The Diagnostic Accuracy of an Intelligent and Automated Fundus Disease Image Assessment System with Lesion Quantitative Function (SmartEye) in Diabetic Patients

Abstract Background With the diabetes mellitus (DM) prevalence increasing annually, the human grading of retinal images to evaluate DR has posed a substantial burden worldwide. SmartEye is a recently developed fundus image processing and analysis system with lesion quantification function for DR screening. It is sensitive to the lesion area and can automatically identify the lesion position and size. We reported the diabetic retinopathy (DR) grading results of SmartEye versus ophthalmologists in analyzing images captured with non-mydriatic fundus cameras in community healthcare centers, as well as DR lesion quantitative analysis results on different disease stages. Methods This is a cross-sectional study. All the fundus images were collected from the Shanghai Diabetic Eye Study in Diabetics (SDES) program from Apr 2016 to Aug 2017. 19904 fundus images were acquired from 6013 diabetic patients. The grading results of ophthalmologists and SmartEye are compared. Lesion quantification of several images at different DR stages is also presented. Results The sensitivity for diagnosing no DR, mild NPDR (non-proliferative diabetic retinopathy), moderate NPDR, severe NPDR, PDR (proliferative diabetic retinopathy) are 86.19%, 83.18%, 88.64%, 89.59%, and 85.02%. The specificity are 63.07%, 70.96%, 64.16%, 70.38%, and 74.79%, respectively. The AUC are PDR, 0.80 (0.79, 0.81); severe NPDR, 0.80 (0.79, 0.80); moderate NPDR, 0.77 (0.76, 0.77); and mild NPDR, 0.78 (0.77, 0.79). Lesion quantification results showed that the total hemorrhage area, maximum hemorrhage area, total exudation area, and maximum exudation area increase with DR severity. Conclusions SmartEye has a high diagnostic accuracy in DR screening program using non-mydriatic fundus cameras. SmartEye quantitative analysis may be an innovative and promising method of DR diagnosis and grading.

Download Full-text

The Diagnostic Accuracy of an Intelligent and Automated Fundus Disease Image Assessment System with Lesion Quantitative Function (SmartEye) in Diabetic Patients

10.21203/rs.2.275/v3 ◽

2019 ◽

Author(s):

Yi XU ◽

Yongyi WANG ◽

Bin LIU ◽

Lin TANG ◽

Liangqing LV ◽

...

Keyword(s):

Diabetic Retinopathy ◽

Quantitative Analysis ◽

Diagnostic Accuracy ◽

Proliferative Diabetic Retinopathy ◽

Screening Program ◽

Assessment System ◽

Diabetic Patients ◽

Fundus Images ◽

Cross Sectional ◽

Community Healthcare

Abstract Background With the diabetes mellitus (DM) prevalence increasing annually, the human grading of retinal images to evaluate DR has posed a substantial burden worldwide. SmartEye is a recently developed fundus image processing and analysis system with lesion quantification function for DR screening. It is sensitive to the lesion area and can automatically identify the lesion position and size. We reported the diabetic retinopathy (DR) grading results of SmartEye versus ophthalmologists in analyzing images captured with non-mydriatic fundus cameras in community healthcare centers, as well as DR lesion quantitative analysis results on different disease stages. Methods This is a cross-sectional study. All the fundus images were collected from the Shanghai Diabetic Eye Study in Diabetics (SDES) program from Apr 2016 to Aug 2017. 19904 fundus images were acquired from 6013 diabetic patients. The grading results of ophthalmologists and SmartEye are compared. Lesion quantification of several images at different DR stages is also presented. Results The sensitivity for diagnosing no DR, mild NPDR (non-proliferative diabetic retinopathy), moderate NPDR, severe NPDR, PDR (proliferative diabetic retinopathy) are 86.19%, 83.18%, 88.64%, 89.59%, and 85.02%. The specificity are 63.07%, 70.96%, 64.16%, 70.38%, and 74.79%, respectively. The AUC are PDR, 0.80 (0.79, 0.81); severe NPDR, 0.80 (0.79, 0.80); moderate NPDR, 0.77 (0.76, 0.77); and mild NPDR, 0.78 (0.77, 0.79). Lesion quantification results showed that the total hemorrhage area, maximum hemorrhage area, total exudation area, and maximum exudation area increase with DR severity. Conclusions SmartEye has a high diagnostic accuracy in DR screening program using non-mydriatic fundus cameras. SmartEye quantitative analysis may be an innovative and promising method of DR diagnosis and grading.

