Showing 2 results for Machine Learning
Jafar Bakhtiar Shohani, Morteza Hajimahmoodzadeh, Hamidreza Fallah,
Volume 16, Issue 2 (7-2022)
Abstract
In this paper, we investigate the usage of machine learning in the detection and recognition of double stars. To do this, numerous images including one star and double stars are simulated. Then, 100 terms of Zernike expansion with random coefficients are considered as aberrations to impose on the aforementioned images. Also, a telescope with a specific aperture is simulated. In this work, two kinds of intensity are used, one is in-focus and the other is out-of-focus of the telescope. After these simulations, a convolutional neural network (CNN) is configured and designed and its input is simulated intensity patterns. After learning the network, we could recognize double stars at severe turbulence without needing phase correction with a very high accuracy level of more than 98%.
Neda Yaghoubi, Hassan Masumi, Mohammad Hossein Fatehi, Fereshteh Ashtari, Rahele Kafieh,
Volume 17, Issue 1 (1-2023)
Abstract
Background: Multiple Sclerosis (MS) is a chronic immune-mediated disease affecting the central nervous system, leading to various disturbances, including visual impairment. Early and accurate diagnosis of MS is critical for effective treatment and management. Scanning Laser Ophthalmoscopy (SLO) is a non-invasive technique that provides high-quality retinal images, serving as a promising resource for the early detection of MS. This research investigates a vessel-based approach for MS detection in SLO images using Long Short-Term Memory (LSTM) networks.
Material and Methods: A total of 106 Healthy Controls (HCs) and 39 MS patients (78 eyes) were enrolled. After implementing quality control measures and removing poor-quality or damaged images, the research utilized a total of 265 photos (73 MS and 192 HC). An approach for the early detection of MS in SLO images using LSTM network is introduced. This approach involves two steps: 1.It involves pre-processing and extracting vessels and then pre-training a deep neural network using the source dataset, and 2. tuning the network on the target dataset of SLO images.
The significance of vessel segmentation in MS detection is examined, and the application of the proposed method in improving diagnostic models is explored. The proposed approach achieves an accuracy rate of 97.44% when evaluated on a test dataset consisting of SLO pictures.
Through experiments on SLO datasets and employing the proposed vessel-based approach with LSTM, empirical results demonstrate that this approach contributes to the early detection of MS with high accuracy. These models exhibit the capability to accurately detect the disease with high precision and appropriate sensitivity.
