Translational Vision Science & Technology最新文献

Purpose: Accurate diagnosis of retinal disease based on optical coherence tomography (OCT) requires scrutiny of both B-scan and en face images. The aim of this study was to investigate the effectiveness of fusing en face and B-scan images for better diagnostic performance of deep learning models.

Methods: A multiview fusion network (MVFN) with a decision fusion module to integrate fast-axis and slow-axis B-scans and en face information was proposed and compared with five state-of-the-art methods: a model using B-scans, a model using en face imaging, a model using three-dimensional volume, and two other relevant methods. They were evaluated using the OCTA-500 public dataset and a private multicenter dataset with 2330 cases; cases from the first center were used for training and cases from the second center were used for external validation. Performance was assessed by averaged area under the curve (AUC), accuracy, sensitivity, specificity, and precision.

Results: In the private external test set, our MVFN achieved the highest AUC of 0.994, significantly outperforming the other models (P < 0.01). Similarly, for the OCTA-500 public dataset, our proposed method also outperformed the other methods with the highest AUC of 0.976, further demonstrating its effectiveness. Typical cases were demonstrated using activation heatmaps to illustrate the synergy of combining en face and B-scan images.

Conclusions: The fusion of en face and B-scan information is an effective strategy for improving the diagnostic accuracy of deep learning models.

Translational relevance: Multiview fusion models combining B-scan and en face images demonstrate great potential in improving AI performance for retina disease diagnosis.

Purpose: To investigate the efficacy of a novel approach using a sterile caliper for anterior chamber (AC) decompression in reducing post-intravitreal injection (IVI) intraocular pressure (IOP) spikes.

Methods: A prospective interventional case series conducted at the Iowa City Veterans Affairs Medical Center (VAMC) with Institutional Review Board approval. Patients were randomized to receive conventional IVI or IVI with sterile caliper decompression. Fifty eyes from 47 patients underwent IVI for various retinal pathologies. Subjects were randomly assigned to the intervention or control arm. Two resident physician providers performed injections, with one applying sterile caliper decompression (intervention) and the other following the standard technique (control). Baseline and postinjection IOP were measured using Tonopen (Reichert, Depew, NY).

Results: In both groups there was a significant IOP rise following IVI (P < 0.0001). There was no significant difference in baseline IOP between groups (P = 0.082), but postinjection IOP was significantly lower in the intervention group (23.52 ± 5.98 mm Hg) compared to the control group (44.08 ± 8.48 mm Hg). There were no patients with an IOP spike >25 mm Hg in the intervention arm. The technique was effective regardless of lens status.

Conclusions: Sterile caliper AC decompression significantly reduced post-IVI IOP spikes presenting an efficient and cost-effective alternative to previously proposed methods of IOP reduction. Further studies are warranted to validate these findings and explore broader applications in ophthalmic interventions.

Translational relevance: The caliper decompression technique presents potential benefit in preventing short-term morbidity associated with IOP spikes after IVI and addressing long-term concerns in patients with pre-existing glaucoma.

Purpose: To investigate the motility of the primary cilia of corneal endothelial cells (CECs), which exist like a hair on the cell surface, using our new in vitro method.

Methods: A white glass rod was heated with a gas burner to produce a rod approximately 0.5 mm in diameter and 20 mm in length and then coated with collagen. A suspension of cultured human CECs (HCECs) was then added to the rod and cultured for 20 days. Cells on the rod's side were then observed using phase-contrast microscopy, and videos and images of the primary cilia were obtained. After fixing the cells cultured on the rod's surface, immunofluorescence staining was performed and fluorescence and phase contrast images were taken.

Results: Hair-like structures were observed on the surface of live HCECs on the rod's surface. Video images revealed that the structures sometimes swayed owing to slight convection of the medium, yet had no motile function, and immunostaining with acetylated α-tubulin antibody confirmed that the structures were primary cilia.

Conclusions: Our new method using white glass rods provided the ability to observe the movement of primary cilia in cultured living HCECs, and the findings clearly showed that the primary cilia of HCECs are passive rather than motile. This novel procedure can be applied widely to other cultured cells as a method to observe the movement of primary cilia from the lateral aspect of the cell.

Translational relevance: This method may help to clarify the role of primary cilia in the anterior chamber.

Purpose: To develop a short form of Chinese ULV-VFQ-50 based on the ULV-VFQ-150 and compare the psychometrical properties of the two questionnaires.

Methods: We selected candidate items from the ULV-VFQ-150, considering the item response among ultra-low vision (ULV) participants, the even distribution of item measures, visual aspects, and visual domains, to construct a 50-item ULV-VFQ-50 questionnaire. Then, ULV participants were recruited to evaluate its psychometric characteristics by using Rasch analysis.

