Article
Multidisciplinary Sciences
Ruochen Liu, Song Gao, Hengsheng Zhang, Simin Wang, Lun Zhou, Jiaming Liu
Summary: This paper introduces a deep learning-based algorithm for automatic segmentation of retinal vessels, which achieves better segmentation results with limited dataset capacity. Based on this, a combined diagnosis algorithm for vessel segmentation and diabetic retinopathy classification of retinal images is proposed.
Article
Endocrinology & Metabolism
Nicola Quinn, Laima Brazionis, Benjamin Zhu, Chris Ryan, Rossella D'Aloisio, Hongying Lilian Tang, Tunde Peto, Alicia Jenkins
Summary: The study evaluated an automated retinal image analysis system for the classification of diabetic retinopathy, finding high sensitivity in detecting retinal disorders, which could serve as a triage tool for human graders. The system also showed high sensitivity in detecting proliferative diabetic retinopathy, suggesting its potential benefits in diabetes care in resource-constrained regions.
Article
Computer Science, Information Systems
German Pinedo-Diaz, Susana Ortega-Cisneros, Eduardo Ulises Moya-Sanchez, Jorge Rivera, Pedro Mejia-Alvarez, Francisco J. Rodriguez-Navarrete, Abraham Sanchez
Summary: Diabetic retinopathy is a major cause of blindness, and automated diabetes screening plays a crucial role in identifying lesions and reducing the risk of blindness. This study proposes a deep learning-based algorithm that analyzes spatial domain features and image quality assessment metrics to perform efficient diabetes screening and detect significant features with low computational cost, achieving high sensitivity and specificity.
Article
Multidisciplinary Sciences
Niloy Sikder, Mehedi Masud, Anupam Kumar Bairagi, Abu Shamim Mohammad Arif, Abdullah-Al Nahid, Hesham A. Alhumyani
Summary: Diabetic Retinopathy (DR) is a serious global health issue caused by diabetes affecting the retina. This study presents a novel diagnosis method based on decision tree-based ensemble learning technique, achieving 94.20% classification accuracy and 93.51% F-measure on the Asia Pacific Tele-Ophthalmology Society 2019 Blindness Detection dataset.
Article
Endocrinology & Metabolism
Aki Kato, Keiichiro Fujishima, Kazuhisa Takami, Naomi Inoue, Noriaki Takase, Norihiro Suzuki, Katsuya Suzuki, Soichiro Kuwayama, Akiko Yamada, Katsuhisa Sakai, Ryosuke Horita, Miho Nozaki, Munenori Yoshida, Yoshio Hirano, Tsutomu Yasukawa, Yuichiro Ogura
Summary: The study aimed to investigate the feasibility of a remote interpretation system for retinal images using an ultra-widefield retinal imaging device. Remote interpretation proved to be useful for diabetic retinopathy screening, especially during the COVID-19 pandemic where it can ensure social distancing and reduce the number of ophthalmic visits for both patients and clinicians.
DIABETES RESEARCH AND CLINICAL PRACTICE
(2021)
Review
Medicine, General & Internal
Warren W. Pan, Thomas W. Gardner, Jennifer L. Harder
Summary: DRD remains a common cause of vision loss in working-age adults, with limited progress on new therapies due to the complexity of the human eye. Utilizing systems biology approaches that have been successful in DKD research may help clarify the molecular basis and progression of DRD.
JOURNAL OF CLINICAL MEDICINE
(2021)
Article
Computer Science, Information Systems
Meysam Tavakoli, Alireza Mehdizadeh, Afshin Aghayan, Reza Pourreza Shahri, Tim Ellis, Jamshid Dehmeshki
Summary: Detection of red lesions in color retinal images is crucial for preventing vision loss and blindness in diabetic retinopathy. Microaneurysms (MAs) are the most common lesions associated with DR and their detection is challenging due to low image contrast and variable imaging conditions. This study developed unsupervised and supervised techniques to intelligently address the MAs detection problem, achieving high sensitivity and specificity in detecting DR lesions.
Article
Engineering, Electrical & Electronic
Fathi Kallel, Amira Echtioui
Summary: Diabetic retinopathy is a major cause of blindness, and current diagnosis and treatment methods are challenging. Therefore, a comprehensive and automated screening method is needed for early detection and management of the disease. This study introduces a transfer learning-based optical image classification approach using four pretrained models, and the InceptionV3 model achieved the highest accuracy.
