Retinal Image Analysis for Diabetes-Based Eye Disease Detection Using Deep Learning

Nazir, Tahira; Irtaza, Aun; Javed, Ali; Malik, Hafiz; Hussain, Dildar; Naqvi, Rizwan Ali

doi:10.3390/app10186185

Cited by 82 publications

(51 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hence, the detection accuracy of the ML-based techniques is dependent on the quality and representation of the extracted features, thus is limited and prone to errors in dealing with large datasets [24]. Meanwhile, DL-based algorithms have shown high performance in various industries including medical imaging [25][26][27]. The most common or well-known DL model is the convolutional neural network (CNN) that can instinctively learn dense characteristics directly from the training data due to its weight-sharing nature [28].…”

Section: Introductionmentioning

confidence: 99%

A Novel Deep Learning Method for Recognition and Classification of Brain Tumors from MRI Images

et al. 2021

Self Cite

View full text Add to dashboard Cite

A brain tumor is an abnormal growth in brain cells that causes damage to various blood vessels and nerves in the human body. An earlier and accurate diagnosis of the brain tumor is of foremost important to avoid future complications. Precise segmentation of brain tumors provides a basis for surgical planning and treatment to doctors. Manual detection using MRI images is computationally complex in cases where the survival of the patient is dependent on timely treatment, and the performance relies on domain expertise. Therefore, computerized detection of tumors is still a challenging task due to significant variations in their location and structure, i.e., irregular shapes and ambiguous boundaries. In this study, we propose a custom Mask Region-based Convolution neural network (Mask RCNN) with a densenet-41 backbone architecture that is trained via transfer learning for precise classification and segmentation of brain tumors. Our method is evaluated on two different benchmark datasets using various quantitative measures. Comparative results show that the custom Mask-RCNN can more precisely detect tumor locations using bounding boxes and return segmentation masks to provide exact tumor regions. Our proposed model achieved an accuracy of 96.3% and 98.34% for segmentation and classification respectively, demonstrating enhanced robustness compared to state-of-the-art approaches.

show abstract

Section: Introductionmentioning

confidence: 99%

A Novel Deep Learning Method for Recognition and Classification of Brain Tumors from MRI Images

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…The most commonly used methods for depression appraisal were: Patient Health Questionnaire (PHQ) [ 43 , 44 , 45 ], Manual for Depression Epidemic Study Center (CESD) [ 46 ], and the BDI. However, there are already major variations in different studies [ 14 , 33 , 47 , 48 , 49 , 50 ]. The key importance of this research was to determine patients who were considered to not be stressed but have diabetes distress according to the DSM manual guidelines.…”

Section: Discussionmentioning

confidence: 99%

An Empirical Study on Diabetes Depression over Distress Evaluation Using Diagnosis Statistical Manual and Chi-Square Method

Noman

Arshad

Zeeshan

et al. 2021

IJERPH

View full text Add to dashboard Cite

Diabetes distress is an alternative disorder that is often associated with depression syndromes. Psychosocial distress is an alternative disorder that acts as a resistance to diabetes self-care management and compromises diabetes control. Yet, in Nigeria, the focus of healthcare centers is largely inclined toward the medical aspect of diabetes that neglects psychosocial care. In this retrospective study, specific distress was measured by the Diabetes Distress Screening (DDS) scale, and depression was analyzed by the Beck Depression Inventory (BDI) and Diagnosis Statistics Manual (DSM) criteria in type 2 diabetes mellitus (T2DM) patients of Northwestern Nigeria. Additionally, we applied the Chi-square test and linear regression to measure the forecast prevalence ratio and evaluate the link between the respective factors that further determine the odd ratios and coefficient correlations in five nonintrusive variables, namely age, gender, physical exercise, diabetes history, and smoking. In total, 712 sample patients were taken, with 51.68% male and 47.31% female patients. The mean age and body mass index (BMI) was 48.6 years ± 12.8 and 45.6 years ± 8.3. Based on the BDI prediction, 90.15% of patients were found depressed according to the DSM parameters, and depression prevalence was recorded around 22.06%. Overall, 88.20% of patients had DDS-dependent diabetes-specific distress with a prevalence ratio of 24.08%, of whom 45.86% were moderate and 54.14% serious. In sharp contrast, emotion-related distress of 28.96% was found compared to interpersonal (23.61%), followed by physician (16.42%) and regimen (13.21%) distress. The BDI-based matching of depression signs was also statistically significant with p < 0.001 in severe distress patients. However, 10.11% of patients were considered not to be depressed by DSM guidelines. The statistical evidence indicates that depression and distress are closely correlated with age, sex, diabetes history, physical exercise, and smoking influences. The facts and findings in this work show that emotional distress was found more prevalent. This study is significant because it considered several sociocultural and religious differences between Nigeria and large, undeveloped, populated countries with low socioeconomic status and excessive epidemiological risk. Finally, it is important for the clinical implications of T2DM patients on their initial screenings.