Results: In total, 79 out of 79 completed questionnaires of ULV-VFQ-50 were collected, of which 11 filled questionnaires were excluded because the participants' vision did not meet the inclusion criteria. Thus 68 valid questionnaires were analyzed (valid response rate 91.9%). The average age of the eligible responders was 45.0 years (standard deviation [SD] = 16.7), with 42.6% females (29/68). As per Rasch analysis, the person measures ranged from -1.74 to 4.91 logits, and the item measures ranged from -1.56 to 1.15 logits. The mean value of item difficulty was 0.00 logits, whereas the mean value of personnel distribution was -0.35 logits. The item reliability was 0.95, and the person reliability was 0.98. The items conform to unidimensionality as indicated by principal component analysis of the residuals, which showed that the first principal component unexplained only 5.5% of the total variance, and each component after that unexplained even less.

Conclusions: The Chinese ULV-VFQ-50 exhibits excellent psychometric properties. The short form of Chinese ULV-VFQ, with fewer items and less administration time is better suited for clinical practice and research settings.

Translational relevance: The Chinese version of ULV-VFQ-50 is a reliable assessment of the visual function for people with ULV in China.

Purpose: To distinguish between multiple sclerosis (MS) and glaucoma by nerve fiber layer (NFL) thinning patterns.

Methods: MS patients were diagnosed by the 2017 McDonald Criteria; glaucoma patients had disc rim thinning or an NFL defect, with or without perimetric defect. The peripapillary NFL thickness was divided into eight sectors, and percentage reduction (% reduction) was calculated relative to normative reference values. The MS and glaucoma eyes were grouped based on the severity of NFL thinning in the worst sector: significant reduction (<1 percentile of normal reference), borderline reduction (1%∼5%), and no reduction (>5%). We devised four diagnostic indexes, and the area under the curve of receiver operating characteristics (AROC) and accuracy were used to evaluate the indexes.

Results: We enrolled 58 control subjects (58 eyes), 56 MS subjects (112 eyes), and 92 glaucoma subjects (92 eyes) at two centers. The most pronounced percent reduction in MS eyes occurred in the temporal-upper and temporal-lower sectors. In glaucoma eyes, this occurred in the inferior-temporal, inferior-nasal, and superior-temporal sectors. The temporal pattern index had the best AROC (0.96, 0.91-1.00) and accuracy (92.6%) in the significant reduction group. It had good AROC (0.88, 0.78-0.99) and accuracy (76.7%) in the borderline reduction group.

Conclusions: Normalizing NFL reduction as a percentage of normal reference accentuated patterns characteristic of MS and glaucoma. Quantitative pattern indexes were effective in differentiating the two diseases.

Translational relevance: The utility of optical coherence tomography in the differential diagnosis of optic neuropathies is enhanced by analyzing the retinal nerve fiber layer percentage reduction pattern.

Purpose: To evaluate the association of choroidal thickness (ChT) with best-corrected visual acuity (BCVA) in patients with high myopia (HM) and without myopic maculopathy.

Methods: This study was a retrospective, cross-sectional study of participants aged 7 to 70 years with bilateral HM but without myopic maculopathy. Swept-source optical coherence tomography was used to measure ChT at the fovea. A BCVA of logMAR 0.1 was regarded as the benchmark for normal. The association between ChT and BCVA was evaluated by linear regression models with confounders fully adjusted. Subgroup analyses were performed across sex and age groups.

Results: A total of 412 eligible participants were enrolled in this study. The mean age, spherical equivalence, and subfoveal (SF) ChT of the included participants were 21.17 ± 9.55 years, -9.77 ± 2.40 diopters, and 171.56 ± 61.33 µm, respectively. The SF ChT was thinner in participants with abnormal BCVA (normal. 176.74 ± 60.24 µm; abnormal, 139.30 ± 58.63 µm). A thinner ChT in all subregions of the posterior pole of the Early Treatment Diabetic Retinopathy Study grid was associated significantly with a worse BCVA after adjusting for age, sex, and axial length (SF ChT: coefficient, × 10-4, -2.64; 95% confidence interval, -4.73 to -0.55; P < 0.05 in all subregions). The strongest correlation was observed in the outer inferior region, where a per 21 µm thinning of ChT led to a 0.01 worsening of BCVA. This correlation presented a stronger magnitude in male aged more than 40 years.

Conclusions: In patients with HM without myopic maculopathy, a thinner ChT was associated independently with a worse BCVA.

Translational relevance: The findings of this study suggest that thinning ChT should be considered a vital risk factor for irreparable visual acuity impairment.

Purpose: We investigated the association between metabolically healthy obesity (MHO) and retinal age gap and explored potential sex differences in this association.