SIGNAL IMAGE AND VIDEO PROCESSING
(2023)
Article
Biology
Zhike Han, Bin Yang, Shuiguang Deng, Zhuorong Li, Zhou Tong
Summary: Diabetic retinopathy (DR) is a major cause of blindness in adults. Incorporating machine learning into DR grading can enhance medical diagnosis accuracy. This study proposes a category weighted network (CWN) to achieve data balance at the model level and introduces relation weighted labels to investigate label distance relationships. Experimental results demonstrate the excellent performance of the CWN on different DR datasets and the broad applicability of relation weighted labels in improving methods using one-hot labels. The proposed method achieves high kappa scores and accuracy on DDR and APTOS datasets.
COMPUTERS IN BIOLOGY AND MEDICINE
(2023)
Article
Biology
Manisha Saini, Seba Susan
Summary: Screening and diagnosis of diabetic retinopathy disease is a significant problem in the biomedical domain. The use of medical imagery from a patient's eye for computer-aided diagnosis has greatly advanced with the success of deep learning. However, challenges with imbalanced datasets, inconsistent annotations, limited samples, and inappropriate evaluation metrics have impacted the performance of deep learning models.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Medicine, General & Internal
Doaa Hassan, Hunter Mathias Gill, Michael Happe, Ashay D. Bhatwadekar, Amir R. Hajrasouliha, Sarath Chandra Janga
Summary: Diabetic retinopathy (DR) is a late microvascular complication of Diabetes Mellitus (DM) that could lead to permanent blindness in patients if not detected early. AI systems can accurately detect DR, improve accuracy, and assist in rapid diagnoses.
FRONTIERS IN MEDICINE
(2022)
Article
Biology
Xiaoling Luo, Honggang Zhang, Jingyong Su, Wai Keung Wong, Jinkai Li, Yong Xu
Summary: Diabetic retinopathy is a common and serious complication of diabetes, which can lead to blindness. This paper proposes a novel computer-aided shape analysis system for retinal vessels to detect structural and functional changes before the development of diabetic retinopathy.
COMPUTERS IN BIOLOGY AND MEDICINE
(2023)
Article
Neurosciences
Maria H. Madeira, Ines P. Marques, Sonia Ferreira, Diana Tavares, Torcato Santos, Ana Rita Santos, Joao Figueira, Conceicao Lobo, Jose Cunha-Vaz
Summary: Diabetic retinopathy is not only a microvascular disease, but also involves neurodegeneration, with different phenotypes identified based on the characteristics of neurodegeneration. Understanding these phenotypes may offer insights for improved treatment and management of diabetic individuals with retinopathy.
FRONTIERS IN NEUROSCIENCE
(2021)
Article
Medicine, General & Internal
K. K. Mujeeb Rahman, Mohamed Nasor, Ahmed Imran
Summary: Diabetic retinopathy is a vision impairment caused by degeneration of blood vessels in the retina. It can lead to blindness and is associated with diabetes. Using pre-recorded digital fundus images and machine learning models, accurate predictions of diabetic retinopathy can be made.
Article
Computer Science, Information Systems
Yuhao Niu, Lin Gu, Yitian Zhao, Feng Lu
Summary: In this study, we propose an interpretable deep learning approach for medical diagnosis. By isolating and visualizing neuron activation patterns, we can determine the symptoms that the deep learning model uses for prediction. We also introduce a new network, Patho-GAN, to synthesize medically plausible images, and demonstrate that we can manipulate the generated images to control the properties of lesions. Our approach is faster and produces higher-quality images than previous methods, and has the potential to be an effective solution for data augmentation.
