Eye

Abstract Purpose To apply a mathematical model integrating vascular permeability, lesion transmural pressure, choroidal excess pressure, and wall stability to incomplete fluid response and massive submacular hemorrhage in neovascular age-related macular degeneration. Methods Permeability was calculated from aqueous vascular endothelial growth factor (VEGF). Windkessel gave usual lesion pressure from choriocapillaris pressure and high inflow from an arterial-level puncture pressure used as a working value. Choroidal excess pressure was calculated from choroidal thickness and a pachy index as increased outflow resistance. Exudative change used a revised Starling equation scaled to early central-retinal-thickness reduction, with 40-m threshold. A relative rupture-hazard index was calculated from Laplace wall stress and wall stability. VEGF, pressures, and that reduction were from the literature. Pressure was isolated from permeability and choroidal excess. Results In the model, raising lesion pressure from choroidal-excess lesion pressure to high inflow increased untreated exudative change from 150 to 760 m. Residual under saturation was 64 m as a lower-end reference (Monte Carlo 69 - 286 m; median, 170) and 90 m with high-hazard choroidal excess; both above 40 m. Reducing choroidal thickness in the thickened choroidal-excess state added 13 - 14 m, smaller than that residual. Lowering lesion pressure from high inflow to usual lesion pressure (21 to 10 mmHg) reduced the index to 0.64 of the high-hazard reference, and to 0.34 with wall stabilization. Conclusion Lesion transmural pressure due to high inflow left residual exudation even after permeability suppression with anti-VEGF. Fluid resolution on optical coherence tomography may not parallel a reduction in the rupture-hazard index, depending on wall stability.

Purpose: Post-marketing reports have associated faricimab, the first intravitreal bispecific antibody, with intraocular inflammation (IOI). Its inflammatory phenotype, temporal pattern and systemic safety relative to established anti-VEGF agents remain unclear. We compared faricimab, aflibercept and ranibizumab across three independent pharmacovigilance databases. Methods: We analysed FAERS reports from Q1 2022 to Q2 2026, with external validation in JADER (PMDA, Japan) and EudraVigilance. Reports were de-duplicated, restricted to primary-suspect reports, and a pre-specified list of 36 non-adverse-event preferred terms (PTs) was excluded. We computed reporting odds ratios (ROR), proportional reporting ratios and Bayesian information components, with Benjamini-Hochberg false discovery rate control. Head-to-head RORs were adjusted for demographic confounders by logistic regression. Time to onset (TTO) was modelled with Weibull distributions. Sensitivity analyses addressed masking and differential market duration. Signals were graded with a pre-specified bidirectional cross-database evidence framework (Confirmed, Consistent, Unconfirmed, Discordant). Results: Faricimab was reported more often for anterior-chamber and vitreal inflammation. Adjusted RORs (comparator versus faricimab, values <1 indicating faricimab excess) were 0.27 (95% CI 0.18-0.40) for iridocyclitis and 0.45 (0.35-0.58) for uveitis versus aflibercept, and 0.13 (0.06-0.24) and 0.12 (0.06-0.20) versus ranibizumab. All four PTs were Confirmed in JADER against both comparators; for iridocyclitis, RORs were 0.048 versus aflibercept and 0.109 versus ranibizumab. Faricimab IOI arose late (median TTO 100 days, IQR 31-182, versus 7 days for both comparators; P = 2.6E-14) and its Weibull onset distribution was markedly flatter (shape 0.81 versus {approx}0.42). In contrast, aflibercept predominated for endophthalmitis-spectrum events and intraocular pressure elevation, and ranibizumab for vitreous haemorrhage, the latter confirmed in JADER. Faricimab showed lower reporting across cardiac, nervous system, vascular and renal system organ classes, confirmed in EudraVigilance. An unadjusted FAERS death signal (ROR 3.16, aflibercept versus faricimab) attenuated after adjustment (aROR {approx}1.8) and was absent in outcome-based JADER analyses. Conclusions: Faricimab exhibits a reproducible, delayed-onset, anterior-predominant IOI phenotype alongside a comparatively favourable systemic safety profile. Clinicians should be aware that faricimab-associated IOI may arise weeks to months after injection. Prospective studies with exposure denominators are needed to quantify absolute risk.

