DFTSA-Net: Deep Feature Transfer-Based Stacked Autoencoder Network for DME Diagnosis

Detection of Ocular Surface Squamous Neoplasia Using Artificial Intelligence With Anterior Segment Optical Coherence Tomography

PURPOSE

To develop and validate a deep learning (DL) model to differentiate ocular surface squamous neoplasia (OSSN) from pterygium and pinguecula using high-resolution anterior segment optical coherence tomography (AS-OCT).

DESIGN

Retrospective Diagnostic Accuracy Study.

METHODS

Setting: Single-center.

STUDY POPULATION

All eyes with a clinical or biopsy-proven diagnosis of OSSN, pterygium, or pinguecula that received AS-OCT imaging.

PROCEDURES

Imaging data was extracted from Optovue AS-OCT (Fremont, CA) and patients' clinical or biopsy-proven diagnoses were collected from electronic medical records. A DL classification model was developed using two methodologies: (1) a masked autoencoder was trained with unlabeled data from 105,859 AS-OCT images of 5746 eyes and (2) a Vision Transformer supervised model coupled to the autoencoder used labeled data for fine-tuning a binary classifier (OSSN vs non-OSSN lesions). A sample of 2022 AS-OCT images from 523 eyes (427 patients) were classified by expert graders into "OSSN or suspicious for OSSN" and "pterygium or pinguecula." The algorithm's diagnostic performance was evaluated in a separate test sample using 566 scans (62 eyes, 48 patients) with biopsy-proven OSSN and compared with expert clinicians who were masked to the diagnosis. Analysis was conducted at the scan-level for both the DL model and expert clinicians, who were not provided with clinical images or supporting clinical data.

MAIN OUTCOME

Diagnostic performance of expert clinicians and the DL model in identifying OSSN on AS-OCT scans.

RESULTS

The DL model had an accuracy of 90.3% (95% confidence intervals [CI]: 87.5%-92.6%), with sensitivity of 86.4% (95% CI: 81.4%-90.4%) and specificity of 93.2% (95% CI: 89.9%-95.7%) compared to the biopsy-proven diagnosis. Expert graders had a lower sensitivity 69.8% (95% CI: 63.6%-75.5%) and slightly higher specificity 98.5% (95% CI: 96.4%-99.5%) than the DL model. The area under the receiver operating characteristic curve for the DL model was 0.945 (95% CI: 0.918-0.972) and significantly greater than expert graders (area under the receiver operating characteristic curve = 0.688, P < .001).

CONCLUSIONS

A DL model applied to AS-OCT scans demonstrated high accuracy, sensitivity, and specificity in differentiating OSSN from pterygium and pinguecula. Interestingly, the model had comparable diagnostic performance to expert clinicians in this study and shows promise for enhancing clinical decision-making. Further research is warranted to explore the integration of this artificial intelligence-driven approach in routine screening and diagnostic protocols for OSSN.

A bibliometric analysis of artificial intelligence applications in macular edema: exploring research hotspots and Frontiers

Background: Artificial intelligence (AI) is used in ophthalmological disease screening and diagnostics, medical image diagnostics, and predicting late-disease progression rates. We reviewed all AI publications associated with macular edema (ME) research Between 2011 and 2022 and performed modeling, quantitative, and qualitative investigations. Methods: On 1st February 2023, we screened the Web of Science Core Collection for AI applications related to ME, from which 297 studies were identified and analyzed (2011-2022). We collected information on: publications, institutions, country/region, keywords, journal name, references, and research hotspots. Literature clustering networks and Frontier knowledge bases were investigated using bibliometrix-BiblioShiny, VOSviewer, and CiteSpace bibliometric platforms. We used the R "bibliometrix" package to synopsize our observations, enumerate keywords, visualize collaboration networks between countries/regions, and generate a topic trends plot. VOSviewer was used to examine cooperation between institutions and identify citation relationships between journals. We used CiteSpace to identify clustering keywords over the timeline and identify keywords with the strongest citation bursts. Results: In total, 47 countries published AI studies related to ME; the United States had the highest H-index, thus the greatest influence. China and the United States cooperated most closely between all countries. Also, 613 institutions generated publications - the Medical University of Vienna had the highest number of studies. This publication record and H-index meant the university was the most influential in the ME field. Reference clusters were also categorized into 10 headings: retinal Optical Coherence Tomography (OCT) fluid detection, convolutional network models, deep learning (DL)-based single-shot predictions, retinal vascular disease, diabetic retinopathy (DR), convolutional neural networks (CNNs), automated macular pathology diagnosis, dry age-related macular degeneration (DARMD), class weight, and advanced DL architecture systems. Frontier keywords were represented by diabetic macular edema (DME) (2021-2022). Conclusion: Our review of the AI-related ME literature was comprehensive, systematic, and objective, and identified future trends and current hotspots. With increased DL outputs, the ME research focus has gradually shifted from manual ME examinations to automatic ME detection and associated symptoms. In this review, we present a comprehensive and dynamic overview of AI in ME and identify future research areas.