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
(2022)
Article
Computer Science, Information Systems
Mohammad Alishahi, Mohammad Hossein Yaghmaee Moghaddam, Hamid Reza Pourreza
PEER-TO-PEER NETWORKING AND APPLICATIONS
(2018)
Article
Optics
Sedigheh Chaji, Alireza Pourreza, Hamidreza Pourreza, Modjtaba Rouhani
JOURNAL OF THE OPTICAL SOCIETY OF AMERICA A-OPTICS IMAGE SCIENCE AND VISION
(2018)
Article
Multidisciplinary Sciences
Mojtaba Masoudi, Hamid-Reza Pourreza, Mahdi Saadatmand-Tarzjan, Noushin Eftekhari, Fateme Shafiee Zargar, Masoud Pezeshki Rad
Article
Engineering, Biomedical
Noushin Eftekhari, Hamid-Reza Pourreza, Mojtaba Masoudi, Kamaledin Ghiasi-Shirazi, Ehsan Saeedi
BIOMEDICAL ENGINEERING ONLINE
(2019)
Article
Computer Science, Artificial Intelligence
Malihe Javidi, Ahad Harati, HamidReza Pourreza
ARTIFICIAL INTELLIGENCE IN MEDICINE
(2019)
Article
Computer Science, Artificial Intelligence
Shahaboddin Askarian Bajestani, Hamidreza Pourreza, Shayan Nalbandian
MACHINE VISION AND APPLICATIONS
(2019)
Article
Food Science & Technology
Morteza Mohamadzadeh Moghadam, Masoud Taghizadeh, Hassan Sadrnia, Hamid Reza Pourreza
FOOD SCIENCE & NUTRITION
(2020)
Article
Biology
Fateme Bameri, Hamid-Reza Pourreza, Amir-Hossein Taherinia, Mansour Aliabadian, Hamid-Reza Mortezapour, Raziyeh Abdilzadeh
Article
Behavioral Sciences
Elaheh Imani, Ahad Harati, Hamidreza Pourreza, Morteza Moazami Goudarzi
Summary: The study investigated the neural and behavioral representations of perceptual decision-making and found that decision-making was more influenced by coherence factor under different stimulus conditions and attention levels. The decision stage was identified to play a crucial role in the performance of the decision-making process.
Article
Environmental Sciences
Farzaneh Boroumand, Mohammad Taghi Shakeri, Touka Banaee, Hamidreza Pourreza, Hassan Doosti
Summary: The study evaluated the vascular density of ocular surface microvasculature and different clinical parameters in diabetic patients, finding significant linear effects of age and mean arterial pressure on the Area Occupied by Vessels (AOV), while showing nonlinear associations between BMI, daily UPE, HbA1C, and BUN with AOV. The study concluded that a tilted additive model can better elucidate the effects of different predictors, with AOV peaking at HbA1C of 9% and increasing when daily UPE exceeds 600 mg. These findings should be considered in future studies of diabetic patients' ocular surface vessels.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2021)
Article
Computer Science, Information Systems
Saeideh Yousefzadeh, Hamidreza Pourreza, Hamidreza Mahyar
Summary: Content-based image retrieval is the process of retrieving a subset of images based on visual contents, and this paper presents a simple and efficient image retrieval system that offers acceptable accuracy. The proposed method uses a dilated residual convolutional neural network with triplet loss to extract high-resolution representations. To enhance robustness, candidate regions of interest are obtained from each feature map and Generalized-Mean pooling is applied. Experimental results on challenging datasets show high accuracy.
JOURNAL OF SIGNAL PROCESSING SYSTEMS FOR SIGNAL IMAGE AND VIDEO TECHNOLOGY
(2023)
Article
Engineering, Biomedical
Mehdi Sadeghibakhi, Hamidreza Pourreza, Hamidreza Mahyar
Summary: The paper proposes a method for segmenting lesions in Multiple Sclerosis using FLAIR images. A patch-based Convolutional Neural Network is designed and experimental results show that the proposed method outperforms other methods.
IEEE JOURNAL OF TRANSLATIONAL ENGINEERING IN HEALTH AND MEDICINE
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Seyyed Hamed Naghavi, Hamidreza Pourreza
2018 8TH INTERNATIONAL CONFERENCE ON COMPUTER AND KNOWLEDGE ENGINEERING (ICCKE)
(2018)
Correction
Ophthalmology
Touka Banaee, Hamidreza Pourreza, Hassan Doosti, Mojtaba Abrishami, Asieh Ehsaei, Mohsen Basiry, Reza Pourreza
JOURNAL OF OPHTHALMIC & VISION RESEARCH
(2018)
Article
Computer Science, Artificial Intelligence
Malihe Javidi, Hadi Sadoghi Yazdi, H. R. Pourreza
JOURNAL OF MULTIPLE-VALUED LOGIC AND SOFT COMPUTING
(2018)