Prognostic potentials of AI in ophthalmology: systemic disease forecasting via retinal imaging.

BACKGROUND: Artificial intelligence (AI) that utilizes deep learning (DL) has potential for systemic disease prediction using retinal imaging. The retina's unique features enable non-invasive visualization of the central nervous system and microvascular circulation, aiding early detection and personalized treatment plans for personalized care. This review explores the value of retinal assessment, AI-based retinal biomarkers, and the importance of longitudinal prediction models in personalized care. MAIN TEXT: This narrative review extensively surveys the literature for relevant studies in PubMed and Google Scholar, investigating the application of AI-based retina biomarkers in predicting systemic diseases using retinal fundus photography. The study settings, sample sizes, utilized AI models and corresponding results were extracted and analysed. This review highlights the substantial potential of AI-based retinal biomarkers in predicting neurodegenerative, cardiovascular, and chronic kidney diseases. Notably, DL algorithms have demonstrated effectiveness in identifying retinal image features associated with cognitive decline, dementia, Parkinson's disease, and cardiovascular risk factors. Furthermore, longitudinal prediction models leveraging retinal images have shown potential in continuous disease risk assessment and early detection. AI-based retinal biomarkers are non-invasive, accurate, and efficient for disease forecasting and personalized care. CONCLUSION: AI-based retinal imaging hold promise in transforming primary care and systemic disease management. Together, the retina's unique features and the power of AI enable early detection, risk stratification, and help revolutionizing disease management plans. However, to fully realize the potential of AI in this domain, further research and validation in real-world settings are essential.

Effect of vascular endothelial growth factor 165 on dopamine level in the retinas of guinea pigs with form-deprivation myopia.

Background: Myopia is the most common refractive error because excessive increase in the axial length of a myopic eye leads to the thinning of the posterior scleral pole and can cause serious complications resulting in blindness. Thus, myopia has become a great concern worldwide. Dopamine (DA) plays a role in the development of myopia. Moreover, in Parkinson's disease, it has been proved that vascular endothelial growth factor 165 (VEGF165) can promote the survival and recovery of DA neurons, resulting in increased DA secretion in the striatum, thereby treating neuropathy. Therefore, we speculate that VEGF165 can also promote the release of DA in the retina to inhibit the occurrence and development of myopia. We aimed to investigate the effect of VEGF165 on DA levels in the retinas of guinea pigs with form-deprivation myopia (FDM) and the effects of DA on myopia prevention and control. Methods: Healthy 3-week-old pigmented guinea pigs were randomly divided into blank, FDM, phosphate buffer saline (PBS), 1, 5, and 10 ng groups. The FDM model was established by covering the right eye continuously with a translucent latex balloon pullover for 14 days. The pigs in the PBS, 1, 5, and 10 ng groups were injected with PBS buffer and 1, 5, and 10 ng of VEGF165 recombinant human protein, respectively, in the vitreous of the right eye before masking. The refractive error and axial length were measured before and after modeling. All retinas were used for biomolecular analyses after 14 days. Results: We found that the intravitreal injection of VEGF165 elevated DA levels in the retina and was effective in slowing the progression of myopia, and 1 ng of VEGF165 was the most effective. Moreover, the number of vascular endothelial cell nuclei in the 1 ng group was lower than that in the other VEGF165 groups. Conclusions: Our data suggest that VEGF165 has a promoting effect on DA in the retinas of guinea pigs with FDM, potentially controlling the development of myopia.

Validation of a deep-learning-based retinal biomarker (Reti-CVD) in the prediction of cardiovascular disease: data from UK Biobank.

