Impact of senescence cell signature in patients with non-small cell carcinoma and melanoma receiving PD-L1/PD-1 inhibitors.

Tumor cell senescence plays a crucial role in tumor immunity. We investigated whether the senescent cell signature (SCS) could predict prognosis in non-small cell carcinoma (NSCLC) and melanoma datasets treated with PD-L1/PD-1 inhibitors. Patients with high SCS expression exhibited elevated levels of interferon-gamma and T cell-inflamed signatures in three lung adenocarcinomas (LUAD), two squamous cell carcinoma (LUSC) and three melanoma datasets. The high SCS group was associated with PD-L1-related pathways such as IL6/JAK/STAT3 and TNF-alpha signaling via NF-kB in LUAD, LUSC, and melanoma datasets. A positive correlation was observed between several immune checkpoint markers and the SCS, indicating an immunosuppressive state in LUAD, LUSC and melanoma datasets. In patients treated with PD-1/PD-L1 inhibitors, a higher SCS was associated with a better prognosis, and a positive correlation between SCS and PD-L1 was observed in six independent NSCLC and three independent melanoma datasets. We used the LASSO Cox regression model to build a risk model focusing on the SCS genes that particularly predict prognosis. We confirmed that the model accurately predicts prognosis. However, the senescent immunohistochemical markers p16 and p21 could predict the response to PD-1/PD-L1 inhibitors in patients with LUSC and melanoma but not in patients with LUAD. SCS could serve as a valuable biomarker to complement PD-L1 expression in patients receiving PD-L1/PD-1 inhibitors.

Differentiating loss of consciousness causes through artificial intelligence-enabled decoding of functional connectivity.

Differential diagnosis of acute loss of consciousness (LOC) is crucial due to the need for different therapeutic strategies despite similar clinical presentations among etiologies such as nonconvulsive status epilepticus, metabolic encephalopathy, and benzodiazepine intoxication. While altered functional connectivity (FC) plays a pivotal role in the pathophysiology of LOC, there has been a lack of efforts to develop differential diagnosis artificial intelligence (AI) models that feature the distinctive FC change patterns specific to each LOC cause. Three approaches were applied for extracting features for the AI models: three-dimensional FC adjacency matrices, vectorized FC values, and graph theoretical measurements. Deep learning using convolutional neural networks (CNN) and various machine learning algorithms were implemented to compare classification accuracy using electroencephalography (EEG) data with different epoch sizes. The CNN model using FC adjacency matrices achieved the highest accuracy with an AUC of 0.905, with 20-s epoch data being optimal for classifying the different LOC causes. The high accuracy of the CNN model was maintained in a prospective cohort. Key distinguishing features among the LOC causes were found in the delta and theta brain wave bands. This research advances the understanding of LOC's underlying mechanisms and shows promise for enhancing diagnosis and treatment selection. Moreover, the AI models can provide accurate LOC differentiation with a relatively small amount of EEG data in 20-s epochs, which may be clinically useful.

Changes in the expression of cell interaction-related pathways during brain metastasis in lung adenocarcinoma: Gene expression and immunohistochemical analysis.

BACKGROUND: Brain metastasis (BM) is a prevalent prognostic event in the development of lung adenocarcinoma (LUAD) with a poor prognosis. Alterations in gene or protein expression during various phases of BM remain unclear. METHODS: We performed gene expression and pathway analyses using a metastasis-related gene panel on 12 lung tissues from patients with confirmed BM, 12 lung tissues from patients without BM, and 12 matched brain tissues from patients with confirmed BM during follow-up after LUAD surgery. The results of the gene expression analysis were validated by immunohistochemistry. RESULTS: Cell interaction-related pathways (such as focal adhesion, extracellular matrix-receptor interaction, and proteoglycans in cancer) showed the greatest differences among the three groups. Expression of the cell interaction-related pathway was highest in the lung sample of BM group and lowest in the matched brain tissue. Using a machine learning model, a signature of 20 genes from cell interaction-related pathways accurately predicted BM (area under the curve score of 0.792 and an accuracy rate of 0.875). Immunohistochemical analysis showed higher expression of proteins associated with cell interaction-related genes and a mesenchymal phenotype in the lung sample of BM group than in those without BM or matched brain tissue. CONCLUSIONS: LUAD acquires the characteristics of the cell interaction-related pathway that leads to the development of BM, with a significant decrease in expression following brain colonization.

