Retinal microvascular complexity as a putative biomarker of biological age: a pilot study.

Physiological changes associated with aging increase the risk for the development of age-related diseases. This increase is non-specific to the type of age-related disease, although each disease develops through a unique pathophysiologic mechanism. People who age at a faster rate develop age-related diseases earlier in their life. They have an older "biological age" compared to their "chronological age". Early detection of individuals with accelerated aging would allow timely intervention to postpone the onset of age-related diseases. This would increase their life expectancy and their length of good quality life. The goal of this study was to investigate whether retinal microvascular complexity could be used as a biomarker of biological age. Retinal images of 68 participants ages ranging from 19 to 82 years were collected in an observational cross-sectional study. Twenty of the old participants had age-related diseases such as hypertension, type 2 diabetes, and/or Alzheimer's dementia. The rest of the participants were healthy. Retinal images were captured by a hand-held, non-mydriatic fundus camera and quantification of the microvascular complexity was performed by using Sholl's, box-counting fractal, and lacunarity analysis. In the healthy subjects, increasing chronological age was associated with lower retinal microvascular complexity measured by Sholl's analysis. Decreased box-counting fractal dimension was present in old patients, and this decrease was 2.1 times faster in participants who had age-related diseases (p = 0.047). Retinal microvascular complexity could be a promising new biomarker of biological age. The data from this study is the first of this kind collected in Montenegro. It is freely available for use.

Developing Edge AI Computer Vision for Smart Poultry Farms Using Deep Learning and HPC.

This research describes the use of high-performance computing (HPC) and deep learning to create prediction models that could be deployed on edge AI devices equipped with camera and installed in poultry farms. The main idea is to leverage an existing IoT farming platform and use HPC offline to run deep learning to train the models for object detection and object segmentation, where the objects are chickens in images taken on farm. The models can be ported from HPC to edge AI devices to create a new type of computer vision kit to enhance the existing digital poultry farm platform. Such new sensors enable implementing functions such as counting chickens, detection of dead chickens, and even assessing their weight or detecting uneven growth. These functions combined with the monitoring of environmental parameters, could enable early disease detection and improve the decision-making process. The experiment focused on Faster R-CNN architectures and AutoML was used to identify the most suitable architecture for chicken detection and segmentation for the given dataset. For the selected architectures, further hyperparameter optimization was carried out and we achieved the accuracy of AP = 85%, AP50 = 98%, and AP75 = 96% for object detection and AP = 90%, AP50 = 98%, and AP75 = 96% for instance segmentation. These models were installed on edge AI devices and evaluated in the online mode on actual poultry farms. Initial results are promising, but further development of the dataset and improvements in prediction models is needed.

Forecasting Day-Ahead Electricity Metrics with Artificial Neural Networks.

As artificial neural network architectures grow increasingly more efficient in time-series prediction tasks, their use for day-ahead electricity price and demand prediction, a task with very specific rules and highly volatile dataset values, grows more attractive. Without a standardized way to compare the efficiency of algorithms and methods for forecasting electricity metrics, it is hard to have a good sense of the strengths and weaknesses of each approach. In this paper, we create models in several neural network architectures for predicting the electricity price on the HUPX market and electricity load in Montenegro and compare them to multiple neural network models on the same basis (using the same dataset and metrics). The results show the promising efficiency of neural networks in general for the task of short-term prediction in the field, with methods combining fully connected layers and recurrent neural or temporal convolutional layers performing the best. The feature extraction power of convolutional layers shows very promising results and recommends the further exploration of temporal convolutional networks in the field.

TREND database: Retinal images of healthy young subjects visualized by a portable digital non-mydriatic fundus camera.

Topological characterization of the Retinal microvascular nEtwork visualized by portable fuNDus camera (TREND) is a database comprising of 72 color digital retinal images collected from the students of the Faculty of Medicine at the University of Montenegro, in the period from February 18th to March 11th 2020. The database also includes binarized images of manually segmented microvascular networks associated with each raw image. The participant demographic characteristics, health status, and social habits information such as age, sex, body mass index, smoking history, alcohol use, as well as previous medical history was collected. As proof of the concept, a smaller set of 10 color digital fundus images from healthy older participants is also included. Comparison of the microvascular parameters of these two sets of images demonstrate that digital fundus images recorded with a hand-held portable camera are able to capture the changes in patterns of microvascular network associated with aging. The raw images from the TREND database provide a standard that defines normal retinal anatomy and microvascular network geometry in young healthy people in Montenegro as it is seen with the digital hand-held portable non-mydriatic MiiS HORUS Scope DEC 200.This knowledge could facilitate the application of this technology at the primary level of health care for large scale telematic screening for complications of chronic diseases, such as hypertensive and diabetic retinopathy. In addition, it could aid in the development of new methods for early detection of age-related changes in the retina, systemic chronic diseases, as well as eye-specific diseases. The associated manually segmented images of the microvascular networks provide the standard that can be used for development of automatic software for image quality assessment, segmentation of microvascular network, and for computer-aided detection of pathological changes in retina. The TREND database is freely available at https://doi.org/10.5281/zenodo.4521043.

Regional Patterns in Retinal Microvascular Network Geometry in Health and Disease.

