-
1.
A deep learning system for detecting diabetic retinopathy across the disease spectrum.
Dai, L, Wu, L, Li, H, Cai, C, Wu, Q, Kong, H, Liu, R, Wang, X, Hou, X, Liu, Y, et al
Nature communications. 2021;(1):3242
Abstract
Retinal screening contributes to early detection of diabetic retinopathy and timely treatment. To facilitate the screening process, we develop a deep learning system, named DeepDR, that can detect early-to-late stages of diabetic retinopathy. DeepDR is trained for real-time image quality assessment, lesion detection and grading using 466,247 fundus images from 121,342 patients with diabetes. Evaluation is performed on a local dataset with 200,136 fundus images from 52,004 patients and three external datasets with a total of 209,322 images. The area under the receiver operating characteristic curves for detecting microaneurysms, cotton-wool spots, hard exudates and hemorrhages are 0.901, 0.941, 0.954 and 0.967, respectively. The grading of diabetic retinopathy as mild, moderate, severe and proliferative achieves area under the curves of 0.943, 0.955, 0.960 and 0.972, respectively. In external validations, the area under the curves for grading range from 0.916 to 0.970, which further supports the system is efficient for diabetic retinopathy grading.
-
2.
Quantitative CEST imaging of amide proton transfer in acute ischaemic stroke.
Msayib, Y, Harston, GWJ, Tee, YK, Sheerin, F, Blockley, NP, Okell, TW, Jezzard, P, Kennedy, J, Chappell, MA
NeuroImage. Clinical. 2019;:101833
Abstract
BACKGROUND Amide proton transfer (APT) imaging may help identify the ischaemic penumbra in stroke patients, the classical definition of which is a region of tissue around the ischaemic core that is hypoperfused and metabolically stressed. Given the potential of APT imaging to complement existing imaging techniques to provide clinically-relevant information, there is a need to develop analysis techniques that deliver a robust and repeatable APT metric. The challenge to accurate quantification of an APT metric has been the heterogeneous in-vivo environment of human tissue, which exhibits several confounding magnetisation transfer effects including spectrally-asymmetric nuclear Overhauser effects (NOEs). The recent literature has introduced various model-free and model-based approaches to analysis that seek to overcome these limitations. OBJECTIVES The objective of this work was to compare quantification techniques for CEST imaging that specifically separate APT and NOE effects for application in the clinical setting. Towards this end a methodological comparison of different CEST quantification techniques was undertaken in healthy subjects, and around clinical endpoints in a cohort of acute stroke patients. METHODS MRI data from 12 patients presenting with ischaemic stroke were retrospectively analysed. Six APT quantification techniques, comprising model-based and model-free techniques, were compared for repeatability and ability for APT to distinguish pathological tissue in acute stroke. RESULTS Robustness analysis of six quantification techniques indicated that the multi-pool model-based technique had the smallest contrast between grey and white matter (2%), whereas model-free techniques exhibited the highest contrast (>30%). Model-based techniques also exhibited the lowest spatial variability, of which 4-pool APTR∗ was by far the most uniform (10% coefficient of variation, CoV), followed by 3-pool analysis (20%). Four-pool analysis yielded the highest ischaemic core contrast-to-noise ratio (0.74). Four-pool modelling of APT effects was more repeatable (3.2% CoV) than 3-pool modelling (4.6% CoV), but this appears to come at the cost of reduced contrast between infarct growth tissue and normal tissue. CONCLUSION The multi-pool measures performed best across the analyses of repeatability, spatial variability, contrast-to-noise ratio, and grey matter-white matter contrast, and might therefore be more suitable for use in clinical imaging of acute stroke. Addition of a fourth pool that separates NOEs and semisolid effects appeared to be more biophysically accurate and provided better separation of the APT signal compared to the 3-pool equivalent, but this improvement appeared be accompanied by reduced contrast between infarct growth tissue and normal tissue.
-
3.