Intravitreal injection of Conbercept combined with micropulse laser therapy enhances clinical efficacy in patients with diabetic macular edema

PURPOSE

To analyze the clinical efficacy of intravitreal injection of Conbercept (IVC) combined with micropulse laser (MPL) therapy in the treatment of diabetic macular edema (DME).

METHODS

In this retrospective study, we selected 64 DME patients who visited the First People's Hospital of Yunnan Province between February 2019 and February 2021 for analysis. Based on different intervention methods, 31 cases treated with IVC were included as a control group (the Con group) and 33 cases with IVC + MPL combination therapy were in a research group (the Res group). Data on curative effects, injection frequency, pre- and post-treatment best corrected visual acuity (BCVA) and central macular thickness (CMT), visual field gray value, 30° visual field average light threshold sensitivity, and mean visual field defect (VFD) were collected for inter-group comparisons. Further, Cox multivariate regression analysis was performed to identify factors affecting the curative efficacy of DME patients.

RESULTS

Compared with the Con group, the Res group had a higher total response rate and a lower injection frequency. In addition, higher BCVA and lower CMT were determined in the Res after 6 months of treatment. Moreover, Res group exhibited statistically lower visual field gray value and mean VFD, as well as higher 30° visual field average light threshold sensitivity than the Con at 1 month postoperatively. All the above differences were statistically significant. According to the Cox multivariate regression analysis, treatment modality was the influencing factor for the efficacy of DME patients.

CONCLUSIONS

IVC + MPL have better clinical efficacy than IVC alone for DME. The combined modality can improve patients' visual quality, inhibit DME, and reduce medication frequency.

Clinical Decision Support Framework for Segmentation and Classification of Brain Tumor MRIs Using a U-Net and DCNN Cascaded Learning Algorithm

Brain tumors (BTs) are an uncommon but fatal kind of cancer. Therefore, the development of computer-aided diagnosis (CAD) systems for classifying brain tumors in magnetic resonance imaging (MRI) has been the subject of many research papers so far. However, research in this sector is still in its early stage. The ultimate goal of this research is to develop a lightweight effective implementation of the U-Net deep network for use in performing exact real-time segmentation. Moreover, a simplified deep convolutional neural network (DCNN) architecture for the BT classification is presented for automatic feature extraction and classification of the segmented regions of interest (ROIs). Five convolutional layers, rectified linear unit, normalization, and max-pooling layers make up the DCNN's proposed simplified architecture. The introduced method was verified on multimodal brain tumor segmentation (BRATS 2015) datasets. Our experimental results on BRATS 2015 acquired Dice similarity coefficient (DSC) scores, sensitivity, and classification accuracy of 88.8%, 89.4%, and 88.6% for high-grade gliomas. When it comes to segmenting BRATS 2015 BT images, the performance of our proposed CAD framework is on par with existing state-of-the-art methods. However, the accuracy achieved in this study for the classification of BT images has improved upon the accuracy reported in prior studies. Image classification accuracy for BRATS 2015 BT has been improved from 88% to 88.6%.