BACKGROUND: Currently in the United Kingdom, cardiovascular disease (CVD) risk assessment is based on the QRISK3 score, in which 10% 10-year CVD risk indicates clinical intervention. However, this benchmark has limited efficacy in clinical practice and the need for a more simple, non-invasive risk stratification tool is necessary. Retinal photography is becoming increasingly acceptable as a non-invasive imaging tool for CVD. Previously, we developed a novel CVD risk stratification system based on retinal photographs predicting future CVD risk. This study aims to further validate our biomarker, Reti-CVD, (1) to detect risk group of ≥ 10% in 10-year CVD risk and (2) enhance risk assessment in individuals with QRISK3 of 7.5-10% (termed as borderline-QRISK3 group) using the UK Biobank. METHODS: Reti-CVD scores were calculated and stratified into three risk groups based on optimized cut-off values from the UK Biobank. We used Cox proportional-hazards models to evaluate the ability of Reti-CVD to predict CVD events in the general population. C-statistics was used to assess the prognostic value of adding Reti-CVD to QRISK3 in borderline-QRISK3 group and three vulnerable subgroups. RESULTS: Among 48,260 participants with no history of CVD, 6.3% had CVD events during the 11-year follow-up. Reti-CVD was associated with an increased risk of CVD (adjusted hazard ratio [HR] 1.41; 95% confidence interval [CI], 1.30-1.52) with a 13.1% (95% CI, 11.7-14.6%) 10-year CVD risk in Reti-CVD-high-risk group. The 10-year CVD risk of the borderline-QRISK3 group was greater than 10% in Reti-CVD-high-risk group (11.5% in non-statin cohort [n = 45,473], 11.5% in stage 1 hypertension cohort [n = 11,966], and 14.2% in middle-aged cohort [n = 38,941]). C statistics increased by 0.014 (0.010-0.017) in non-statin cohort, 0.013 (0.007-0.019) in stage 1 hypertension cohort, and 0.023 (0.018-0.029) in middle-aged cohort for CVD event prediction after adding Reti-CVD to QRISK3. CONCLUSIONS: Reti-CVD has the potential to identify individuals with ≥ 10% 10-year CVD risk who are likely to benefit from earlier preventative CVD interventions. For borderline-QRISK3 individuals with 10-year CVD risk between 7.5 and 10%, Reti-CVD could be used as a risk enhancer tool to help improve discernment accuracy, especially in adult groups that may be pre-disposed to CVD.

Detection of Systemic Diseases From Ocular Images Using Artificial Intelligence: A Systematic Review.

PURPOSE: Despite the huge investment in health care, there is still a lack of precise and easily accessible screening systems. With proven associations to many systemic diseases, the eye could potentially provide a credible perspective as a novel screening tool. This systematic review aims to summarize the current applications of ocular image-based artificial intelligence on the detection of systemic diseases and suggest future trends for systemic disease screening. METHODS: A systematic search was conducted on September 1, 2021, using 3 databases-PubMed, Google Scholar, and Web of Science library. Date restrictions were not imposed and search terms covering ocular images, systemic diseases, and artificial intelligence aspects were used. RESULTS: Thirty-three papers were included in this systematic review. A spectrum of target diseases was observed, and this included but was not limited to cardio-cerebrovascular diseases, central nervous system diseases, renal dysfunctions, and hepatological diseases. Additionally, one- third of the papers included risk factor predictions for the respective systemic diseases. CONCLUSIONS: Ocular image - based artificial intelligence possesses potential diagnostic power to screen various systemic diseases and has also demonstrated the ability to detect Alzheimer and chronic kidney diseases at early stages. Further research is needed to validate these models for real-world implementation.

Clinical Relevance of Body Fluid Volume Status in Diabetic Patients With Macular Edema.

Objective: To investigate body fluid status in diabetic macular edema (DME) patients and the extent to which it is affected by renal function. Methods: One hundred and thirty-two eyes from 132 patients with diabetes mellitus (DM) were prospectively collected in this cross-sectional, observational study. Thirty-five were DM patients without diabetic retinopathy (DR), 31 were DR patients without DME, and 66 were DME patients. The fluid status of each participant was quantified with extracellular water-to-total body water ratio (ECW/TBW) using a body composition monitor. Central subfield thickness (CST) and macular volume (MV) were obtained using optical coherence tomography (OCT). Urine albumin-to-creatinine ratio (UACR), estimated glomerular filtration rate (eGFR), and albumin was obtained using serum and urine laboratory data. Results: ECW/TBW was significantly increased in DME patients (39.2 ± 0.9, %) compared to DM (38.1 ± 0.7, %, P = 0.003) and DR patients without DME (38.7 ± 0.9, %, P < 0.001). In multilinear regression, fluid overload was positively related to DME and UACR (DME vs. DM: ß = 2.418, P < 0.001; DME vs. DR: ß = 1.641, P = 0.001; UACR, per 102, ß = 1.017, P = 0.01). In the binary logistic regression for DME risk, the area under the receiver operating characteristic curve (AUROC) increased significantly by adding ECW/TBW along with UACR and age (AUC: 0.826 vs. 0.768). Conclusion: DME patients had elevated body fluid volume independent of kidney functions. The assessment of extracellular fluid status may help in the management of DME.