Purpose To evaluate the safety, tolerability, and preliminary bioactivity of a single suprachoroidal injection of AXT107 (Gersizangitide), a multimodal integrin-disrupting peptide that acts on both the VEGFA pathway and angiopoietin-Tie2 pathway, in eyes with neovascular age-related macular degeneration (nAMD). Design Phase 1/2a, open-label, dose-escalation clinical trial. Participants Fifteen subjects with active nAMD, including 11 treatment-experienced partial responders to intravitreal anti-vascular endothelial growth factor (VEGF) therapy and 4 treatment-naive subjects; 14 eyes met the protocol-specified baseline imaging criterion and were included in efficacy analyses. Methods Study eyes received a single suprachoroidal injection of AXT107 microparticles at doses of 0.125, 0.25, or 0.5 mg and were followed for 40 weeks. Protocol-specified rescue intravitreal anti-VEGF treatment was permitted from week 12 based on protocol-specified best-corrected visual acuity (BCVA) or central subfield thickness (CST) criteria; anti-VEGF treatment before Week 12 was investigator-initiated. Primary outcomes were ocular and systemic safety through week 40; secondary outcomes included change in CST, change in BCVA, and time to first protocol-specified rescue injection. Main Outcome Measures Incidence of ocular and systemic adverse events, intraocular pressure (IOP), CST, BCVA, and time to first protocol-specified rescue injection. Results Twenty subjects were screened and 15 enrolled across 4 U.S. sites; 3 eyes received 0.125 mg, 3 received 0.25 mg, and 9 received 0.5 mg of AXT107. No serious adverse events related to AXT107 occurred, and no cases of endophthalmitis, retinal detachment, suprachoroidal hemorrhage, or clinically significant intraocular inflammation were observed through week 40. No clinically meaningful intraocular pressure (IOP) elevations occurred at any dose, and no subject required chronic IOP-lowering therapy. Among the 14 eligible eyes included in efficacy analyses, 2 eyes (01-04 and 01-05) completed 40 weeks without protocol-specified anti-VEGF rescue, and 1 eye (02-04) first received protocol-specified rescue at week 36. At the last on-treatment visit, BCVA changes ranged from +15 to -31 ETDRS letters; 3 eyes gained[&ge;]5 letters, 3 gained[&ge;]10 letters, and 1 gained[&ge;]15 letters. Among the 14 eligible eyes, anatomical improvements in CST were observed in at least 1 eye in each dose cohort. The most durable responses were seen in treatment-experienced eyes receiving 0.5 mg (median durability of 36 weeks), whereas treatment-naive 0.5-mg eyes showed more variable courses and generally earlier protocol-specified rescue in this small cohort. Conclusions In this first-in-human study, a single suprachoroidal injection of AXT107 up to 0.5 mg was well tolerated through Week 40, with no drug-related serious adverse events, no clinically significant intraocular inflammation, and no sustained IOP elevations. After a single suprachoroidal AXT107 injection, 2 eyes completed 40 weeks without protocol-specified rescue anti-VEGF treatment, and 1 additional eye received its first protocol-specified rescue treatment at Week 36. The exploratory observation that durability appeared greatest in treatment-experienced eyes supports further controlled study of AXT107, particularly in treatment-experienced nAMD populations.