BO-ALLCNN: Bayesian-Based Optimized CNN for Acute Lymphoblastic Leukemia Detection in Microscopic Blood Smear Images

Acute lymphoblastic leukemia (ALL) is a deadly cancer characterized by aberrant accumulation of immature lymphocytes in the blood or bone marrow. Effective treatment of ALL is strongly associated with the early diagnosis of the disease. Current practice for initial ALL diagnosis is performed through manual evaluation of stained blood smear microscopy images, which is a time-consuming and error-prone process. Deep learning-based human-centric biomedical diagnosis has recently emerged as a powerful tool for assisting physicians in making medical decisions. Therefore, numerous computer-aided diagnostic systems have been developed to autonomously identify ALL in blood images. In this study, a new Bayesian-based optimized convolutional neural network (CNN) is introduced for the detection of ALL in microscopic smear images. To promote classification performance, the architecture of the proposed CNN and its hyperparameters are customized to input data through the Bayesian optimization approach. The Bayesian optimization technique adopts an informed iterative procedure to search the hyperparameter space for the optimal set of network hyperparameters that minimizes an objective error function. The proposed CNN is trained and validated using a hybrid dataset which is formed by integrating two public ALL datasets. Data augmentation has been adopted to further supplement the hybrid image set to boost classification performance. The Bayesian search-derived optimal CNN model recorded an improved performance of image-based ALL classification on test set. The findings of this study reveal the superiority of the proposed Bayesian-optimized CNN over other optimized deep learning ALL classification models.

A Hybrid Deep Transfer Learning of CNN-Based LR-PCA for Breast Lesion Diagnosis via Medical Breast Mammograms

One of the most promising research areas in the healthcare industry and the scientific community is focusing on the AI-based applications for real medical challenges such as the building of computer-aided diagnosis (CAD) systems for breast cancer. Transfer learning is one of the recent emerging AI-based techniques that allow rapid learning progress and improve medical imaging diagnosis performance. Although deep learning classification for breast cancer has been widely covered, certain obstacles still remain to investigate the independency among the extracted high-level deep features. This work tackles two challenges that still exist when designing effective CAD systems for breast lesion classification from mammograms. The first challenge is to enrich the input information of the deep learning models by generating pseudo-colored images instead of only using the input original grayscale images. To achieve this goal two different image preprocessing techniques are parallel used: contrast-limited adaptive histogram equalization (CLAHE) and Pixel-wise intensity adjustment. The original image is preserved in the first channel, while the other two channels receive the processed images, respectively. The generated three-channel pseudo-colored images are fed directly into the input layer of the backbone CNNs to generate more powerful high-level deep features. The second challenge is to overcome the multicollinearity problem that occurs among the high correlated deep features generated from deep learning models. A new hybrid processing technique based on Logistic Regression (LR) as well as Principal Components Analysis (PCA) is presented and called LR-PCA. Such a process helps to select the significant principal components (PCs) to further use them for the classification purpose. The proposed CAD system has been examined using two different public benchmark datasets which are INbreast and mini-MAIS. The proposed CAD system could achieve the highest performance accuracies of 98.60% and 98.80% using INbreast and mini-MAIS datasets, respectively. Such a CAD system seems to be useful and reliable for breast cancer diagnosis.

Hybrid Feature Reduction Using PCC-Stacked Autoencoders for Gold/Oil Prices Forecasting under COVID-19 Pandemic

The financial markets have been influenced by the emerging spread of Coronavirus disease, COVID-19. The oil, and gold as well have experienced a downward trend due to the increased rate in the number of confirmed COVID-19 cases. Lately, the published COVID data comprised new variables that may influence the accuracy of the oil/gold prices forecasting models including the Stringency index, Reproduction rate, Positive Rate, and Vaccinations. In this study, Deep Autoencoders are introduced and combined with the well-known approach: Pearson Correlation Coefficient, PCC, analysis in selecting the key features that affect the accuracy of the forecasting models of gold and oil prices with respect to COVID-19 pandemic. We have utilized a hybrid approach of PCC along with a 2-Stage Stacked Autoencoder, SA, to extract the latent features which are then submitted to Neural Network, NN, regression model. The NN regressor has been trained using the Bayesian Regularization-backpropagation algorithm which provides a good generalization for small noisy datasets. The hybrid approach has yielded the minimum MSE values of 8.97 × 10−3 and 5.356 × 10−2 on the oil/gold validation set, respectively. Compared to the existing approaches, the proposed approach has outperformed the ARIMA, ML based regression models in forecasting the oil/gold prices. In addition, the introduced framework has yielded lower Mean Absolute Error, MAE, than the Recurrent Neural Network, RNN, and the Principal Component Analysis, PCA, for dimension reduction. The retrieved results showed that the hybrid method produced more robust features by considering the relationship between the input features.