The impact of CDCA5 expression on the immune microenvironment and its potential utility as a biomarker for PD-L1/PD-1 inhibitors in lung adenocarcinoma.

BACKGROUND: Studies have highlighted the important role of cell division cycle associated 5 (CDCA5) in tumor-associated immune dysfunction. We studied immune dysfunction based on CDCA5 expression in lung adenocarcinoma and investigated its potential as a biomarker for patients undergoing anti-programmed death protein-1/ programmed death ligand-1 (PD-1/PD-L1) inhibitor therapy. METHODS: We used the CIBERSORTx algorithm to investigate the immune cell distribution based on CDCA5 and explored its potential as a biomarker for PD-1/PD-L1 therapy using Tumor Immune Dysfunction and Exclusion in three lung adenocarcinoma datasets. Thus, we validated the role of CDCA5 as a biomarker in patients treated with PD-1/PD-L1 inhibitors. We also investigated the pathways through which CDCA5 regulates PD-L1 expression in a cell line. RESULTS: The high CDCA5 expression group showed elevated interferon gamma signature, CD274 expression, CD8+ T cell levels, tumor mutation burden, and microsatellite instability. Higher CDCA5 expression was associated with poorer prognosis in patients not treated with PD-1/PD-L1 inhibitors. However, in patients treated with PD-1/PD-L1 inhibitors, higher CDCA5 expression correlated with better response rates and prognosis. CDCA5 expression positively correlated with inhibitory immune checkpoint molecules. CDCA5 regulated the expression of PD-L1 through the ANXA/AKT pathway, and combined suppression of CDCA5 and PD-L1 synergistically inhibited cell proliferation. CONCLUSIONS: CDCA5 served as a promising biomarker for patients undergoing PD-L1/PD-1 inhibitor treatment, and co-inhibition of CDCA5 and PD-L1 could serve as an effective therapeutic strategy.

Prediction of Responsiveness to PD-L1/PD-1 Inhibitors Using miRNA Profiles Associated With PD-L1 Expression in Lung Adenocarcinoma and Squamous Cell Carcinoma.

BACKGROUND/AIM: MicroRNAs (miRNAs) regulate programmed cell death ligand 1 (PD-L1) and play a crucial role in tumor immune response. However, the relationship between miRNA expression patterns and PD-L1 remains unclear in both lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC). We investigated PD-L1-related miRNAs that can predict treatment response in patients treated with PD-L1/PD-1 inhibitors. PATIENTS AND METHODS: We selected miRNAs that were correlated with PD-L1 expression within the LUAD and LUSC datasets obtained from The Cancer Genome Atlas (TCGA) and Clinical Proteomic Tumor Analysis Consortium (CPTAC). We validated whether the miRNA profile could be used to predict the prognosis of patients treated with PD-L1/PD-1 inhibitors. RESULTS: Based on four public datasets, we selected 66 and 23 miRNAs associated with PD-L1 expression in LUAD and LUSC, respectively. From the above miRNAs, we identified 5 miRNAs in LUSC and 1 miRNA in LUAD that could predict the response to PD-L1/PD-1 inhibitors in a validation set of patients treated with PD-L1/PD-1 inhibitors. In LUSC, the miRNA profile exhibited a high predictive capability for the response to PD-L1/PD-1 treatment [area under the curve (AUC)=0.963] and accurately predicted prognosis (p=0.031). In LUAD, the miRNA profile was relatively less predictive than in LUSC (AUC=0.691 and p=0.213). Additionally, we observed variations in the PD-L1-associated miRNA profiles, as well as in the associated pathways, between LUAD and LUSC. CONCLUSION: The PD-L1-associated miRNA profile may predict treatment response in LUSC patients treated with PD-L1/PD-1 inhibitors and help select the PD-L1/PD-1 inhibitor treatment group.

Machine learning-driven prediction of brain metastasis in lung adenocarcinoma using miRNA profile and target gene pathway analysis of an mRNA dataset.