Delineation of Human Carotid Plaque Features In Vivo by Exploiting Displacement Variance.
Torres, G, Czernuszewicz, TJ, Homeister, JW, Caughey, MC, Huang, BY, Lee, ER, Zamora, CA, Farber, MA, Marston, WA, Huang, DY, et al
IEEE transactions on ultrasonics, ferroelectrics, and frequency control. 2019;(3):481-492
-
-
Free full text
-
Abstract
While in vivo acoustic radiation force impulse (ARFI)-induced peak displacement (PD) has been demonstrated to have high sensitivity and specificity for differentiating soft from stiff plaque components in patients with carotid plaque, the parameter exhibits poorer performance for distinguishing between plaque features with similar stiffness. To improve discrimination of carotid plaque features relative to PD, we hypothesize that signal correlation and signal-to-noise ratio (SNR) can be combined, outright or via displacement variance. Plaque feature detection by displacement variance, evaluated as the decadic logarithm of the variance of acceleration and termed "log(VoA)," was compared to that achieved by exploiting SNR, cross correlation coefficient, and ARFI-induced PD outcome metrics. Parametric images were rendered for 25 patients undergoing carotid endarterectomy, with spatially matched histology confirming plaque composition and structure. On average, across all plaques, log(VoA) was the only outcome metric with values that statistically differed between regions of lipid-rich necrotic core (LRNC), intraplaque hemorrhage (IPH), collagen (COL), and calcium (CAL). Further, log(VoA) achieved the highest contrast-to-noise ratio (CNR) for discriminating between LRNC and IPH, COL and CAL, and grouped soft (LRNC and IPH) and stiff (COL and CAL) plaque components. More specifically, relative to the previously demonstrated ARFI PD parameter, log(VoA) achieved 73% higher CNR between LRNC and IPH and 59% higher CNR between COL and CAL. These results suggest that log(VoA) enhances the differentiation of LRNC, IPH, COL, and CAL in human carotid plaques, in vivo, which is clinically relevant to improving stroke risk prediction and medical management.
-
4.
Subclinical left ventricular systolic dysfunction in diabetic patients and its association with retinopathy: A 2D speckle tracking echocardiography study.
Karagöz, A, Bezgin, T, Kutlutürk, I, Külahçıoğlu, S, Tanboğa, IH, Güler, A, Karabay, CY, Oduncu, V, Aksoy, H, Kırma, C
Herz. 2015;:240-6
Abstract
AIM: It is important to diagnose diabetic cardiomyopathy in the early stages to prevent development of evident heart failure in the future. The primary objective of this study was to evaluate the presence of subclinical left ventricular (LV) dysfunction with two-dimensional (2D) speckle tracking echocardiography (STE) and the secondary objective was to compare retinopathy-positive and -negative diabetic patients. PATIENTS AND METHODS A total of 82 patients with type II diabetes mellitus (DM) and 90 gender-matched healthy controls were included. Retinopathy was present in 55 patients in the study group. All study participants underwent conventional 2D echocardiography and STE. RESULTS All diabetic patients had preserved LV ejection fraction (LV-EF ≥ 50). Compared with the control group, diabetic patients had a larger left atrium (47.3 ± 19.9 mm vs. 36.9 ± 17.8 mm, p < 0.001) and a higher E/Em ratio (12.0 ± 2.9 vs. 10.5 ± 3.7, p = 0.004). The LV-EF, LV end diastolic and end systolic volumes, E/A ratios, deceleration times, and tissue Doppler parameters were compared between groups. The study group was observed to have statistically significant lower four-chamber (4C; 17.7 ± 3.0 % vs. 19.3 ± 3.5 %, p = 0.002), three-chamber (3C; 17.5 ± 3.0 % vs. 19.2 ± 3.4 %, p = 0.001), and two-chamber (2C; 18.5 ± 3.5 % vs. 20.1 ± 2.4 %, p = 0.001) peak longitudinal strain values compared with the control group. Moreover, LV global strain values were found to be significantly lower in the DM group than in the control group (17.9 ± 2.7 % vs. 21.1 ± 3.2 %, p < 0.001). By contrast, basal rotation (4.9 ± 3.3° vs.2.8 ± 4.5°, p = 0.001), apical rotation (15.3 ± 6.7° vs. 12.1 ± 5.3°, p = 0.001) and LV twist (20.2 ± 7.2° vs. 16.9 ± 6.5°, p = 0.002) in the DM group were significantly increased compared with those of controls. CONCLUSION The STE procedure can be a useful novel technique in the determination of subclinical LV dysfunction in diabetic patients. Diabetic patients have lower longitudinal myocardial mechanics, and circumferential and rotational mechanics are impaired. There was no significant association between diabetic retinopathy and LV function.