The study explores the regional differences in microvascular geometry between the optic disc (O) and the macular area (M) in health and disease. Skeletonized manually segmented vascular networks from 15 healthy, 15 retinas with diabetic retinopathy (DR), and 15 retinas with glaucoma from publicly available High-Resolution Fundus (HRF) image database were used. When visualized by a digital fundus camera, O has a substantial proportion of small arteries and larger arterioles, while M contains smaller arterioles at the periphery and avascular zone in the center. We hypothesized that in pathological conditions the vascular network remodelling patterns in these two regions may be different. The analysis of box-counting fractal dimension (Db), lacunarity (Λ), and microvascular density showed that in healthy retinas, Λ and vessel density were lower in the M compared to the O, while the Db did not change. In retinas with DR, the Db was the lowest in the M, which was different from all other groups. The vessel density followed this trend. Lacunarity was the highest in the O of DR group compared to all other groups. The results show that in DR various regions of retinal microvascular network remodel in a different manner and to different extent.

Fractal characterization of retinal microvascular network morphology during diabetic retinopathy progression.

OBJECTIVE: The study aimed to characterize morphological changes of the retinal microvascular network during the progression of diabetic retinopathy. METHODS: Publicly available retinal images captured by a digital fundus camera from DIARETDB1 and STARE databases were used. The retinal microvessels were segmented using the automatic method, and vascular network morphology was analyzed by fractal parametrization such as box-counting dimension, lacunarity, and multifractals. RESULTS: The results of the analysis were affected by the ability of the segmentation method to include smaller vessels with more branching generations. In cases where the segmentation was more detailed and included a higher number of vessel branching generations, increased severity of diabetic retinopathy was associated with increased complexity of microvascular network as measured by box-counting and multifractal dimensions, and decreased gappiness of retinal microvascular network as measured by lacunarity parameter. This association was not observed if the segmentation method included only 3-4 vessel branching generations. CONCLUSIONS: Severe stages of diabetic retinopathy could be detected noninvasively by using high resolution fundus photography and automatic microvascular segmentation to the high number of branching generations, followed by fractal analysis parametrization. This approach could improve risk stratification for the development of microvascular complications, cardiovascular disease, and dementia in diabetes.

Manually segmented vascular networks from images of retina with proliferative diabetic and hypertensive retinopathy.

In this article we present a data set that contains 37 image files obtained by manual vessel segmentation of raw retinal images from Structured Analysis of the Retina (STARE) database ("The STARE Project", 2018) [1]. Our expert segmented 8 images that are associated with the single diagnosis of hypertensive retinopathy and 9 images with the single diagnosis of proliferative diabetic retinopathy (Popovic et al., 2018) [2]. To validate the manual segmentation, the same expert additionally segmented a gold standard set of 20 raw images from the STARE database. Raw images of retinas associated with either diabetic proliferative retinopathy or hypertensive retinopathy display the intricate and very different morphologies of retinal microvascular networks. Very frequently, they also have pathological changes such as exudates and hemorrhages. The presence of these changes, as well as neovascularization in proliferative diabetic retinopathy, poses a significant challenge for researchers who are developing automatic methods for retinal vessel segmentation. Therefore, this data set can be useful for the development of methods for automatic segmentation. In addition, the data can be used for development of methods for quantitation of microvascular morphology of the retina in various pathological conditions.

Fractal dimension and lacunarity analysis of retinal microvascular morphology in hypertension and diabetes.

Hypertension and diabetes mellitus represent modifiable risk factors for vascular disease. They cause microvascular remodeling, and ultimately result in end-organ damage. Therefore, development of methods for noninvasive quantification of the effects of hypertension and diabetes mellitus on microvasculature is of paramount importance. The two goals of the study were: 1) to characterize the geometric complexity and inhomogeneity of retinal vasculature in hypertensive retinopathy (HR) and in proliferative diabetic retinopathy (PDR) by using box counting fractal dimension and lacunarity analysis, and 2) to determine if the combination of these two parameters can be used to describe differences in the vascular tree geometry between HR and PDR. The extended set of retinal images from the publicly available STARE database was manually segmented by our expert, validated, and made available for other researchers to use. The healthy retinal vascular network has a higher complexity (fractal dimension) compared to that in HR and in PDR. However, there is no difference in microvascular complexity between HR and PDR. The inhomogeneity of the retinal microvascular tree (lacunarity) was higher in PDR compared to HR. Lacunarity and fractal dimension together quantitatively characterize microvascular geometry in the retina with higher specificity than fractal analysis alone.

A Moodle-based blended learning solution for physiology education in Montenegro: a case study.

This study evaluates the impact of web-based blended learning in the physiology course at the Faculty of Medicine, University of Montenegro. The two main goals of the study were: to determine the impact of e-learning on student success in mastering the course, and to assess user satisfaction after the introduction of e-learning. The study compared a group of students who attended the physiology course before, with a group of students who attended the physiology course after the Moodle platform was fully implemented as an educational tool. Formative and summative assessment scores were compared between these two groups. The impact of high vs. low Moodle use on the assessment scores was analyzed. The satisfaction among Moodle users was assessed by the survey. The study found that attendance of face-to-face lectures had a positive impact on academic performance. The introduction of Moodle in the presented model of teaching increased interest of students, attendance of face-to-face lectures, as well as formative and summative scores. High frequency of Moodle use was not always associated with better academic performance, suggesting that the introduction of a new method of teaching was most likely equally accepted by low- and high-achieving students. Most of the students agreed that Moodle was easy to use and it complemented traditional teaching very well, but it could not completely replace traditional face-to-face lectures. The study supports continuing the use of web-based learning in a form of blended learning for physiology, as well as for other courses in medical education.