BACKGROUND: Brain metastasis (BM) is common in lung adenocarcinoma (LUAD) and has a poor prognosis, necessitating predictive biomarkers. MicroRNAs (MiRNAs) promote cancer cell growth, infiltration, and metastasis. However, the relationship between the miRNA expression profiles and BM occurrence in patients with LUAD remains unclear. METHODS: We conducted an analysis to identify miRNAs in tissue samples that exhibited different expression levels between patients with and without BM. Using a machine learning approach, we confirmed whether the miRNA profile could be a predictive tool for BM. We performed pathway analysis of miRNA target genes using a matched mRNA dataset. RESULTS: We selected 25 miRNAs that consistently exhibited differential expression between the two groups of 32 samples. The 25-miRNA profile demonstrated a strong predictive potential for BM in both Group 1 and Group 2 and the entire dataset (area under the curve [AUC] = 0.918, accuracy = 0.875 in Group 1; AUC = 0.867, accuracy = 0.781 in Group 2; and AUC = 0.908, accuracy = 0.875 in the entire group). Patients predicted to have BM, based on the 25-miRNA profile, had lower survival rates. Target gene analysis of miRNAs suggested that BM could be induced through the ErbB signaling pathway, proteoglycans in cancer, and the focal adhesion pathway. Furthermore, patients predicted to have BM based on the 25-miRNA profile exhibited higher expression of the epithelial-mesenchymal transition signature, TWIST, and vimentin than those not predicted to have BM. Specifically, there was a correlation between EGFR mRNA levels and BM. CONCLUSIONS: This 25-miRNA profile may serve as a biomarker for predicting BM in patients with LUAD.

Bimodal Transformer with Regional EEG Data for Accurate Gameplay Regularity Classification.

As games have been applied across various fields, including education and healthcare, numerous new games tailored to each field have emerged. Therefore, understanding user behavior has become crucial in securing the right players for each type of game. This study provides valuable insights for improving game development by measuring the electroencephalography (EEG) of game users and classifying the frequency of game usage. The multimodal mobile brain-body imaging (MOBI) dataset was employed for this study, and the frequency of game usage was categorized into "often" and "sometimes". To achieve decent classification accuracy, a novel bimodal Transformer architecture featuring dedicated channels for the frontal (AF) and temporal (TP) lobes is introduced, wherein convolutional layers, self-attention mechanisms, and cross-attention mechanisms are integrated into a unified model. The model, designed to differentiate between AF and TP channels, exhibits functional differences between brain regions, allowing for a detailed analysis of inter-channel correlations. Evaluated through five-fold cross-validation (CV) and leave-one-subject-out cross-validation (LOSO CV), the proposed model demonstrates classification accuracies of 88.86% and 85.11%, respectively. By effectively classifying gameplay frequency, this methodology provides valuable insights for targeted game participation and contributes to strategic efforts to develop and design customized games for player acquisition.

Characteristics of the immune microenvironment associated with RRM2 expression and its application to PD-L1/PD-1 inhibitors in lung adenocarcinoma.

Recent studies have indicated that RRM2 plays a crucial part in the tumor immune microenvironment. According to the expression of RRM2, we evaluated immune cell infiltration, immunotherapy biomarkers, and the expression of immune checkpoint molecules in four lung adenocarcinoma (LUAD) datasets. We employed the Tumor Immune Dysfunction and Exclusion (TIDE) and CIBERSORTx algorithms to examine the patterns of immune cell distribution and evaluate the responses to anti-programmed death protein-1/programmed death ligand-1 (PD-1/PD-L1) therapy in three publicly available LUAD datasets. These findings were corroborated using a validation group comprising patients who received treatment with PD-1/PD-L1 inhibitors. Additionally, we conducted experiments using LUAD cell lines to investigate how RRM2 affects the expression of PD-L1. In comparison to the low RRM2 group, the high RRM2 group exhibited a high interferon gamma signature, high T-cell-inflamed signature, high CD274 expression, high CD8+ T cell levels, low cancer-associated fibroblasts, and low M2 macrophages, according to TIDE analysis in the three LUAD datasets. Analysis of the three LUAD datasets using CIBERSORTx confirmed a positive correlation between RRM2 and CD8+ T cells, and this finding was validated by immunohistochemistry in a separate validation set. In the three LUAD datasets without PD-1/PD-L1 inhibitor treatment, higher RRM2 expression was associated with a poorer prognosis. However, in the LUAD dataset treated with PD-1/PD-L1 inhibitors, higher RRM2 expression was associated with better prognosis. In the three datasets, the high-RRM2 group exhibited higher expression of inhibitory immune checkpoint molecules. In a LUAD cell line study, we discovered that RRM2 regulates PD-L1 expression through the ANXA1/AKT pathway. The expression of RRM2 shows promise as a predictive biomarker for PD-1/PD-L1 inhibitors in LUAD patients, and it may represent a new target to overcome resistance to PD-L1/PD-1 therapies.