Development and Validation of Retinal Vasculature Nomogram in Suspected Angina Due to Coronary Artery Disease.

AIMS: To develop and validate a nomogram using retinal vasculature features and clinical variables to predict coronary artery disease (CAD) in patients with suspected angina. METHODS: The prediction model consisting of 795 participants was developed in a training set of 508 participants with suspected angina due to CAD, and data were collected from January 2018 to June 2019. The held-out validation was conducted with 287 consecutive patients from July 2019 to November 2019. All patients with suspected CAD received optical coherence tomography angiography (OCTA) examination before undergoing coronary CT angiography. LASSO regression model was used for data reduction and feature selection. Multivariable logistic regression analysis was used to develop the retinal vasculature model for predicting the probability of the presence of CAD. RESULTS: Three potential OCTA parameters including vessel density of the nasal and temporal perifovea in the superficial capillary plexus and vessel density of the inferior parafovea in the deep capillary plexus were further selected as independent retinal vasculature predictors. Model clinical electrocardiogram (ECG) OCTA (clinical variables＋ECG＋OCTA) was presented as the individual prediction nomogram, with good discrimination (AUC of 0.942 [95% CI, 0.923-0.961] and 0.897 [95% CI, 0.861-0.933] in the training and held-out validation sets, respectively) and good calibration. Decision curve analysis indicated the clinical applicability of this retinal vasculature nomogram. CONCLUSIONS: The presented retinal vasculature nomogram based on individual probability can accurately identify the presence of CAD, which could improve patient selection and diagnostic yield of aggressive testing before determining a diagnosis.

Retinal microvasculature impairments in patients with coronary artery disease: An optical coherence tomography angiography study.

PURPOSE: To investigate the association between retinal microvasculature and the presence and severity of coronary artery disease (CAD) using optical coherence tomography angiography (OCTA). METHODS: The cross-sectional study was conducted in Guangdong Provincial People's Hospital, China. Retinal microvasculature parameters were measured by OCTA of the optic disc, including the vessel density (VD) and retinal nerve fibre thickness of the radial peripapillary capillary. In terms of the entire macula, VD of the superficial capillary plexus (SCP), deep capillary plexus (DCP) and foveal density (FD-300) were included. The Gensini score was used to evaluate the severity of coronary artery obstructive lesions in CAD patients. RESULTS: A total of 410 participants (270 CAD patients and 140 controls) were included. Overall, participants showed significantly greater odds of having CAD in the lower versus higher VD for mean SCP, OR = 2.33 (95% CI 1.49-3.65); in the parafoveal SCP, OR = 2.68 (95% CI 1.70-4.23); and in the perifoveal SCP, OR = 2.36 (95% CI 1.49-3.72). Additionally, participants showed significantly greater odds of having CAD in the lower versus higher VD for mean DCP, OR = 4.04 (95% CI 2.53-6.45); in the parafoveal DCP, OR = 4.08 (95% CI 2.54-6.55); and in the perifoveal DCP, OR = 3.88 (95% CI 2.43-6.19). Among CAD patients, lower VD of DCP was associated with significantly greater adjusted Gensini scores (p = 0.004 for mean DCP; p = 0.035 for parafoveal DCP; p = 0.006 for perifoveal DCP). CONCLUSIONS: SCP and DCP were found to be associated with the presence of CAD among the whole population, while DCP was found to be associated with Gensini scores in CAD patients. Retinal microvasculature was associated with the presence and severity of coronary artery stenosis in CAD patients.