Abstract Purpose: To use a minimally invasive lid treatment (MILT) utilizing Compressive Skin Contouring (CoCo) technology to compare patient- and surgeon-reported ecchymosis, edema, and incision appearance following MILT and traditional upper eyelid blepharoplasty (tBULB). Methods: Prospective, single-surgeon intra-patient study comparing MILT and tBULB in patients seen for bilateral blepharoplasty. Patients and the operating surgeon rated MILT and tBULB outcomes post-operatively on the day of surgery (T1) and at 2 months (T2) using surveys. Patient photographs taken 3 and 9 days post-operatively were blindly assessed by 3 independent surgeons to compare ecchymosis, edema, and incision appearance on the MILT and tBULB sides. Results: Seven patients participated in the study. At T1, 6/7 patients reported [&le;] ecchymosis and edema, and all patients reported better or equivalent incision appearance on the MILT side compared to the tBULB side. At T2, 6/7 of patients reported [&le;] edema; 4/7 reported < ecchymosis and better incision appearance on the MILT side. The surgeon reported < ecchymosis and [&le;] edema for 7/7 patients and better or equivalent incision appearance for 6/7 patients at T1; [&le;] ecchymosis and better or equivalent incision appearance for 7/7 and [&le;] edema for 6/7 patients at T2 on the MILT side. In the blinded evaluation, mean ecchymosis scores were lower, edema scores were lower or equivalent, and incision appearance was better for MILT vs. tBULB at days 3 and 9. Conclusions: In this intra-patient study, MILT outcomes compared favorably with tBULB, especially for ecchymosis and edema; two important aesthetic measures of patient perception.

Purpose: To understand the functional impact of macular atrophy (MA) and fibrosis in eyes with treated neovascular age-related macular degeneration (nAMD) based on deep visual sensitivity losses. Design: Cross-sectional study. Participants: One-hundred and nine eyes from 97 individuals with nAMD that had been treated for >=12 months without significant retinal fluid. Methods: All participants underwent defect-mapping microperimetry (DMP) testing - a strategy optimized to quantify the spatial extent of deep visual sensitivity losses through single 10 decibel stimuli presentations - where 208 locations were sampled in the central 8 degrees radius region. Participants also underwent color fundus photography, fundus autofluorescence and OCT imaging, which were co-registered and manually annotated for MA and fibrosis to determine their two-dimensional pointwise extent at individual test locations, and global extent in the corresponding region where DMP was performed. Main Outcome Measures: Pointwise association between the presence of MA and fibrosis with missing a stimulus on DMP and global association between the extent of MA and fibrosis with the proportion of locations missed (PLM) on DMP. Results: At a pointwise level, the presence of MA and fibrosis were independently associated with an increased likelihood of missing a stimulus on DMP (odds ratio = 42.5 and 2.5 respectively; P < 0.001 for both). However, at a global level, only an increasing extent of MA was associated with a significantly higher PLM on DMP (P < 0.001), but not fibrosis (P <= 0.227). MA extent explained a substantial proportion of the variance in the PLM (R2 = 0.97) and there was a very strong correlation between the two parameters ({rho} = 0.93). Conclusions: This study confirmed the expected functional impact of MA and fibrosis on DMP testing at a pointwise level. However, it showed at the global level that only the extent of MA, and not fibrosis, was independently associated with functional loss on DMP, and that there was a very strong structure-function correlation based on MA. These findings thus suggest that evaluating MA alone may be sufficient for capturing structural changes strongly associated with the global extent of deep visual sensitivity losses in those with treated nAMD.

Published artificial intelligence for keratoconus is unproven in fellow eyes that topography and tomography call normal, the eyes where surgical decisions are made, although it reports near-perfect accuracy for early disease. In a registered systematic review of 573 studies (PROSPERO CRD420261441197), none of 2,035 claim-by-stratum units from 212 adjudicated studies established discrimination in these eyes. Most units (90.0%) answered a different question, never constructed the target population or certified its normality; external validation did not test label generation, and five systematic reviews did not assess it. In a public cohort, a frozen score yielded an area under the receiver operating characteristic curve of 0.996 against a same-eye index label and 0.549 against a contralateral proxy; its inputs reconstructed the label-generating index (r = 0.991), so accuracy there measured agreement with the index. Anchor-gap evaluation measures accuracy in these eyes against the contralateral clinical diagnosis.