Correlated insulator collapse due to quantum avalanche via in-gap ladder states.

The significant discrepancy observed between the predicted and experimental switching fields in correlated insulators under a DC electric field far-from-equilibrium necessitates a reevaluation of current microscopic understanding. Here we show that an electron avalanche can occur in the bulk limit of such insulators at arbitrarily small electric field by introducing a generic model of electrons coupled to an inelastic medium of phonons. The quantum avalanche arises by the generation of a ladder of in-gap states, created by a multi-phonon emission process. Hot-phonons in the avalanche trigger a premature and partial collapse of the correlated gap. The phonon spectrum dictates the existence of two-stage versus single-stage switching events which we associate with charge-density-wave and Mott resistive phase transitions, respectively. The behavior of electron and phonon temperatures, as well as the temperature dependence of the threshold fields, demonstrates how a crossover between the thermal and quantum switching scenarios emerges within a unified framework of the quantum avalanche.

Immune profiles according to EGFR mutant subtypes and correlation with PD-1/PD-L1 inhibitor therapies in lung adenocarcinoma.

Background: We examined the distributions of 22 immune cell types and the responses to PD-1/PD-L1 inhibitors according to EGFR mutation profile, in three independent datasets of lung adenocarcinoma (LUAD). Methods: We used CIBERSORTx to analyze the distributions of immune cells, and tumor immune dysfunction and exclusion (TIDE) or tumor mutation burden (TMB) to analyze responses to anti-PD-1/PD-L1 therapy, in two public LUAD datasets. The results were verified with a validation set that included patients treated with PD-1/PD-L1 inhibitors. Results: Compared to EGFR mutants, EGFR wild-type carcinomas had higher numbers of CD8+ T cells, CD4 memory activated T cells and neutrophils, and lower numbers of resting dendritic cells and resting mast cells, in two of the datasets. In our subgroup analyses, CD8+ T cells and CD4 memory activated T cells were more numerous in EGFR rare variants than in wild-types, L858R mutants, and exon 19 deletion mutants. In our TIDE or TMB analyses, EGFR rare variants were predicted to respond better to PD-1/PD-L1 inhibitors than wild-types, L858R mutants, and exon 19 deletion mutants. In the validation set verified by immunohistochemical staining, levels of CD8+ T cells in the EGFR rare variant or wild-type groups were significantly higher than in the EGFR L858R and exon 19 deletion groups. In patients treated with PD-1/PD-L1 inhibitors, the survival rates of patients with EGFR wild-type and rare mutant carcinomas were higher than those with L858R and exon 19 deletion carcinomas. Conclusion: The EGFR rare mutation form of LUAD shows a higher immune activation state compared to wild-type, L858R, and exon 19 deletion variants, indicating it as a potential target for PD-1/PD-L1 inhibitor therapy.

Comparison of histological and molecular features of pediatric-type follicular lymphoma and pediatric nodal marginal zone lymphoma.

Pediatric-type follicular lymphoma and pediatric nodal marginal zone lymphoma are pediatric B cell lymphomas with similar clinical characteristics but distinct histological features. We investigated the differences between pediatric-type follicular lymphoma and pediatric nodal marginal zone lymphoma by comparing their histological and molecular characteristics. A total of 5 pediatric-type follicular lymphoma and 11 pediatric nodal marginal zone lymphoma patients were included in the study. In the histological review, 5 of the 16 cases showed overlapping morphological features of pediatric-type follicular lymphoma and pediatric nodal marginal zone lymphoma; hence, they were reclassified as "mixed type." In molecular analysis, using panel-based massively parallel sequencing, MAP2K1, TNFRSF14, and IRF8 mutations were found in 6, 3, and 2 of the 11 pediatric nodal marginal zone lymphoma patients, respectively, and IRF8 mutation was found in one of the five pediatric-type follicular lymphoma patients. There were no significant differences in genetic alterations established from the histologically reclassified diagnosis as well as the initial diagnosis. Pediatric-type follicular lymphoma and pediatric nodal marginal zone lymphoma showed morphological overlap in some cases, and no difference between the two was found upon molecular analysis. These findings suggest the possibility that pediatric-type follicular lymphoma and pediatric nodal marginal zone lymphoma are single entity pediatric B-cell lymphoma with broad morphological spectrum.

