of GC, with AUC values of 0.83 in the training cohort, 0.84 in the internal validation cohort, and 0.73 in the external cohort, respectively. By integrating the independent clinical risk factors (N stage, bile acid duodenogastric reflux and nodular or irregular outer layer of the gastric wall) into the radiomics signature, the radiomics nomogram achieved the highest accuracy in predicting LR, with AUC values of 0.89, 0.89 and 0.80 in the three cohorts, respectively. DCA in the validation cohort showed that radiomics nomogram added more net benefit than the clinical model within the range of 0.01-0.98.

Conclusion

The CT-based radiomics nomogram has the potential to predict the LR of GC after radical resection.