Retinal vascular features are proposed biomarkers of systemic health, yet their longitudinal behaviour and relationship to systemic health remain poorly understood. We quantified retinal vascular morphology from serial colour fundus photographs in the Canadian Longitudinal Study on Aging, including 21,562 participants at baseline with good quality images. On average, vascular complexity and density decreased between baseline visit and follow-up 2 (mean interval 5.7 years), whereas tortuosity, arteriolar widths, and venular widths increased. Baseline vascular features were associated with demographic, cardiovascular, metabolic and behavioural traits, whereas altered longitudinal trajectories were associated principally with older age and higher low-density lipoprotein cholesterol. Within-person increases in mean arterial pressure covaried with lower vascular complexity, density and vessel widths. In survival analyses, independently of baseline vascular features, increases in fractal dimension were associated with lower hazard of subsequent all-cause mortality (hazard ratio = 0.92 per standard deviation, 95% confidence interval (CI) = 0.86 - 0.98), whereas arteriolar widening was associated with higher mortality (hazard ratio = 1.09, 95% CI = 1.02 - 1.17). No retinal change metric was associated with major adverse cardiovascular events after false-discovery-rate correction. These findings identify retinal vascular remodelling as a dynamic phenotype associated with systemic health and subsequent mortality, and motivate evaluation of multi-time-point retinal imaging in oculomics.

Purpose: To characterize the biomechanical response of the optic nerve head (ONH) in myopic versus non-myopic individuals with normal-tension glaucoma (NTG), and to evaluate the association between ONH strains and the rate of visual field (VF) progression. Design: Retrospective clinical cohort study. Participants: A total of 118 NTG eyes from 77 patients with [&ge;] 2 years of follow-up and 5 VF tests were enrolled, including 66 eyes with myopia and 52 eyes without. Methods: Each participant underwent three-dimensional (3D) optical coherence tomography (OCT) imaging of the ONH under three mechanical loading conditions: 20{degrees} adduction, 30{degrees} abduction and acute intraocular pressure (IOP) elevation by applying a 0.80 N force to the lower eyelid. Effective strains across the global ONH complex and within specific segmented tissues (pre-lamina, lamina cribrosa, choroid, and sclera) were quantified using a validated 3D strain-mapping algorithm. Results: Myopic NTG eyes exhibited significantly higher ONH strains compared to non-myopic eyes under acute IOP elevation (10.1{+/-}2.7% vs. 9.3{+/-}1.8%, P = 0.044) and abduction (9.2{+/-}2.3% vs. 8.4{+/-}2.1%, P = 0.048). In the non-myopic group, ONH strains varied significantly across conditions (P < 0.001), with adduction (9.9{+/-}2.0%) and acute IOP elevation (9.3{+/-}1.8%) eliciting greater strains than abduction (8.4{+/-}2.1%). These condition-dependent variations were less pronounced in the myopic group (10.1{+/-}2.7% vs. 9.8{+/-}2.7% vs. 9.2{+/-}2.3%, P = 0.07). Multivariable linear mixed-effects analysis revealed that in myopic NTG, higher ONH strains induced by adduction were significantly associated with faster MD decline (estimate: -0.139 dB/year per 1% increase, P = 0.03, adjusted for age, spherical equivalent and baseline perimetric mean defect). Conversely, larger strains during acute IOP elevation were associated with slower progression (estimate: 0.184 dB/year per 1% increase, P = 0.004). Tissue-specific analysis confirmed that adduction-induced strains in the choroid and pre-lamina tissue as well as IOP-induced strains across all ONH tissues were significantly correlated with the rate of perimetric progression in myopic eyes (all P < 0.05). Conclusions: In this cohort, greater adduction-induced ONH strain was associated with faster perimetric deterioration in myopic NTG. These findings support further investigation of gaze-related optic nerve traction as a potential non-IOP-related biomechanical factor in myopic NTG. Keywords: Normal tension glaucoma, Myopia, Optic nerve head, Ocular biomechanics, Visual field Progression, adduction.