Antiproliferative Activity of Gibbosic Acid H through Induction of G0/G1 Cell Cycle Arrest and Apoptosis in Human Lung Cancer Cells.

Lung cancer is one of the most common causative cancers worldwide. Particularly, non-small cell lung cancer (NSCLC) accounts for approximately 85% of all lung cancer cases. NSCLC is a serious form of lung cancer that requires prompt diagnosis, and the 5-year survival rate for patients with this disease is only 24%. Gibbosic acid H (GaH), a natural lanostanoid obtained from the Ganoderma species (Ganodermataceae), has antiproliferative activities against colon and lung cancer cells. The aim of the present study was to evaluate the antiproliferative activity of GaH in NSCLC cells and to elucidate the underlying molecular mechanisms. GaH was found to induce G0/G1 cell cycle arrest and autophagy by activating adenosine monophosphate-activated protein kinase in A549 and H1299 cells. The induction of this cell cycle arrest was associated with the downregulation of cyclin E1 and CDK2. Additionally, the induction of autophagy by GaH was correlated with the upregulation of LC3B, beclin-1, and p53 expression. GaH also induced apoptosis by upregulating cleaved caspase-3 and Bax in the lung cancer cells. These findings suggest that GaH has a potential in the growth inhibition of human lung cancer cells.

Prognostic and predictive value of YTHDF1 and YTHDF2 and their correlation with tumor-infiltrating immune cells in non-small cell carcinoma.

Background: YTH domain-containing family protein 1 (YTHDF1) or YTHDF2 play crucial roles in cancer immunotherapy. We examine the expression of YTHDF1, YTHDF2, CD8, CD4, and FOXP3 to identify their prognostic or predictive role for PD-1/PD-L1 inhibitor in non-small cell lung cancer (NSCLC). Methods: Immunohistochemical expression of YTHDF1, YTHDF2, CD8, CD4, and FOXP3 was investigated in 266 patients not receiving PD-1/PD-L1 inhibitors and in 59 patients receiving PD-1/PD-L1 inhibitors. Immunohistochemical results were verified using mRNA dataset obtained from The Cancer Genome Atlas (TCGA) database. Results: Immunohistochemical expression of YTHDF1 or YTHDF2 was negatively associated with CD8- and CD4-positive T cells; however, the same expression was positively associated with FOXP3-positive T cells. YTHDF1 or YTHDF2 mRNA expression was also negatively associated with CD8- and CD4-positive T cells. Gene set enrichment analysis revealed that low YTHDF1 was related to immune hot tumor gene sets. Expression of YTHDF1 or YTHDF2 was negatively associated with expression of most immune checkpoints. YTHDF1 and YTHDF2 were predictive markers of response to PD-1/PD-L1 inhibitors. YTHDF1 or YTHDF2 expression was associated with better prognosis. YTHDF1 has an immune hot profile in both cell types, whereas YTHDF2 is only seen in adenocarcinoma. Conclusion: Low YTHDF1 or YTHDF2 reflects an immune hot tumor signature and may serve as a predictor or prognostic marker.

Artificial Intelligence in Eye Disease: Recent Developments, Applications, and Surveys.

Artificial intelligence (AI) has expanded by finding applications in medical diagnosis for clinical support systems [...].

Epoxinnamide: An Epoxy Cinnamoyl-Containing Nonribosomal Peptide from an Intertidal Mudflat-Derived Streptomyces sp.

Cinnamoyl-containing nonribosomal peptides (CCNPs) form a unique family of actinobacterial secondary metabolites and display various biological activities. A new CCNP named epoxinnamide (1) was discovered from intertidal mudflat-derived Streptomyces sp. OID44. The structure of 1 was determined by the analysis of one-dimensional (1D) and two-dimensional (2D) nuclear magnetic resonance (NMR) data along with a mass spectrum. The absolute configuration of 1 was assigned by the combination of advanced Marfey's method, 3JHH and rotating-frame overhauser effect spectroscopy (ROESY) analysis, DP4 calculation, and genomic analysis. The putative biosynthetic pathway of epoxinnamide (1) was identified through the whole-genome sequencing of Streptomyces sp. OID44. In particular, the thioesterase domain in the nonribosomal peptide synthetase (NRPS) biosynthetic gene cluster was proposed as a bifunctional enzyme, which catalyzes both epimerization and macrocyclization. Epoxinnamide (1) induced quinone reductase (QR) activity in murine Hepa-1c1c7 cells by 1.6-fold at 5 µM. It also exhibited effective antiangiogenesis activity in human umbilical vein endothelial cells (IC50 = 13.4 µM).