Improved Retinal Microcirculation After Cardiac Surgery in Patients With Congenital Heart Disease.

Background: Microcirculatory changes in congenital heart disease (CHD) patients undergoing cardiac surgery are not fully understood. We aimed to investigate the changes of retinal microcirculation in CHD patients after cardiac surgery by optical coherence tomography angiography (OCTA) and explore the association between retinal microcirculation and surgical outcome. Methods: This prospective observational study consisted of 71 CHD patients aged ≥6 years undergoing cardiac surgery including 19 cyanotic CHD (CCHD) and 52 acyanotic CHD (ACHD). Optical coherence tomography angiography (OCTA) was used to measure vessel density (VD) and capillary density (CD) of radial peripapillary capillary (RPC) and peripapillary, VD of superficial capillary plexus (SCP) and deep capillary plexus (DCP), thickness of retinal nerve fiber layer (RNFL) and ganglion cell complex (GCC) preoperatively and 1 month postoperatively. Transthoracic echocardiography was conducted to measure macrocirculation. Results: In CCHD patients, VD and CD of RPC and peripapillary increased postoperatively (all P < 0.05). In ACHD patients, VD of peripapillary, CD of RPC and peripapillary, and RNFL thickness increased postoperatively (all P < 0.05). VD of SCP and DCP, and GCC thickness did not change significantly in CHD patients after surgery. Lower preoperative retinal microvascular density was associated with longer cardiopulmonary bypass (CPB) time and postoperative length of stay (PLOS). No correlation was found between microcirculatory and macrohemodynamic parameters (all P > 0.05). Conclusions: Improved retinal microcirculation was observed after congenital cardiac surgery and impaired preoperative retinal microvasculature was associated with prolonged CPB time and PLOS, which might provide potential information about the outcome of congenital cardiac surgery.

Epigenetic inhibition assisted chemotherapeutic treatment of lung cancer based on artificial exosomes.

We adopted a novel strategy by combining histone deacetylase (HDAC) inhibitors with traditional chemotherapeutics to treat solid tumors. However, chemotherapeutics often have a narrow therapeutic index and need multiple administrations with undesired side effects that lead to the intolerance. To reduce the non-specificity of chemotherapeutics, targeted therapy was introduced to restrict such agents in the tumor with minimum effects on other tissues. We developed bioinspired artificial exosomes (AE), which enabled to deliver chemotherapeutics to the tumors effectively after systemic administration. AE were produced by incorporating membrane proteins from cancer cells into phospholipid liposomes that mimicked the plasma membrane. The synthesized AE were used for the delivery of broad-spectrum chemotherapeutic doxorubicin (DOX) and vorinostat (SAHA), an epigenetic inhibitor. The combination of DOX and SAHA showed synergistic effects on suppressing non-small cell lung cancer cells and xenograft tumors without apparent adverse effects. AE facilitated the delivery of drugs to tumor tissue and extended the retention time of drugs within tumors. Taken together, these studies suggest that the bioengineered artificial exosomes may serve as novel delivery strategy for chemotherapeutics to treat non-small cell lung cancer.

Hybrid artificial cell-mediated epigenetic inhibition in metastatic lung cancer.

HYPOTHESIS: Histone deacetylase inhibitors (HDACIs), such as vorinostat (suberoylanilide hydroxamic acid, SAHA), has become a promising approach for the treatment of metastatic lung cancer. However, HDACIs usually showed a short circulation lifetime, low specificity, and low bioavailability, which limited their therapeutic effect in this field. We supposed that the use of biomimetic nanoparticles enabled to overcome the disadvantages of HDACIs, and improved the inhibition of metastatic lung cancer. EXPERIMENTS: SAHA was encapsulated into a pH-sensitive core constructed with Poly(lactic-co-glycolic acid) (PLAG) and 1,2-dioleoyloxy-3-(trimethylammonium) propane (DOTAP), followed by the camouflage with hybrid membranes derived from red blood cells and metastatic NCI-H1299 lung cancer cells (HRPDS). The physical and chemical properties were characterized with Transmission electron microscope (TEM), Size & Zeta potential analyzer. The cellular uptake was analyzed with Confocal laser scanning microscope (CLSM) and Flow cytometry (FACS). The biological effect analysis was performed with Western blotting (WB), RNA-Sequencing (RNA-Seq), and ChIP-Sequencing (ChIP-Seq). FINDINGS: HRPDS exhibited enhanced circulation lifetime in vivo and homotypic targeting to metastatic cells in the metastatic foci, which induced significant suppression of lung cancer liver metastasis. Our work opens a new avenue for the treatment of metastatic lung cancer by epigenetic inhibition based on this style of biomimetic nanovehicle.