-
5.
Intra- and inter-examination repeatability of magnetic resonance spectroscopy, magnitude-based MRI, and complex-based MRI for estimation of hepatic proton density fat fraction in overweight and obese children and adults.
Tyagi, A, Yeganeh, O, Levin, Y, Hooker, JC, Hamilton, GC, Wolfson, T, Gamst, A, Zand, AK, Heba, E, Loomba, R, et al
Abdominal imaging. 2015;(8):3070-7
-
-
Free full text
-
Abstract
PURPOSE Determine intra- and inter-examination repeatability of magnitude-based magnetic resonance imaging (MRI-M), complex-based magnetic resonance imaging (MRI-C), and magnetic resonance spectroscopy (MRS) at 3T for estimating hepatic proton density fat fraction (PDFF), and using MRS as a reference, confirm MRI-M and MRI-C accuracy. METHODS Twenty-nine overweight and obese pediatric (n = 20) and adult (n = 9) subjects (23 male, 6 female) underwent three same-day 3T MR examinations. In each examination MRI-M, MRI-C, and single-voxel MRS were acquired three times. For each MRI acquisition, hepatic PDFF was estimated at the MRS voxel location. Intra- and inter-examination repeatability were assessed by computing standard deviations (SDs) and intra-class correlation coefficients (ICCs). Aggregate SD was computed for each method as the square root of the average of first repeat variances. MRI-M and MRI-C PDFF estimation accuracy was assessed using linear regression with MRS as a reference. RESULTS For MRI-M, MRI-C, and MRS acquisitions, respectively, mean intra-examination SDs were 0.25%, 0.42%, and 0.49%; mean intra-examination ICCs were 0.999, 0.997, and 0.995; mean inter-examination SDs were 0.42%, 0.45%, and 0.46%; and inter-examination ICCs were 0.995, 0.992, and 0.990. Aggregate SD for each method was <0.9%. Using MRS as a reference, regression slope, intercept, average bias, and R (2), respectively, for MRI-M were 0.99%, 1.73%, 1.61%, and 0.986, and for MRI-C were 0.96%, 0.43%, 0.40%, and 0.991. CONCLUSION MRI-M, MRI-C, and MRS showed high intra- and inter-examination hepatic PDFF estimation repeatability in overweight and obese subjects. Longitudinal hepatic PDFF change >1.8% (twice the maximum aggregate SD) may represent real change rather than measurement imprecision. Further research is needed to assess whether examinations performed on different days or with different MR technologists affect repeatability of MRS voxel placement and MRS-based PDFF measurements.
-
6.
Enhanced axonal metabolism during early natalizumab treatment in relapsing-remitting multiple sclerosis.