Nyuzenamide C, an Antiangiogenic Epoxy Cinnamic Acid-Containing Bicyclic Peptide from a Riverine Streptomyces sp.

A new nonribosomal peptide, nyuzenamide C (1), was discovered from riverine sediment-derived Streptomyces sp. DM14. Comprehensive analysis of the spectroscopic data of nyuzenamide C (1) revealed that 1 has a bicyclic backbone composed of six common amino acid residues (Asn, Leu, Pro, Gly, Val, and Thr) and four nonproteinogenic amino acid units, including hydroxyglycine, ß-hydroxyphenylalanine, p-hydroxyphenylglycine, and 3,ß-dihydroxytyrosine, along with 1,2-epoxypropyl cinnamic acid. The absolute configuration of 1 was proposed by J-based configuration analysis, the advanced Marfey's method, quantum mechanics-based DP4 calculations, and bioinformatic analysis of its nonribosomal peptide synthetase biosynthetic gene cluster. Nyuzenamide C (1) displayed antiangiogenic activity in human umbilical vein endothelial cells and induced quinone reductase in murine Hepa-1c1c7 cells.

Underwater versus conventional cold snare polypectomy of colorectal polyps 4-9 mm in diameter: a prospective randomized controlled trial.

BACKGROUND AND AIMS: Cold snare polypectomy (CSP) has been reported as safe and effective method for the removal of small colorectal polyps. However, some studies showed low R0 resection rate. Underwater endoscopic mucosal resection is an effective technique to increase the proportions of complete resection. Therefore, the aim was to compare the rate of R0 resection of colorectal polyps 4-9 mm in diameter between conventional CSP (C-CSP) and underwater CSP (U-CSP). METHODS: This study was a prospective randomized controlled trial. A total of 198 polyps (4-9 mm) in 110 patients were enrolled between December 2019 and June 2020. The polyps were randomized to be treated with either C-CSP (100 polyps) or U-CSP (98 polyps). RESULTS: The R0 resection rate was significantly higher in U-CSP group than in C-CSP groups (84.7% vs. 59.0%; p < 0.001). The polyp retrieval rate of C-CSP and U-CSP was 94.5% and 100% (p = 0.030). The rate of polyp fragmentation of C-CSP and U-CSP group was 5.3% and 0% (p = 0.027). The resection time and retrieval time were longer in C-CSP than U-CSP (45.0 ± 37.7 s vs. 34.1 ± 21.2 s, p = 0.032 and 51.9 ± 67.7 s vs. 12.7 ± 12.4 s, p < 0.001). No clinically significant bleeding or perforation occurred in either group. CONCLUSIONS: The results of this study were excellent with U-CSP of 4-9 mm colorectal polyps in terms of R0 resection, polyp retrieval and fragmentation rate, and procedure/retrieval time. Therefore, U-CSP is a safe and effective technique for removing colorectal polyps 4-9 mm in diameter. KCT (0004530).

Review of Machine Learning Applications Using Retinal Fundus Images.

Automating screening and diagnosis in the medical field saves time and reduces the chances of misdiagnosis while saving on labor and cost for physicians. With the feasibility and development of deep learning methods, machines are now able to interpret complex features in medical data, which leads to rapid advancements in automation. Such efforts have been made in ophthalmology to analyze retinal images and build frameworks based on analysis for the identification of retinopathy and the assessment of its severity. This paper reviews recent state-of-the-art works utilizing the color fundus image taken from one of the imaging modalities used in ophthalmology. Specifically, the deep learning methods of automated screening and diagnosis for diabetic retinopathy (DR), age-related macular degeneration (AMD), and glaucoma are investigated. In addition, the machine learning techniques applied to the retinal vasculature extraction from the fundus image are covered. The challenges in developing these systems are also discussed.