Automatic prediction of treatment outcomes in patients with diabetic macular edema using ensemble machine learning.

BACKGROUND: This study aimed to predict the treatment outcomes in patients with diabetic macular edema (DME) after 3 monthly anti-vascular endothelial growth factor (VEGF) injections using machine learning (ML) based on pretreatment optical coherence tomography (OCT) images and clinical variables. METHODS: An ensemble ML system consisting of four deep learning (DL) models and five classical machine learning (CML) models was developed to predict the posttreatment central foveal thickness (CFT) and the best-corrected visual acuity (BCVA). A total of 363 OCT images and 7,587 clinical data records from 363 eyes were included in the training set (304 eyes) and external validation set (59 eyes). The DL models were trained using the OCT images, and the CML models were trained using the OCT images features and clinical variables. The predictive posttreatment CFT and BCVA values were compared with true outcomes obtained from the medical records. RESULTS: For CFT prediction, the mean absolute error (MAE), root mean square error (RMSE), and R2 of the best-performing model in the training set was 66.59, 93.73, and 0.71, respectively, with an area under receiver operating characteristic curve (AUC) of 0.90 for distinguishing the eyes with good anatomical response. The MAE, RMSE, and R2 was 68.08, 97.63, and 0.74, respectively, with an AUC of 0.94 in the external validation set. For BCVA prediction, the MAE, RMSE, and R2 of the best-performing model in the training set was 0.19, 0.29, and 0.60, respectively, with an AUC of 0.80 for distinguishing eyes with a good functional response. The external validation achieved a MAE, RMSE, and R2 of 0.13, 0.20, and 0.68, respectively, with an AUC of 0.81. CONCLUSIONS: Our ensemble ML system accurately predicted posttreatment CFT and BCVA after anti-VEGF injections in DME patients, and can be used to prospectively assess the efficacy of anti-VEGF therapy in DME patients.

Retinal Microvasculature Changes in Patients With Coronary Total Occlusion on Optical Coherence Tomography Angiography.

Background: Retinal microvasculature has been associated with coronary artery disease (CAD), but the exact contributory role in coronary total occlusion (CTO) is unclear. We aimed to investigate whether retinal vasculature is associated with CTO and could provide incremental value in the assessment of CTO. Methods: A total of 218 CAD patients including 102 CTO and 116 non-CTO were enrolled. Retinal vasculature was measured by optical coherence tomography angiography (OCTA) for all patients. Receiver operating characteristic (ROC) curve was used to assess the performance of retinal vasculature in differentiating CTO from non-CTO patients. Results: In non-CTO CAD patients, vessel density (VD) of mean superficial capillary plexus (SCP) and parafovea SCP were 49.85 and 52.56%, respectively; in CTO patients, VD of mean SCP and parafovea SCP were 47.77, and 49.58%, respectively. After multiple adjustment, VD in the SCP was significantly lower in CTO patients compared to non-CTO patients. VD of superior hemi in the parafovea SCP combined with the clinical variates showed the best ability to predict CTO from CAD with an area under the curve (AUC) of 0.812 (specificity of 89.0% and sensitivity of 65.9%). Conclusions: In CTO patients, retinal VD was significantly decreased, and microvascular damage might specifically target to arterioles than capillaries. Retinal vasculature could thus be a surrogate for detecting the microvascular damage and assist in the assessment of CTO patients. OCTA examination could be suggested to monitor the process of coronary arteries lesions.

DETECTION OF MORPHOLOGIC PATTERNS OF DIABETIC MACULAR EDEMA USING A DEEP LEARNING APPROACH BASED ON OPTICAL COHERENCE TOMOGRAPHY IMAGES.