Wiebenga, OT, Klauser, AM, Schoonheim, MM, Nagtegaal, GJ, Steenwijk, MD, van Rossum, JA, Polman, CH, Barkhof, F, Pouwels, PJ, Geurts, JJ
AJNR. American journal of neuroradiology. 2015;(6):1116-23
Abstract
BACKGROUND AND PURPOSE The considerable clinical effect of natalizumab in patients with relapsing-remitting multiple sclerosis might be explained by its possible beneficial effect on axonal functioning. In this longitudinal study, the effect of natalizumab on absolute concentrations of total N-acetylaspartate, a marker for neuronal integrity, and other brain metabolites is investigated in patients with relapsing-remitting multiple sclerosis by using MR spectroscopic imaging. MATERIALS AND METHODS In this explorative observational study, 25 patients with relapsing-remitting multiple sclerosis initiating natalizumab treatment were included and scanned every 6 months for 18 months. Additionally 18 matched patients with relapsing-remitting multiple sclerosis continuing treatment with interferon-β or glatiramer acetate were included along with 12 healthy controls. Imaging included short TE 2D-MR spectroscopic imaging with absolute metabolite quantification of total N-acetylaspartate, creatine and phosphocreatine, choline-containing compounds, myo-inositol, and glutamate. Concentrations were determined for lesional white matter, normal-appearing white matter, and gray matter. RESULTS At baseline in both patient groups, lower concentrations of total N-acetylaspartate and creatine and phosphocreatine were found in lesional white matter compared with normal-appearing white matter and additionally lower glutamate in lesional white matter of patients receiving natalizumab. In those patients, a significant yearly metabolite increase was found for lesional white matter total N-acetylaspartate (7%, P < .001), creatine and phosphocreatine (6%, P = .042), and glutamate (10%, P = .028), while lesion volumes did not change. In patients receiving interferon-β/glatiramer acetate, no significant change was measured in lesional white matter for any metabolite, while whole-brain normalized lesion volumes increased. CONCLUSIONS Patients treated with natalizumab showed an increase in total N-acetylaspartate, creatine and phosphocreatine, and glutamate in lesional white matter. These increasing metabolite concentrations might be a sign of enhanced axonal metabolism.
-
7.
Interexamination repeatability and spatial heterogeneity of liver iron and fat quantification using MRI-based multistep adaptive fitting algorithm.
Sofue, K, Mileto, A, Dale, BM, Zhong, X, Bashir, MR
Journal of magnetic resonance imaging : JMRI. 2015;(5):1281-90
Abstract
PURPOSE To assess the interexamination repeatability and spatial heterogeneity of liver iron and fat measurements using a magnetic resonance imaging (MRI)-based multistep adaptive fitting algorithm. MATERIALS AND METHODS This prospective observational study was Institutional Review Board-approved and Health Insurance Portability and Accountability Act-compliant. Written informed consent was waived. In all, 150 subjects were imaged on 3T MRI systems. A whole-liver volume acquisition was performed twice using a six-echo 3D spoiled gradient echo sequence during two immediately adjacent examinations. Colocalized regions of interest (ROIs) in three different hepatic segments were placed for R2 * and proton density fat fraction (PDFF) measurements by two readers independently. Mean R2 * and PDFF values between readers and acquisitions were compared using the Wilcoxon signed-rank test, intraclass correlation coefficients (ICCs), linear regression, Bland-Altman analysis, and analysis of variance (ANOVA). RESULTS The mean R2 * and PDFF values across all ROIs and measurements were 51.2 ± 25.2 s(-1) and 6.9 ± 6.4%, respectively. Mean R2 * and PDFF values showed no significant differences between the two acquisitions (P = 0.05-0.87). Between the two acquisitions, R2 * and PDFF values demonstrated almost perfect agreement (ICCs = 0.979-0.994) and excellent correlation (R(2) = 0.958-0.989). Bland-Altman analysis also demonstrated excellent agreement. In the ANOVA, the individual patient and ROI location were significant effects for both R2 * and PDFF values (P < 0.05). CONCLUSION MRI-based R2 * and PDFF measurements are repeatable between examinations. Between-measurement changes in R2 * of more than 10.1 s(-1) and in PDFF of more than 1.7% are likely due to actual tissue changes. Liver iron and fat content are variable between hepatic segments.