Diabetic retinopathy (DR) is a common chronic fundus disease, which has four different kinds of microvessel structure and microvascular lesions: microaneurysms (MAs), hemorrhages (HEs), hard exudates, and soft exudates. Accurate detection and counting of them are a basic but important work. The manual annotation of these lesions is a labor-intensive task in clinical analysis. To solve the problem, we proposed a novel segmentation method for different lesions in DR. Our method is based on a convolutional neural network and can be divided into encoder module, attention module, and decoder module, so we refer it as EAD-Net. After normalization and augmentation, the fundus images were sent to the EAD-Net for automated feature extraction and pixel-wise label prediction. Given the evaluation metrics based on the matching degree between detected candidates and ground truth lesions, our method achieved sensitivity of 92.77%, specificity of 99.98%, and accuracy of 99.97% on the e_ophtha_EX dataset and comparable AUPR (Area under Precision-Recall curve) scores on IDRiD dataset. Moreover, the results on the local dataset also show that our EAD-Net has better performance than original U-net in most metrics, especially in the sensitivity and F1-score, with nearly ten percent improvement. The proposed EAD-Net is a novel method based on clinical DR diagnosis. It has satisfactory results on the segmentation of four different kinds of lesions. These effective segmentations have important clinical significance in the monitoring and diagnosis of DR.
This study is aimed at developing an intelligent algorithm based on deep learning and discussing its application for the classification and diagnosis of retinal vein occlusions (RVO) using fundus images. A total of 501 fundus images of healthy eyes and patients with RVO were used for model training and testing to investigate an intelligent diagnosis system. The images were first classified into four categories by fundus disease specialists: (i) healthy fundus (group 0), (ii) branch RVO (BRVO) (group 1), (iii) central RVO (CRVO) (group 2), and (iv) macular branch RVO (MBRVO) (group 3), before being diagnosed using the ResNet18 network model. Intelligent diagnoses were compared with clinical diagnoses. The specificity of the intelligent diagnosis system under each attention mechanism was 100% in group 0 and also revealed a high sensitivity of over 95%, score of over 97%, and an accuracy of over 97% in this group. For the other three groups, the specificities of diagnosis ranged from 0.45 to 0.91 with different attention mechanisms, in which the ResNet18+coordinate attention (CA) model had the highest specificities of 0.91, 0.88, and 0.83 for groups 1, 2, and 3, respectively. It also provided a high accuracy of over 94% with a coordinate attention mechanism in all four groups. The intelligent diagnosis and classifier system developed herein based on deep learning can determine the presence of RVO and classify disease according to the site of occlusion. This proposed system is expected to provide a new tool for RVO diagnosis and screening and will help solve the current challenges due to the shortage of medical resources.
The eye is one of the most important organs of the human body. Eye diseases are closely related to other systemic diseases, both of which influence each other. Numerous systemic diseases lead to special clinical manifestations and complications in the eyes. Typical diseases include diabetic retinopathy, hypertensive retinopathy, thyroid associated ophthalmopathy, optic neuromyelitis, and Behcet’s disease. Systemic disease-related ophthalmopathy is usually a chronic disease, and the analysis of imaging markers is helpful for a comprehensive diagnosis of these diseases. Recently, artificial intelligence (AI) technology based on deep learning has rapidly developed, leading to numerous achievements and arousing widespread concern. Presently, AI technology has made significant progress in research on imaging markers of systemic disease-related ophthalmopathy; however, there are also many limitations and challenges. This article reviews the research achievements, limitations, and future prospects of AI image analysis technology in systemic disease-related ophthalmopathy.
21 May 2009 in Nephrology Dialysis Transplantation
Background. Continuous renal replacement therapy (CRRT) techniques have occupied an important position in the intensive care units (ICU). Serum proteome alteration and protein removal in this process are not clear. Since it has a poor understanding of mechanism of the treatment, there is a specific need of proteomics research for CRRT. The aim of this research was to study the serum proteome alterations of severe sepsis patients in the treatment of CRRT. Improved knowledge of proteome alteration could lead to the development of more efficient treatment strategies. Methods. In this study, 20 severe sepsis patients were enrolled. A proteomic approach with two-dimensional gel electrophoresis (2-DE) coupled with mass spectrometry and bioinformatics methods was utilized to identify proteins with altered expression at different times in the treatment of continuous veno-venous haemofiltration (CVVH). All proteins were identified on the appearance of the 2-DE gel at the appropriate molecular size and pI and score from peptide mass fingerprinting. Protein identifications were confirmed by sequencing of the tryptic peptides and an independent database search based on the sequence. A further validation study was performed by western blot. Results. Thirty-four protein spots expressed differentially were separated. Ten proteins were identified to be the commonly differentially expressed proteins in the treatment. Seven proteins decreased in the serum and three increased. Conclusions. This study gives a novel overview of serum proteome alteration of severe sepsis patient in the treatment of CVVH. Potentially interesting proteins have been revealed that are different from those identified by method of traditional biology.
6 May 2022 in Biochemical and Biophysical Research Communications
Background: Recently, periumbilical fat (PF) mass, an autologous material with a high survival rate, has been transplanted to treat sunken or dissatisfactory double eyelids. However, the intricate complications of PF grafts and associated reconstructive strategies are infrequently discussed. Methods: During 3 years, 20 patients (33 eyes) with eyelid malformations caused by PF grafts into the orbital septum or on the surface of the levator aponeurosis underwent corrective blepharoplasty. We recorded patients’ subjective feelings and identified deformities from crease abnormalities, bloated appearance, and problems with the eyelid’s height. Then, we categorize them into three types based on their complexity: type I, swollen appearance; type II, obvious adhesion; type III, severe comprehensive damage. The relevant management included removing fat implants, releasing the adhesion, and rebuilding the physical structure according to the anatomic damage mechanism. The improvement effect was assessed with a satisfaction survey from patients and doctors at 6 months of follow-up. Results: The swollen appearance was observed in 26 eyes (78.8%), an unsmooth double-eyelid line was in 23 eyes (69.7%), and the incidence of adhesion was in 22 eyes (66.7%). Following a comprehensive evaluation, 15 eyes (45.5%) and 13 (39.4%) were classified as type I and type II respectively. After the 6-month follow-up, 22 eyes (66.7%) showed exceptional aesthetic results, whereas only 2 eyes as type III had a poor outcome. Conclusions: The deformities emerging from periumbilical fat into the upper eyelid are associated with the shape of the fat and the adhesion in tissues. Graft removal, adhesion release, and restoration of the natural anatomic structure can have positive outcomes.