Entry tear hemodynamics using 4D flow MRI in a patient with acute type B aortic dissection.

Stanford type B aortic dissection (TBAD) is a potentially fatal condition involving a tear in the descending aorta. As TBAD can be managed with medical therapy or surgical repair, identifying predictors of adverse outcomes is important to risk-stratify patients for preemptive surgical procedures. 4D flow magnetic resonance imaging (MRI) has shown to be useful in characterizing the complex hemodynamics seen in TBAD patients and correlating flow patterns with adverse outcomes. We report a case of a 58-year-old man who presented to the hospital with acute TBAD and a large primary entry tear. He was initially managed with medical therapy due to his stable clinical status and computed tomographic angiography showing a stable dissection. However, 4D flow MRI showed high velocity flow through the entry tear, which foreshadowed the later clinical decompensation of the patient. Our case demonstrates that performing 4D flow MRI on TBAD patients is feasible and can provide valuable information in the decision to pursue medical or surgical management.

Association between macrophage-like cell density and ischemia metrics in diabetic eyes.

We previously showed that macrophage-like cells (MLCs) are increased in eyes with advanced diabetic retinopathy (DR). Here, we hypothesized that MLC density was correlated with ischemia using optical coherence tomography angiography (OCTA) and ultra-widefield fluorescein angiography (UWF-FA). Treatment-naïve diabetic eyes were prospectively imaged with repeated OCTA (average 5.3 scans per eye) and UWF-FA imaging. OCTA images were registered and averaged to generate a superficial capillary plexus (SCP), deep capillary plexus (DCP), and MLC slab. We calculated geometric perfusion deficit (GPD), vessel length density, and vessel density for the SCP and DCP. MLC density was quantified by two masked graders and averaged. Ischemia on UWF-FA was measured to generate a non-perfusion area (NPA) and index (NPI). Since MLC density was non-parametrically distributed, MLC density was correlated with ischemia metrics using Spearman correlations. Forty-five treatment-naïve eyes of 45 patients (59 ± 12 years of age; 56% female) were imaged. We included 6 eyes with no DR, 7 eyes with mild non-proliferative DR (NPDR), 22 moderate NPDR, 4 severe NPDR, and 6 PDR eyes. MLC density between graders was highly correlated (r = 0.9592, p < 0.0001). MLC density was correlated with DCP GPD (r = 0.296, p = 0.049), but no other OCTA ischemia metrics. MLC density was also correlated with UWF-FA NPA (r = 0.330, p = 0.035) and NPI (r = 0.332, p = 0.034). MLC density was correlated with total ischemia on UWF-FA and local DCP GPD. Since both UWF-FA and DCP non-perfusion are associated with higher risk for DR progression, MLC density could be another potential biomarker for DR progression.

Input layer regularization for magnetic resonance relaxometry biexponential parameter estimation.

Many methods have been developed for estimating the parameters of biexponential decay signals, which arise throughout magnetic resonance relaxometry (MRR) and the physical sciences. This is an intrinsically ill-posed problem so that estimates can depend strongly on noise and underlying parameter values. Regularization has proven to be a remarkably efficient procedure for providing more reliable solutions to ill-posed problems, while, more recently, neural networks have been used for parameter estimation. We re-address the problem of parameter estimation in biexponential models by introducing a novel form of neural network regularization which we call input layer regularization (ILR). Here, inputs to the neural network are composed of a biexponential decay signal augmented by signals constructed from parameters obtained from a regularized nonlinear least-squares estimate of the two decay time constants. We find that ILR results in a reduction in the error of time constant estimates on the order of 15%-50% or more, depending on the metric used and signal-to-noise level, with greater improvement seen for the time constant of the more rapidly decaying component. ILR is compatible with existing regularization techniques and should be applicable to a wide range of parameter estimation problems.