Download Full-text

The Diagnostic Accuracy of an Intelligent and Automated Fundus Disease Image Assessment System with Lesion Quantitative Function (SmartEye) in Diabetic Patients

10.21203/rs.2.275/v2 ◽

2019 ◽

Author(s):

Yi XU ◽

Yongyi WANG ◽

Bin LIU ◽

Lin TANG ◽

Liangqing LV ◽

...

Keyword(s):

Diabetic Retinopathy ◽

Quantitative Analysis ◽

Diagnostic Accuracy ◽

Proliferative Diabetic Retinopathy ◽

Screening Program ◽

Assessment System ◽

Diabetic Patients ◽

Fundus Images ◽

Cross Sectional ◽

Community Healthcare

Abstract Background With the diabetes mellitus (DM) prevalence increasing annually, the human grading of retinal images to evaluate DR has posed a substantial burden worldwide. SmartEye is a recently developed fundus image processing and analysis system with lesion quantification function for DR screening. It is sensitive to the lesion area and can automatically identify the lesion position and size. We reported the diabetic retinopathy (DR) grading results of SmartEye versus ophthalmologists in analyzing images captured with non-mydriatic fundus cameras in community healthcare centers, as well as DR lesion quantitative analysis results on different disease stages. Methods This is a cross-sectional study. All the fundus images were collected from the Shanghai Diabetic Eye Study in Diabetics (SDES) program from Apr 2016 to Aug 2017. 19904 fundus images were acquired from 6013 diabetic patients. The grading results of ophthalmologists and SmartEye are compared. Lesion quantification of several images at different DR stages is also presented. Results The sensitivity for diagnosing no DR, mild NPDR (non-proliferative diabetic retinopathy), moderate NPDR, severe NPDR, PDR (proliferative diabetic retinopathy) are 86.19%, 83.18%, 88.64%, 89.59%, and 85.02%. The specificity are 63.07%, 70.96%, 64.16%, 70.38%, and 74.79%, respectively. The AUC are PDR, 0.80 (0.79, 0.81); severe NPDR, 0.80 (0.79, 0.80); moderate NPDR, 0.77 (0.76, 0.77); and mild NPDR, 0.78 (0.77, 0.79). Lesion quantification results showed that the total hemorrhage area, maximum hemorrhage area, total exudation area, and maximum exudation area increase with DR severity. Conclusions SmartEye has a high diagnostic accuracy in DR screening program using non-mydriatic fundus cameras. SmartEye quantitative analysis may be an innovative and promising method of DR diagnosis and grading.

Download Full-text

Parafoveal OCT Angiography Features in Diabetic Patients without Clinical Diabetic Retinopathy: A Qualitative and Quantitative Analysis

Journal of Ophthalmology ◽

10.1155/2017/8676091 ◽

2017 ◽

Vol 2017 ◽

pp. 1-14 ◽

Cited By ~ 16

Author(s):

Mathilde M. Goudot ◽

Anne Sikorav ◽

Oudy Semoun ◽

Alexandra Miere ◽

Camille Jung ◽

...

Keyword(s):

Diabetic Retinopathy ◽

Quantitative Analysis ◽

Capillary Density ◽

Oct Angiography ◽

Foveal Avascular Zone ◽

Diabetic Patients ◽

Cross Sectional ◽

The Mean ◽

Fundus Photographs ◽

And Gender

Purpose. To evaluate the capacity of OCT angiography (OCTA) for detecting infraclinical lesions in parafoveal capillaries in diabetic patients without diabetic retinopathy (DR). Methods. This prospective observational cross-sectional case-control study analyzed the superficial and deep capillary plexuses (SCP and DCP) on macular OCTA scans (3 × 3 mm) centered on the fovea. We compared 22 diabetic patients (34 eyes included) without DR diagnosis on color fundus photographs, with 22 age- and gender-matched nondiabetic controls (40 eyes included). Qualitative analysis concerned morphological ischemic capillary alterations. Quantitative analysis measured foveal avascular zone (FAZ) size, parafoveal capillary density, and enlargement coefficient of FAZ between SCP and DCP. Results. Neither the qualitative nor quantitative parameters were significantly different between both groups. No microaneurysms or venous tortuosity was observed in any of the analyzed images. On the SCP, the mean FAZ area was 0.322 ± 0.125 mm2 in diabetic patients and 0.285 ± 0.150 mm2 in controls, P=0.31. On the DCP, the mean FAZ area was 0.444 ± 0.153 mm2 in cases and 0.398 ± 0.138 mm2 in controls, P=0.20. Conclusion. OCTA did not detect infraclinical qualitative or quantitative differences in parafoveal capillaries of diabetic patients without DR in comparison with nondiabetic controls.