Purpose: To evaluate the associations of the numerical AI-CO center index and visually classified central dark-area extent with Humphrey 10-2 mean deviation (MD). Methods: This single-center retrospective study included 320 eyes of 169 patients with AI-CO imaging, broad-field 24-2F or 30-2 testing, and 10-2 testing. Both visual field examinations were performed within 365 days before or after AI-CO imaging. Two ophthalmologists independently classified central dark-area extent into five Levels (0-4). Analyses assessed Spearman correlations, areas under the receiver operating characteristic curve (AUCs), distributions across final central dark-area Levels, and interrater reliability. Patient-cluster bootstrap confidence intervals and a one-eye-per-patient sensitivity analysis addressed inclusion of both eyes. Results: AI-CO center correlated with 10-2 MD (rho=0.829; 95% CI, 0.773-0.873). AUCs for 10-2 MD below -2, -6, and -12 dB were 0.902, 0.939, and 0.961, respectively. The numbers of eyes in Levels 0-4 were 66, 57, 94, 53, and 50; median 10-2 MD values were -0.59, -1.48, -3.54, -6.24, and -18.15 dB, respectively. Quadratic-weighted kappa was 0.970 (95% CI, 0.958-0.981). Sensitivity analyses yielded broadly similar findings. Conclusions: The numerical AI-CO center index and visually classified central dark-area extent were associated with central visual field damage. Assessing the extent of dark regions in the central image alongside the numerical index may provide useful information for interpreting central visual field status from the AI-CO display.

Ophthalmic decisions require probabilities tied to a clinical question, a candidate set and an operating policy. We present OphJev, an endpoint-aware System One framework that combines frozen image and text representations, candidate scoring, schema-specific calibration and selective generalist queries for fundus and OCT imaging. Across eight external resources, multiple representation families, three generalist sizes and three seeds, controlled interventions separate the effects of wording, supervision and computation. Schema-aware affine calibration lowers mean held-out Brier loss and selective risk in all eight primary image--text configurations. On HYGD, matched-supervision fusion raises AUROC from 0.772 to 0.798, while a separate gain-routing policy lowers Brier from 0.618 to 0.599 using at most 20\% logical generalist image calls. These effects demonstrate complementary decision-level capabilities rather than universal gains: external transfer and larger specialist banks remain cohort dependent. A retrospective text extension evaluates 1,996 clinical notes and 91 ophthalmic questions; validation-based calibration improves note probability scores, while a lexical control remains strong. OphJev provides an auditable decision framework; prospective clinical benefit remains unvalidated.

Purpose: To develop an interpretable automated pipeline for dome-shaped macula (DSM) detection and measurement from optical coherence tomography (OCT) B-scans. Methods: We analyzed one eye from each of 300 children aged 4-18 years in the Shanghai Child and Adolescent Large-scale Eye Study High Myopia (SCALE-HM) cohort (50 DSM-positive; 100 test eyes). A U-Net pretrained on OCTA-500 and fine-tuned on the development set segmented the retinal pigment epithelium (RPE) boundary. A geometric module located the apex, fitted a tangent-line baseline, and computed perpendicular height in physical coordinates. An eye was DSM-positive if any gradable B-scan had height >= 50 um. Two masked graders evaluated the test set; a third adjudicated disagreements. Results: The area under the receiver operating characteristic curve was 0.927 (95% confidence interval [CI], 0.850-0.986), sensitivity was 82.1% (95% CI, 64.4%-92.1%), and specificity was 100% (95% CI, 94.9%-100%) against the adjudicated reference. In concordant-positive eyes, automated height agreed with manual measurements (intraclass correlation coefficient, 0.949; mean absolute error, 6.9 um; bias, -5.7 um, versus Doctor A). Mean RPE-boundary error on pediatric test scans was 1.54 um. Conclusions: The pipeline may be considered a reliable and reproducible tool for automated DSM detection and height measurement on radial OCT, with each value traceable to landmarks on the source B-scan. External validation across populations, devices, and acquisition protocols is needed. Translational Relevance: The pipeline provides an objective and reproducible method to detect DSM and quantify dome height, width, and orientation across OCT meridians. Because each measurement can be verified on the source B-scan, it enables standardized DSM assessment.

