Varun Chowdhary

Background: Intracranial hemorrhage (ICH) requires emergent medical treatment for positive outcomes. While previous artificial intelligence (AI) solutions achieved rapid diagnostics, none were shown to improve the performance of radiologists in detecting ICHs. Here, we show that the Caire ICH artificial intelligence system enhances a radiologist's ICH diagnosis performance. Methods: A dataset of non-contrast-enhanced axial cranial computed tomography (CT) scans (n=532) were labeled for the presence or absence of an ICH. If an ICH was detected, its ICH subtype was identified. After a washout period, the three radiologists reviewed the same dataset with the assistance of the Caire ICH system. Performance was measured with respect to reader agreement, accuracy, sensitivity, and specificity when compared to the ground truth, defined as reader consensus. Results: Caire ICH improved the inter-reader agreement on average by 5.76% in a dataset with an ICH prevalence of 74.3%. Further, radiologists using Caire ICH detected an average of 18 more ICHs and significantly increased their accuracy by 6.15%, their sensitivity by 4.6%, and their specificity by 10.62%. The Caire ICH system also improved the radiologist’s ability to accurately identify the ICH subtypes present. Conclusion: The Caire ICH device significantly improves the performance of a cohort of radiologists. Such a device has the potential to be a tool that can improve patient outcomes and reduce misdiagnosis of ICH.

Subcapsular hepatic steatosis is a rare atypical pattern of fatty deposition of the liver reported in patients with diabetic nephropathy receiving peritoneal dialysis with intraperitoneal insulin. To date, there has been only one pediatric and zero adult cases of subcapsular hepatic steatosis with no history of continuous ambulatory peritoneal dialysis. We report the first published case of subcapsular hepatic steatosis in an adult diabetic patient without any history of peritoneal dialysis or evidence of chronic renal disease. A 46-year-old Caucasian woman with type 2 diabetes mellitus without renal disease presented to our emergency department with vague abdominal symptoms and vomiting. Her blood glucose levels were poorly controlled with a range of 400 to 500 mg/dL. She was diagnosed as having subcapsular hepatic steatosis based on magnetic resonance imaging. Of note, after improved glucose control her subcapsular hepatic steatosis had nearly resolved. Subcapsular hepatic steatosis has been exclusively described in patients with continuous ambulatory peritoneal dialysis and those on intraperitoneal insulin, except for one pediatric case, which was probably due to incorrect insulin administration. Our case demonstrates that a diagnosis of subcapsular hepatic diagnosis should not be restricted to those getting continuous ambulatory peritoneal dialysis, but rather expanded to all patients with uncontrolled blood glucose levels.

The aim of the present study was to test the hypothesis that heart deformation analysis (HDA) is able to discriminate regional myocardial motion patterns on the left ventricle (LV). Totally 21 healthy volunteers (15 men and 6 women) without documented cardiovascular diseases were recruited. Cine MRI was performed on those subjects at four-chamber, two-chamber, and short-axis views. The variations of segmental myocardial motion indices of the LV, which were measured with the HDA tool, were investigated. Regional displacement, velocity, strain and strain rate were compared between lateral wall and septal wall using t tests. There are significant variations (CoV = 18.0–72.4%) of myocardial motion indices (average over 21 subjects) among 16 myocardial segments. There are significant differences (p < 0.05) between displacement, velocity, strain and strain rate measured at lateral and septal areas of the LV. In conclusion, HDA is able to present different regional LV motion patterns from multiple aspects in healthy volunteers.