PURPOSE: To develop a deep learning (DL) model to detect morphologic patterns of diabetic macular edema (DME) based on optical coherence tomography (OCT) images. METHODS: In the training set, 12,365 OCT images were extracted from a public data set and an ophthalmic center. A total of 656 OCT images were extracted from another ophthalmic center for external validation. The presence or absence of three OCT patterns of DME, including diffused retinal thickening, cystoid macular edema, and serous retinal detachment, was labeled with 1 or 0, respectively. A DL model was trained to detect three OCT patterns of DME. The occlusion test was applied for the visualization of the DL model. RESULTS: Applying 5-fold cross-validation method in internal validation, the area under the receiver operating characteristic curve for the detection of three OCT patterns (i.e., diffused retinal thickening, cystoid macular edema, and serous retinal detachment) was 0.971, 0.974, and 0.994, respectively, with an accuracy of 93.0%, 95.1%, and 98.8%, respectively, a sensitivity of 93.5%, 94.5%, and 96.7%, respectively, and a specificity of 92.3%, 95.6%, and 99.3%, respectively. In external validation, the area under the receiver operating characteristic curve was 0.970, 0.997, and 0.997, respectively, with an accuracy of 90.2%, 95.4%, and 95.9%, respectively, a sensitivity of 80.1%, 93.4%, and 94.9%, respectively, and a specificity of 97.6%, 97.2%, and 96.5%, respectively. The occlusion test showed that the DL model could successfully identify the pathologic regions most critical for detection. CONCLUSION: Our DL model demonstrated high accuracy and transparency in the detection of OCT patterns of DME. These results emphasized the potential of artificial intelligence in assisting clinical decision-making processes in patients with DME.

High-Performance Dual Combination Therapy for Cancer Treatment with Hybrid Membrane-Camouflaged Mesoporous Silica Gold Nanorods.

Conventional chemotherapy usually induces significant side effects due to its inability to discriminate between cancer and normal cells. Moreover, the efficacy of cancer elimination is still unsatisfied. Here, we fabricated a nanocomposite enabling high-performance dual combination therapy (chemo/photothermal therapy). This style of novel nanocomposites was constructed with doxorubicin (DOX)-loaded mesoporous silica gold (MSG) nanorods, which were further camouflaged with hybrid membranes derived from HeLa cells and red blood cells (HRMSGD). The hybrid membrane-camouflaged structure showed enhanced circulation lifetime and cell line-specific delivery of chemotherapeutics both in vitro and in vivo. The dual combination therapy by HRMSGD showed an unattainable therapeutic effect, compared with a single treatment, and inhibited tumor growth significantly. Furthermore, the nanoplatforms were photoacoustic-responsive, which showed real-time and noninvasive tracking capability. The present study established nanoplatforms with hybrid cell membrane-camouflaged multifunctional gold nanorods, which realized the combination of homotypic targeting, noninvasive tracking, chemotherapy, and photothermal therapy. To the best of our knowledge, this is the first study to use a natural membrane to camouflage mesoporous silica-modified gold nanorods, which opened a new avenue for cancer treatment.

Retinal microvasculature impairment in patients with congenital heart disease investigated by optical coherence tomography angiography.

IMPORTANCE: A high prevalence of retinal abnormalities have been reported in congenital heart disease (CHD), but quantitative analysis of retinal vasculature is scarce. Optical coherence tomography angiography (OCTA) is a noninvasive method to quantitatively assess the retinal microvasculature. BACKGROUND: To investigate the retinal microvasculature changes in CHD patients by using OCTA. DESIGN: Cross-sectional study. PARTICIPANTS: A total of 158 participants including 57 cyanotic CHD (CCHD) patients, 60 acyanotic CHD (ACHD) patients and 41 control subjects were included. METHODS: All participants underwent a comprehensive ophthalmologic examination, including refraction measurement, intraocular pressure measurement and OCTA. MAIN OUTCOME MEASURES: Vessel density (VD) was measured within the radial peripapillary capillary (RPC), superficial capillary plexus (SCP) and deep capillary plexus (DCP) of the macula. RESULTS: CCHD patients had significantly lower VD in the RPC, SCP and DCP (all P < .01) compared to control subjects, and significantly lower VD in the RPC and DCP (both P < .05) compared to ACHD patients. Besides, among the CHD group, VD in the RPC was positively correlated with oxygen saturation (whole image, ρ = 0.45; peripapillary, ρ = 0.48) and negatively correlated with haematocrit (whole image, ρ = 0.55; peripapillary, ρ = 0.55) (all P < .001). CONCLUSIONS AND RELEVANCE: Retinal VD might be a surrogate to reflect the effect of chronic systemic hypoxemia in CHD patients. OCTA could be a convenient and noninvasive tool to evaluate the retinal structure and function in CHD patients.

