Corneal densitometry changes after femtosecond laser-assisted intracorneal ring segments implantation in keratoconus.

Background: Intrastromal corneal ring segments are commonly implanted in the corneas of eyes with mild-to-moderate keratoconus; however, changes in corneal densitometry (CD) after implantation are a matter of debate in the current literature. We evaluated the changes in CD 1 and 3 months after femtosecond laser-assisted Keraring implantation. Methods: This retrospective, non-comparative, multicenter, case series study included patients with keratoconus who underwent femtosecond laser-assisted implantation of double segments with 90° and 160° arc lengths or two 160° arc length Keraring segments. Demographic and baseline clinical ophthalmic data were recorded. Corneal topography and tomography data acquired using a Pentacam HR Scheimpflug tomography system (Pentacam High Resolution; Oculus, Wetzlar, Germany) with a best-fit sphere were used as a reference surface. Using the Pentacam HR, CD measurements were acquired over a corneal area of 12 mm in total and at four concentric zones (0-2, 2-6, 6-10, and 10-12 mm) of three corneal stromal depths: 120 µm of the anterior corneal stromal layer, 60 µm of the posterior corneal stromal layer, and the central layer of stroma lying between these two layers. Results: We included 40 eyes of 40 patients, including 8 (20%) male and 32 (80%) female individuals, with a mean (standard deviation) age of 21.0 (6.4) years. We observed a significant improvement in the topographic values of steep keratometry (K), flat K, maximum K, and corneal astigmatism (all P < 0.05), but not in the mean K, thinnest corneal pachymetry, corneal thickness at the apex, back elevation, or front elevation (all P > 0.05). The mean total anterior, central, and posterior CD differed significantly among the time points, with a significant increase from the preoperative to the 1-month and 3-month postoperative visits (all P < 0.05) and no difference between those of the 1-month and 3-month postoperative visits (all P > 0.05). The mean CD for the anterior layer in the central, paracentral, and mid-peripheral zones, and the central layer in all four zones, differed significantly among time points, with a significant increase from the preoperative to the 1-month and 3-month postoperative visits (all P < 0.05), which remained unchanged from the 1-month to the 3-month postoperative visit (all P < 0.05), except for the central 2-6-mm zone, which decreased significantly from the 1-month to the 3-month postoperative visit (P < 0.001). The CD of the central 10-12-mm zone did not differ significantly in each pairwise comparison (all P > 0.05). In contrast, CD for the posterior layer in the paracentral zone decreased significantly from the preoperative to the 1-month and 3-month postoperative visits but increased, to a lesser extent, from the 1-month to the 3-month postoperative visit (all P < 0.05). Conclusions: Femtosecond laser-assisted Keraring implantation significantly changes CD, with improvement in most topography parameters. Further longitudinal studies with larger sample sizes are required to verify these preliminary findings.

Venous Thromboembolism Among Hospitalized Patients: Incidence and Adequacy of Thromboprophylaxis - A Retrospective Study.

Purpose: We aimed to determine the incidence of venous thromboembolism among hospitalized patients in Qatar as well as to analyze the adequacy of VTE assessment and prophylaxis in hospitalized patients. Design: Retrospective observational study. Setting: Four hospitals under Hamad Medical Corporation, Qatar. Participants: Patients over the age of 18 who were hospitalized between January 2015 and December 2019 and developed venous thromboembolism during hospitalization or within a month after discharge were included. Results: During the study period, 641,994 individuals were admitted to hospitals. The inclusion criteria were satisfied by 209 of them. The mean age was 51.25 years and 54.5% were males. Hypertension and diabetes mellitus were the most common comorbidities found in the overall group. The incidence of VTE was 32.55 [95% CI 28.4, 37.3] per 100,000 admission per year [0.032%]. The annual incidence was least in 2015 (17.8 per 100,000 admissions) and highest in 2018 (44.4 per 100,000 admissions). Eighty-six subjects had DVT, and 109 had PE, whereas 14 had both. And, 67.5% of the patients developed VTE during admission while, 32.5% developed within 1 month of discharge. Moreover, 22.9% of the patients with PE developed pulmonary embolism after discharge from the hospital. VTE assessment was performed on 64.7% of the patients, and 69.7% received VTE prophylaxis in accordance with guidelines. Conclusion: Although the occurrence of VTE among hospitalized patients in Qatar is low, healthcare providers need additional education and knowledge of VTE assessment and prophylaxis to follow guidelines for all patients at the time of admission. Furthermore, risk assessment for VTE should be done for all patients at the time of discharge to decide on post-discharge prophylaxis so that incidence of VTE after discharge can be minimized. Future studies should focus on patients who developed VTE after discharge from the hospital as well as on various risk factors.

Hyperbolic Deep Neural Networks: A Survey.

Recently, hyperbolic deep neural networks (HDNNs) have been gaining momentum as the deep representations in the hyperbolic space provide high fidelity embeddings with few dimensions, especially for data possessing hierarchical structure. Such a hyperbolic neural architecture is quickly extended to different scientific fields, including natural language processing, single-cell RNA-sequence analysis, graph embedding, financial analysis, and computer vision. The promising results demonstrate its superior capability, significant compactness of the model, and a substantially better physical interpretability than its counterpart in the euclidean space. To stimulate future research, this paper presents a comprehensive review of the literature around the neural components in the construction of HDNN, as well as the generalization of the leading deep approaches to the hyperbolic space. It also presents current applications of various tasks, together with insightful observations and identifying open questions and promising future directions.

Improved Centerline Extraction in Fully Automated Coronary Ostium Localization and Centerline Extraction Framework using Deep Learning.

Coronary artery extraction in cardiac CT angiography (CCTA) image volume is a necessary step for any quantitative assessment of stenoses and atherosclerotic plaque. In this work, we propose a fully automated workflow that depends on convolutional networks to extract the centerlines of the coronary arteries from CCTA image volumes, starting from identifying the ostium points and then tracking the vessel till its end based on its radius and direction. First, a regression U-Net is employed to identify the ostium points in the image volume, then these points are fed to an orientation and radius predictor CNN model to track and extract each artery till its end point. Our results show that an average of 96% of the ostium points were identified and located within less than 5mm from their true location. The coronary arteries centerlines extraction was performed with high accuracy and lower number of training parameters making it suitable for real clinical applications and continuous learning.