Background Asymmetrical caval to pulmonary blood flow is suspected to cause complications in patients with Fontan circulation. The aim of this study was to test the feasibility of 4-D flow MRI for characterizing the relationship between 3-D blood flow distribution and vascular geometry. Objective We hypothesized that both flow distribution and geometry can be calculated with low interobserver variability and will detect a direct relationship between flow distribution and Fontan geometry. Materials and methods Four-dimensional flow MRI was acquired in 10 Fontan patients (age: 16 ± 4 years [mean ± standard deviation], range: 9–21 years). The Fontan connection was isolated by 3-D segmentation to evaluate flow distribution from the inferior vena cava (IVC) and superior vena cava (SVC) to the left and right pulmonary arteries (LPA, RPA) and to characterize geometry (cross-sectional area, caval offset, vessel angle). Results Flow distribution results indicated SVC flow tended toward the RPA while IVC flow was more evenly distributed (SVC to RPA: 78% ± 28 [9–100], IVC to LPA: 54% ± 28 [4–98]). There was a significant relationship between pulmonary artery cross-sectional area and flow distribution (IVC to RPA: R²=0.50, P=0.02; SVC to LPA: R²=0.81, P=0.0004). Good agreement was found between observers and for flow distribution when compared to net flow values. Conclusion Four-dimensional flow MRI was able to detect relationships between flow distribution and vessel geometry. Future studies are warranted to investigate the potential of patient specific hemodynamic analysis to improve diagnostic capability.

Purpose To test the feasibility and effectiveness of using maximum intensity plots (MIPs) based on 4D flow magnetic resonance imaging (MRI) velocity data to assess systolic peak velocities in a cohort of bicuspid aortic valve (BAV) patients. Materials and Methods 4D flow MRI at 1.5T was performed on 51 BAV patients. MIPs were generated from the 4D flow MRI velocity data and used by two users to determine peak velocities in three regions of interest (ROIs): ascending aorta (AAo), aortic arch, and descending aorta. 4D flow MRI peak velocities in the AAo were compared to peak velocities recorded by 2D phase contrast MRI (2D PCMRI) in a subcohort of 36 patients and by Doppler echocardiography in a subcohort of 34 patients. 4D flow MRI peak velocities recorded by each observer were compared for all ROIs to test for interobserver variability. Results 4D flow MRI recorded significantly higher velocities compared to 2D PCMRI (2.04 ± 0.71 m/s vs. 1.69 ± 0.79 m/s, 17.2% difference, P < 0.001) and similar velocities compared to Doppler echocardiography. There was excellent agreement between the observers, with a mean difference of 0.005 m/s and an intraclass correlation coefficient of 0.98. Conclusion 4D flow MRI velocity MIPs allow for efficient measurement of peak velocities in BAV patients with higher accuracy than 2D PCMRI and similar accuracy to Doppler echocardiography. J. Magn. Reson. Imaging 2016;44:1673–1682.

To systematically investigate the reproducibility of global and segmental left ventricular (LV) velocities derived from tissue phase mapping (TPM). Breath held and ECG synchronized TPM data (spatial/temporal resolution = 2 × 2 mm²/20.8 ms) were acquired in 18 healthy volunteers. To analyze scan–rescan variability, TPM was repeated in all subjects during a second visit separated by 16 ± 5 days. Data analysis included LV segmentation, and quantification of global and regional (AHA 16-segment modal) metrics of LV function [velocity–time curves, systolic and diastolic peak and time-to-peak (TTP) velocities] for radial (Vr), long-axis (Vz) and circumferential (VΦ) LV velocities. Mean velocity time curves in basal, mid-ventricular, and apical locations showed highly similar LV motion patterns for all three velocity components (Vr, VΦ, Vz) for scan and rescan. No significant differences for both systolic and diastolic peak and TTP myocardial velocities were observed. Segmental analysis revealed similar regional peak Vr and Vz during both systole and diastole except for three LV segments (p = 0.045, p = 0.033, and p = 0.009). Excellent (p < 0.001) correlations between scans and rescan for peak Vr (R² = 0.92), peak Vz (R² = 0.90), radial TTP (R² = 0.91) and long-axis TTP (R² = 0.88) confirmed good agreement. Bland–Altman analysis demonstrated excellent intra-observer and good inter-observer analysis agreement but increased variability for long axis peak velocities. TPM based analysis of global and regional myocardial velocities can be performed with good reproducibility. Robustness of regional quantification of long-axis velocities was limited but spatial velocity distributions across the LV could reliably be replicated.