show abstract

“…The deep learning (DL) computing paradigm has been deemed the gold standard in the medical image analysis field. It has been exhibiting excellent performance in several medical imaging areas, such as pathology [ 1 ], dermatology [ 2 ], radiology [ 3 , 4 ], and ophthalmology [ 5 , 6 ], which are the most competitive fields requiring human specialists. The recent approaches within DL being adapted to the direction of clinical alteration commonly depend on a large volume of highly reliable annotated images.…”

Section: Introductionmentioning

confidence: 99%

Novel Transfer Learning Approach for Medical Imaging with Limited Labeled Data

Alzubaidi

Al-Amidie

Al-Asadi

et al. 2021

Cancers

183

View full text Add to dashboard Cite

Deep learning requires a large amount of data to perform well. However, the field of medical image analysis suffers from a lack of sufficient data for training deep learning models. Moreover, medical images require manual labeling, usually provided by human annotators coming from various backgrounds. More importantly, the annotation process is time-consuming, expensive, and prone to errors. Transfer learning was introduced to reduce the need for the annotation process by transferring the deep learning models with knowledge from a previous task and then by fine-tuning them on a relatively small dataset of the current task. Most of the methods of medical image classification employ transfer learning from pretrained models, e.g., ImageNet, which has been proven to be ineffective. This is due to the mismatch in learned features between the natural image, e.g., ImageNet, and medical images. Additionally, it results in the utilization of deeply elaborated models. In this paper, we propose a novel transfer learning approach to overcome the previous drawbacks by means of training the deep learning model on large unlabeled medical image datasets and by next transferring the knowledge to train the deep learning model on the small amount of labeled medical images. Additionally, we propose a new deep convolutional neural network (DCNN) model that combines recent advancements in the field. We conducted several experiments on two challenging medical imaging scenarios dealing with skin and breast cancer classification tasks. According to the reported results, it has been empirically proven that the proposed approach can significantly improve the performance of both classification scenarios. In terms of skin cancer, the proposed model achieved an F1-score value of 89.09% when trained from scratch and 98.53% with the proposed approach. Secondly, it achieved an accuracy value of 85.29% and 97.51%, respectively, when trained from scratch and using the proposed approach in the case of the breast cancer scenario. Finally, we concluded that our method can possibly be applied to many medical imaging problems in which a substantial amount of unlabeled image data is available and the labeled image data is limited. Moreover, it can be utilized to improve the performance of medical imaging tasks in the same domain. To do so, we used the pretrained skin cancer model to train on feet skin to classify them into two classes—either normal or abnormal (diabetic foot ulcer (DFU)). It achieved an F1-score value of 86.0% when trained from scratch, 96.25% using transfer learning, and 99.25% using double-transfer learning.

show abstract

Retinal Image Analysis for Diabetes-Based Eye Disease Detection Using Deep Learning

Cited by 82 publications

References 69 publications

A Novel Deep Learning Method for Recognition and Classification of Brain Tumors from MRI Images

A Novel Deep Learning Method for Recognition and Classification of Brain Tumors from MRI Images

An Empirical Study on Diabetes Depression over Distress Evaluation Using Diagnosis Statistical Manual and Chi-Square Method

Novel Transfer Learning Approach for Medical Imaging with Limited Labeled Data

Contact Info

Product

Resources

About