Methods: This study included 30,335 participants from the UK Biobank. Body mass index (BMI) was classified into normal weight, overweight, and obesity. Metabolic health (MH) was defined as meeting the following criteria: systolic blood pressure of <130 mm Hg, no antihypertensive drugs, waist-to-hip ratio of <0.95 for women or 1.03 for men, and the absence of diabetes. Participants were categorized as MH normal weight (MHN), MH overweight (MHOW), MHO, metabolically unhealthy normal weight, metabolically unhealthy (MU) overweight, and MU obesity. Retinal age gap was defined as the difference between retinal age and chronological age. Linear regression models were used to investigate the association of metabolic phenotypes of obesity with retinal age gap.

Results: Compared with MHN, individuals with MHOW (β, 0.17; 95% confidence interval [CI], 0.01-0.32; P = 0.039) and MHO (β, 0.23; 95% CI, 0.02-0.44; P = 0.031) were associated with increased retinal age gap. Furthermore, individuals classified as metabolic unhealthy were also associated with higher retinal age gap, irrespective of body mass index categories (β for MU normal weight, 0.23; 95% CI, 0.08-0.38; P = 0.003; β for MU overweight: 0.31; 95% CI, 0.18-0.45; P < 0.001; β for MU obesity, 0.50; 95% CI, 0.36-0.65; P < 0.001). No significant sex difference was observed in the association between metabolic phenotypes of obesity and retinal age gap (all P for interaction > 0.05).

Conclusions: MHOW and MHO were associated significantly with an increased retinal age gap compared with MHN individuals. Weight management should be recommended for individuals who are overweight or obese, even in the absence of metabolic unhealth.

Translational relevance: Retinal age gap provides a simple tool for identifying early health risks for MHOW and MHO individuals.

Purpose: Automated segmentation software in optical coherence tomography (OCT) devices is usually developed for and primarily tested on common diseases. Therefore segmentation accuracy of automated software can be limited in eyes with rare pathologies.

Methods: We sought to develop a semisupervised deep learning segmentation model that segments 10 retinal layers and four retinal features in eyes with Macular Telangiectasia Type II (MacTel) using a small labeled dataset by leveraging unlabeled images. We compared our model against popular supervised and semisupervised models, as well as conducted ablation studies on the model itself.

Results: Our model significantly outperformed all other models in terms of intersection over union on the 10 retinal layers and two retinal features in the test dataset. For the remaining two features, the pre-retinal space above the internal limiting membrane and the background below the retinal pigment epithelium, all of the models performed similarly. Furthermore, we showed that using more unlabeled images improved the performance of our semisupervised model.

Conclusions: Our model improves segmentation performance over supervised models by leveraging unlabeled data. This approach has the potential to improve segmentation performance for other diseases, where labeled data is limited but unlabeled data abundant.

Translational relevance: Improving automated segmentation of MacTel pathology on OCT imaging by leveraging unlabeled data may enable more accurate assessment of disease progression, and this approach may be useful for improving feature identification and location on OCT in other rare diseases as well.

Purpose: To evaluate the performance of two non-disease-specific patient-reported outcome (PRO) instruments, the National Eye Institute Visual Function Questionnaire-25 (VFQ-25) and the Low Luminance Questionnaire (LLQ), in patients with retinaldehyde-binding protein 1 retinal dystrophy (RLBP1 RD).

Methods: PROs were assessed using the VFQ-25 and LLQ. Rasch analysis was conducted to estimate person and item measures of the VFQ-25 and LLQ questionnaires to determine the association between the two PROs. In addition, the association between these two instruments and their correlations to weighted measures of visual function and disease progression were analyzed in this three-year PRO-focused sub-study of a five-year prospective natural history study.

Results: Forty-two patients participated, with most of them having completed at least two PRO follow-up visits at least one year apart. The mean VFQ-25 scores were lowest for distance activities (39.2-49.0) and peripheral vision (37.5-52.4), with mean LLQ subscale scores generally low (<41), except for the emotional distress subscale. Using Rasch analysis, calibrated item and person measures along with their standard errors were estimated for both ePROs. This indicated that the distribution of the VFQ-25 and LLQ item measures well covered the distribution of person function in this group. This suggests that the item difficulties well cover the person-level performance in this population. As well, the two PROs showed a strong and significant correlation at all assessed time points as assessed with Pearson correlation coefficient (0.81, 0.91, 0.81 and 0.87 at baseline, 1/1.5, 2/2.5 (P < 0.001) and 3/3.5 years (P = 0.002)). The composite scores of both PRO questionnaires strongly correlated with clinical measures of visual function. At 2 to 2.5 years of follow-up, meaningful statistically significant declines in peripheral vision (both VFQ-25 and LLQ), distance vision (VFQ-25), and extreme lighting in dark and bright light (LLQ) subscales were noted.

Conclusions: This study demonstrated a strong association between VFQ-25 and LLQ scores and their association with clinical measures of visual function.

Translational relevance: PRO instruments can provide insights into the specific disabilities of this unique patient population and help to guide appropriate outcome measures for future clinical trials.