Retinal Neurovascular Impairment in Patients with Essential Hypertension: An Optical Coherence Tomography Angiography Study.

Purpose: To investigate retinal neurovascular structural changes in patients with essential hypertension. Methods: This observational cross-sectional study consisted of 199 right eyes from 169 nondiabetic essential hypertensive patients, divided into groups as follows: group A, 113 patients with hypertensive retinopathy (HTNR); group B, 56 patients without HTNR; and a control group of 30 healthy subjects. Peripapillary retinal nerve fiber layer (RNFL), radial peripapillary segmented (RPC), ganglion cell-inner plexiform layer (GC-IPL), and superficial (SVP) and deep (DVP) vascular plexus density at the macula (6 × 6 mm2) were measured by optical coherence tomography angiography (OCTA). Results: DVP density was significantly reduced in groups A and B compared to the control group (group A DVP, P = 0.001; group B DVP P = 0.002). GC-IPL, RNFL thickness, and RPC and SVP density in group A were significantly decreased compared to the control group or group B (all P < 0.05). In hypertensive patients, GC-IPL and RNFL thickness were negatively correlated with severity of HTNR (GC-IPL, r = -0.331, P < 0.001; RNFL, r = -0.583, P < 0.001) and level of home blood pressure monitoring (HBPM) (GC-IPL, r = -0.160, P = 0.050; RNFL, r = -0.282, P = 0.001) and were positively correlated with SVP (GC-IPL, r = 0.267, P = 0.002; RNFL, r = 0.361, P < 0.001) and RPC density (GC-IPL, r = 0.298, P < 0.001; RNFL, r = 0.663, P < 0.001). Among subjects with grade 2 or 3 retinopathy, the superior RNFL was significantly thinner in patients with high HBPM level than in those with normal HBPM level (grade 2, P = 0.016; grade 3, P = 0.006). Conclusions: Reduction of retinal vessel density and RNFL thickness is observed in patients with HTNR and is inversely associated with level of HBPM.

Fully Automatic Arteriovenous Segmentation in Retinal Images via Topology-Aware Generative Adversarial Networks.

Retinal image contains rich information on the blood vessel and is highly related to vascular diseases. Fully automatic and accurate identification of arteries and veins from the complex background of retinal images is essential for analyzing eye-relevant diseases, and monitoring progressive eye diseases. However, popular methods, including deep learning-based models, performed unsatisfactorily in preserving the connectivity of both the arteries and veins. The results were shown to be disconnected or overlapped by the twos and thus manual calibration was needed to refine the results. To tackle the problem, this paper proposes a topological structure-constrained generative adversarial network (topGAN) to automatically identify and differentiate the arteries and veins from retinal images. The introduced topological structure term can automatically delineate the topological structure properties of retinal blood vessels and greatly improves the vascular connectivity of the entire arteriovenous classification results. We train and evaluate our model on both the AV-DRIVE public available dataset and the CVDG home-owned dataset, which consists of 40 images and 3119 images, respectively. Experiments demonstrate that integrating topological structure constraints can significantly improve the performance of arteriovenous classification. Our method achieves excellent performance with an accuracy of 94.3% on the AV-DRIVE dataset and 93.6% on the CVDG dataset.

Restoration of Foveal Bulge after Resolution of Diabetic Macular Edema with Coexisting Serous Retinal Detachment.