Download Full-text

Stringent glycemic control is still the key to evade diabetic retinopathy.

The Professional Medical Journal ◽

10.29309/tpmj/2020.27.05.4077 ◽

2020 ◽

Vol 27 (05) ◽

pp. 1011-1016

Author(s):

Syed Munawar Alam ◽

Sagheer Ahmed ◽

Shazia Bano ◽

Shahneela Perveen

Keyword(s):

Diabetic Retinopathy ◽

Glycemic Control ◽

Proliferative Diabetic Retinopathy ◽

Study Groups ◽

Hplc Method ◽

P Value ◽

Diabetic Patients ◽

Type 2 Diabetic Patients ◽

Cross Sectional ◽

Institutional Review

Objectives: The aim of this study was to evaluate the major determinants of diabetic retinopathy. Study Design: Cross sectional, case control study. Setting: Department of Biochemistry, Basic Medical Sciences Institute, Jinnah Post Graduate Medical Centre, Karachi. Period: March 2015 to April 2016. Material & Methods: Ethical approval was taken from the Institutional Review Board of JPMC. A total of 208 people including type 2 diabetic patients and healthy control subjects; of male gender, aged between ≥30 years and ≤ 60 years were recruited and assigned to four study groups. Each group comprise of 52 individuals, depending on the ophthalmoscopy findings, i.e. healthy controls, diabetic without retinopathy (NDR), diabetic with non-proliferative diabetic retinopathy (NPDR) and diabetic with proliferative diabetic retinopathy (PDR). Fasting blood sugar was estimated using GOD-PAP method, while HbA1c was estimated by HPLC method. Data was analyzed on SPSS software version 16. Results: Diabetics with Diabetic Retinopathy had a poor glycemic control as compare to Diabetics without Diabetic Retinopathy (FBS; 109.12 ± 13.81 vs. 184.29 ± 40.07 vs. 188.6 ± 47.68 vs. 217.06 ± 62.33; p-value = 0.001) (HbA1c; 6.73 ± 0.56 vs. 8.40 ± 1.77 vs. 9.71 ± 1.85 vs. 14.91 ± 3.87; p-value = 0.001). For Diabetic Retinopathy the odds ratio of glycemic control i.e. FBS was observed as 1.019 & HbA1c was recorded as 1.561; which was statistically significant. Conclusion: Glycemic indicators; including FBS and HbA1c, are found to be the major determinants of Diabetic Retinopathy in our study.

Download Full-text

Association Between Diabetic Retinopathy, Dietary Inflammatory Index and Metabolic Syndrome in Azar Cohort Study

10.21203/rs.3.rs-107350/v1 ◽

2020 ◽

Author(s):

Mohammad hossein Somi ◽

Zeinab Nikniaz ◽

Mohammad Asghari Jafarabadi ◽

Amir Taher Eftekharsadat ◽

Mohammad Mirzaei ◽

...

Keyword(s):

Type 2 Diabetes ◽

Metabolic Syndrome ◽

Diabetic Retinopathy ◽

Cohort Study ◽

Proliferative Diabetic Retinopathy ◽

Diabetic Patients ◽

Dietary Inflammatory Index ◽

Cross Sectional ◽

Inflammatory Index

Abstract Background : The aim of present study was to evaluate the association between diabetic retinopathy (DR), dietary inflammatory index (DII), and metabolic syndrome (MetS) in patients with type 2 diabetes in a cohort study in Iran. Methods: This cross-sectional study was a part of the large Azar eye cohort study that included 1378 patients with type 2 diabetes. To diagnose DR, two mydriatric fundus photographs were captured using a digital fundus camera. The DR severity was classified as non-proliferative diabetic retinopathy (NPDR) or proliferative diabetic retinopathy (PDR). MetS was determined on the basis of the ATPIII criteria. DII was calculated according to Shivappa et al. method. Results: Of 1378 diabetic patients, 185 (13.4%) had NPDR and 142 (10.3%) had PDR. The risk of NPDR and PDR increased by 2.65-fold and 2.01-fold, respectively, in patients having blood glucose levels that fell outside the recommended range. There was no statistically significant relationship between Mets, Mets components, and DII in NPDR and PDR. Conclusion: The results suggest that intensive glycemic control, rather than conventional control, may help reduce the progression of DR. It seems that longitudinal studies and clinical trials for evaluating role of DII in DR are necessary.