PURPOSE: To evaluate the impact of restoration of foveal bulge (FB) in optical coherence tomography (OCT) images on visual acuity after resolution of diabetic macular edema with coexisting serous retinal detachment (SRD-DME). METHODS: A total of 52 eyes with resolved SRD-DME and an intact ellipsoid zone at the central fovea were included. All eyes underwent best-corrected visual acuity (BCVA) examination and OCT scanning at baseline and follow-up visits (1, 3, and 6 months). The eyes were divided into two groups according to the presence of FB at 6 months. BCVA, central foveal thickness (CFT), height of SRD (SRDH), outer nuclear layer (ONL) thickness, photoreceptor inner segment (PIS), and outer segment (POS) length were compared between the two groups. RESULTS: A FB was found in 25 of 52 (48%) eyes at 6 months. The FB (+) group had lower SRDH at baseline, and better BCVA, longer POS length at 6 months (all P < 0.05). There was no significant difference in the CFT, ONL thickness, and PIS length at 6 months between the two groups (all P > 0.05). More eyes in the FB (+) group had complete SRD resolution at 1 month (P = 0.009) and 3 months (P = 0.012). Eyes with complete SRD resolution at 1 month (P = 0.009) or 3 months (P = 0.012) were more likely to have a FB at 6 months. CONCLUSIONS: The Presence of the FB is associated with better BCVA after resolution of SRD-DME. Eyes with lower baseline SRDH or faster SRD resolution are more likely to have a FB at 6 months.

Artificial Platelets for Efficient siRNA Delivery to Clear "Bad Cholesterol".

Low-density lipoprotein cholesterol (LDL-C) is usually considered as a "bad cholesterol" for it is one of the major risk factors for coronary heart disease. As a scavenger of LDL-C, the low density lipoprotein receptor (LDLR) binds with LDL-C in the liver. However, the protein levels and function of LDLR are regulated by Proprotein convertase subtilisin/kexin type 9 (PCSK9). Loss of PCSK9 induces the increase of LDLR levels and reduction of plasma LDL-C. Here, we developed a novel style of artificial platelets with biomimetic properties, high stability, and long circulation which enabled the efficient delivery of siRNA targeting Pcsk9. The bioinspired nanoparticles induced Pcsk9 mRNA reduction by 66% in vitro. For in vivo studies, the nanoparticles accumulated in the liver to reduce Pcsk9 transcription, which results in ∼28% reduction in plasma LDL-C concentrations with negligible effects on either high density lipoprotein cholesterol (HDL-C) or triglycerides (TGs). These results demonstrated the use of artificial platelets to deliver siRNA and induce effective RNAi therapeutics to specifically lower LDL-C which provides a potential strategy to lower PCSK9 and treat hypercholesterolemia.

Homotypic Targeting Delivery of siRNA with Artificial Cancer Cells.

The camouflage with cell membrane bestows nanoparticles with cell-like functions, such as specific recognition, long blood circulation, and immune escaping. For cancer therapy, the nanoparticles camouflaged with cancer cell membrane (CCM) from homologous cells show homotypic targeting delivery of small molecule compounds, photosensitizers, or enzymes to the tumors. However, effective gene therapy encounters difficulties by this approach due to the properties of nucleic acids. Herein, a cancer cell-like gene delivery system is developed using an excellent polymer poly(ß-amino ester) (PBAE) to condense small interfering RNA (siRNA) (targeting to Plk1 gene) into nanoparticles (PBAE/siPlk1) as the core, which is further camouflaged with CCM. These novel biomimetic nanoparticles CCM/PBAE/siPlk1 (CCMPP) demonstrate highly specific targeting to homotypic cancer cells, effective downregulation of PLK1 level, and inducing apoptosis of cancer cells. Based on the homotypic binding adhesion molecules on the CCM, the cellular internalization and homotypic-targeting accumulation to the tumors are clearly improved. CCMPP induces highly efficient apoptosis of cancer cells both in vitro and in vivo and results in significant tumor inhibition. The artificial cancer cells with homotypic properties can serve as a biomimetic delivery system for cancer-targeted gene therapy.