Download Full-text

Diagnostic Accuracy of Nonmydriatic Fundus Photography for the Detection of Glaucoma in Diabetic Patients

BioMed Research International ◽

10.1155/2015/892174 ◽

2015 ◽

Vol 2015 ◽

pp. 1-8 ◽

Cited By ~ 5

Author(s):

Francisco J. Muñoz-Negrete ◽

Inés Contreras ◽

Noelia Oblanca ◽

M. Dolores Pinazo-Durán ◽

Gema Rebolleda

Keyword(s):

Diabetic Retinopathy ◽

Diagnostic Accuracy ◽

Screening Program ◽

Fundus Photography ◽

Visual Field Defects ◽

Diabetic Patients ◽

Neuroretinal Rim ◽

Dichotomous Classification ◽

Field Defect ◽

Field Defects

Purpose.To determine the diagnostic accuracy for glaucoma of a set of criteria with nonmydriatic monoscopic fundus photography (NMFP) in diabetics.Methods.Diabetics recruited from a screening program for diabetic retinopathy and diabetic glaucoma patients recruited from our glaucoma unit were included. Any patient with evidence of diabetic retinopathy was excluded. Diabetic patients had to have no visual field defects to be included as controls. Glaucoma patients had to have a glaucomatous field defect in at least one eye to be included. One NMFP was taken per eye for all subjects. These photographs were evaluated by two masked glaucoma specialists for the presence of the following: bilateral cup to disc (C/D) ratio ≥0.6, notching or thinning of the neuroretinal rim, disc hemorrhages, and asymmetry in the C/D ratio between both eyes ≥0.2. This evaluation led to a dichotomous classification: if any of the above criteria was present, the patient was classified as glaucoma. If none were present, the patient was classified as normal.Results.72 control subjects and 72 glaucoma patients were included. Evaluation of NMFP had a sensitivity of 79.17% and a specificity of 80.56% for specialist 1 and a sensitivity of 72.22% and a specificity of 88.88% for specialist 2 for the detection of glaucoma. The overall accuracy was 79.83% and 80.55%, respectively.Discussion.NMFP evaluation by a glaucoma specialist may be useful for the detection of glaucoma in diabetics.

Download Full-text

Serum and vitreous fibulin-1 concentrations in patients with diabetic retinopathy

Journal of Investigative Medicine ◽

10.1136/jim-2016-000130 ◽

2016 ◽

Vol 64 (7) ◽

pp. 1209-1212

Author(s):

Meiling Tian ◽

Jing Wang ◽

Yuqin Wei ◽

Qingle Lu ◽

Baohua Huang

Keyword(s):

Blood Pressure ◽

Diabetic Retinopathy ◽

Correlation Analysis ◽

Proliferative Diabetic Retinopathy ◽

Pearson Correlation ◽

Density Lipoprotein ◽

The Other ◽

Diabetic Patients ◽

Cross Sectional ◽

Pearson Correlation Analysis

Fibulin-1, an extracellular matrix glycoprotein, is closely correlated with angiogenesis. The purpose of this investigation is to determine serum and vitreous fibulin-1 concentrations in diabetic retinopathy (DR). This cross-sectional investigation was carried out in a population of 154 diabetic patients (54 without DR, 42 with non-proliferative diabetic retinopathy (NPDR) and 58 with proliferative diabetic retinopathy (PDR)) and 49 control subjects. The diabetic group showed higher serum and vitreous fibulin-1 concentrations than the controls. Serum and vitreous fibulin-1 concentrations in PDR patients were significantly elevated compared with those in the other 3 groups. NPDR patients showed elevated levels of serum and vitreous fibulin-1 concentrations compared with patients without DR. Logistic regression analysis revealed that serum and vitreous fibulin-1 were risk factors for developing DR. Pearson correlation analysis showed that serum fibulin-1 was correlated with systolic blood pressure (SBP), diastolic blood pressure (DBP), fasting plasma glucose and vitreous fibulin-1. Furthermore, Pearson correlation analysis showed that vitreous fibulin-1 was correlated with SBP, DBP, high-density lipoprotein cholesterol and serum fibulin-1. Serum and vitreous fibulin-1 concentrations are elevated under DR condition.

Download Full-text

Relationship between HbA1c Levels and Severity of Diabetic Retinopathy

Pakistan Journal of Ophthalmology ◽

10.36351/pjo.v37i4.1210 ◽

2021 ◽

Vol 37 (4) ◽

Author(s):

Saima Jamshed ◽

Ayesha Hanif ◽

Irfan Qayyum Malik ◽

Nukhba Zahid ◽

Hafiza Sadia Imtiaz

Keyword(s):

Diabetic Retinopathy ◽

Glycemic Control ◽

Proliferative Diabetic Retinopathy ◽

Hba1c Level ◽

Vitreous Hemorrhage ◽

Cross Sectional Study ◽

Diabetic Patients ◽

Sectional Study ◽

Cross Sectional ◽

Grade 5

Purpose: To determine the relationship between HbA1c and severity of retinopathy in diabetic patients. Study Design: Cross sectional study. Place and Duration of Study: Eye department of DHQ-UTH Gujranwala, from July 2020 to Dec 2020. Methods: An observational cross-sectional study including 100 patients was conducted at eye department of DHQ-UTH Gujranwala from January 2018 to December 2018. After approval from institutional review board, written informed consent with demographic variables was collected from every patient. Patients of either gender between 40-80 years of age with both types I and type II diabetes were included in this study. Complete ophthalmic examination including Best Corrected Visual Acuity, slit lamp biomicroscopy, Goldmann Applanation tonometry and fundus examination carried out. Diabetic retinopathy was classified from grade 0 to grade 5. Results: Out of 100 patients, 46 (46%) were males and 54 (54%) were females. Mean age was 60 ± 2.4 years with a range of 40 – 80 years among males and 40 – 70 years among females. Forty patients presented with grade 1 diabetic retinopathy and 4% presented with grade 5. Thirty two patients had good glycemic control while 20% had glycemic control of grade III. Patients with grade 4 diabetic retinopathy had the worst glycemic control of HbA1c level of 11.5. Conclusion: This study concluded that patients with poor glycemic control had severe diabetic retinopathy as compared to the patients with good diabetic control. There is a direct relation between HbA1c level and severity of diabetic retinopathy. Key Words: HbA1C, Diabetic retinopathy, Non proliferative diabetic retinopathy, Proliferative diabetic retinopathy, Vitreous hemorrhage.

Download Full-text

Effect of glycemic control on diabetic retinopathy and diabetic macular edema: a prospective observational study

International Journal of Advances in Medicine ◽

10.18203/2349-3933.ijam20210011 ◽

2021 ◽

Author(s):

Kajal Seema S. ◽

Jayalekshmi T. ◽

Manasa S. ◽

Prasenna M.

Keyword(s):

Diabetic Retinopathy ◽

Glycemic Control ◽

Observational Study ◽

Macular Edema ◽

Diabetic Macular Edema ◽

Proliferative Diabetic Retinopathy ◽

Fasting Blood Glucose ◽

Diabetic Patients ◽

Cross Sectional ◽

Hba1c Levels

Background: The objective of the study was to examine the effect of long-term glycemic control, as measured by glycated hemoglobin levels (HbA1C), on the onset and progression of diabetic retinopathy (DR) and diabetic macular edema (DME) over a period of 10 years.Methods: Patients who were diagnosed to have type 2 diabetes mellitus participated in this cross-sectional observational study. Duration of onset of diabetes and the glycemic control status were analyzed. Fasting and postprandial blood sugar levels and HbA1C levels were tested for every participant. Diabetic retinopathy was graded as per the ETDRS guidelines using stereoscopic fundus photographs. In addition to the clinical evaluation, optical coherence tomography was done to confirm the presence of DME.Results: A total of 212 diabetic patients were enrolled in this study. One sixty-four patients (78.1%) had DR, out of which 71 patients (43.3%) had mild non proliferative diabetic retinopathy (NPDR), 42 patients (25.6%) had moderate NPDR, 31 patients (18.9%) had severe NPDR, and 20 (12.2%) had proliferative diabetic retinopathy (PDR). Fifty-nine patients with DR (36%) had DME. Duration of diabetes (14.62±6.18 vs 9.72±3.68 years, p<0.001), higher fasting blood glucose (176.79±59.13 vs 138.46±49.44 mg/dl, p<0.001) and higher HbA1c levels (8.21±1.38 vs 7.48±1.25 %, p=0.002) were significantly associated with DR.Conclusions: The stage of diabetic retinopathy rather than metabolic status is a strong predictive factor for the development of diabetic macular edema.

Download Full-text