Background: In 2026, the US Food and Drug Administration (FDA) identified uveitis as a potential signal of serious risk for the B-cell maturation antigen (BCMA) x CD3 bispecific antibodies teclistamab and elranatamab. Uveitis in this setting may reflect opportunistic infection or immune-mediated inflammation, which require opposite management. Methods: We analyzed FDA Adverse Event Reporting System (FAERS) reports from July 2020 to June 2026 (8,690,601 reports after deduplication) using a case-non-case design, calculating reporting odds ratios (RORs) and the information component for primary-suspect reports. Comparators were belantamab mafodotin, talquetamab and CD3 bispecific antibodies used outside myeloma. Every primary-suspect uveitis report (n = 76) was classified as infectious, probable immune-mediated (hereafter immune-mediated) or indeterminate by a physician blinded to a rule-based algorithm. Results: Uveitis was reported in 18 of 2,995 teclistamab and 29 of 1,691 elranatamab reports (ROR 4.10 [95% CI 2.58-6.52] and 11.85 [8.21-17.12]). Of these 47 cases, 20 were infectious (18 cytomegalovirus chorioretinitis; median onset 92 days), and infection-specified uveitis was disproportionately reported for both agents and for CD3 bispecific antibodies used outside myeloma. Thirteen were immune-mediated, arose early (median onset 4 days) and were co-reported with cytokine release syndrome in six; immune-mediated uveitis showed a signal only for elranatamab (8 cases; ROR 3.59, 1.79-7.18), not for teclistamab or any comparator. Six of the eight elranatamab cases originated from Japan. Agreement between physician and algorithm was 98.7%. Conclusions: The FDA uveitis signal appears to comprise a late infectious uveitis shared by most T-cell-redirecting bispecific antibodies examined and an early immune-mediated uveitis that showed a signal only with elranatamab. These hypothesis-generating findings support prompt ophthalmological assessment of new visual symptoms and exclusion of cytomegalovirus retinitis before corticosteroids are started.

Purpose: Electrical stimulation of the occipital cortex evokes visual percepts known as phosphenes. To date, most human studies have examined phosphenes elicited by large surface or depth electrodes. These differ considerably from the small penetrating intracortical microelectrodes capable of stimulating highly localized cortical populations that are the basis of current next-generation devices. Here, we provide a detailed characterization of phosphenes evoked by such intracortical microelectrodes. Methods: As part of the CORTIVIS clinical trial, a Utah Electrode Array with 100 penetrating electrodes was implanted in the early visual cortex of four blind volunteers. Participants completed drawing and description tasks in response to single-electrode stimulation. Two participants also performed forced-choice tasks to determine whether phosphenes elicited by different electrodes could be reliably discriminated. Results: Single-electrode stimulation reliably evoked phosphenes in all participants. Most percepts appeared in the contralateral visual field, although two participants reported near-midline ipsilateral percepts. Phosphenes ranged from simple dots to elongated, textured, circular, or otherwise idiosyncratic forms. One participant reported distinguishable phosphene colors, which depended on electrode and amplitude structure and varied across sessions. The two participants tested on forced-choice discrimination could successfully identify electrode-specific phosphenes across multiple sessions. Conclusions: Intracortical microstimulation of the human visual cortex produces phosphenes with diverse, electrode-specific appearances that are recognizable across repeated trials. These findings indicate that penetrating microelectrodes often evoke more structured percepts than simple points or blobs, providing foundational evidence that electrode-specific encoding strategies will be necessary in future visual prosthetic devices.

Purpose: This study aimed to investigate the association between hypertension and anisometropia and determine whether interocular structural asymmetry mediates this relationship. Methods: In this cross-sectional study, nationally representative samples from the US (N = 12,749; NHANES 1999-2008) and Korea (N = 35,484; KNHANES 2008-2024) were analyzed. Associations between hypertension, blood pressure components (BPC), and anisometropia were examined using weighted multivariable logistic regression and restricted cubic splines. Mediation analysis assessed whether asymmetry in optical coherence tomography (OCT) structural parameters mediated the association between BPC and anisometropia. Results: Hypertensive individuals had higher odds of anisometropia in both cohorts (NHANES: OR = 1.32, 95% CI: 1.12-1.56, P = 0.001; KNHANES: OR = 1.12, 95% CI: 1.05-1.20, P < 0.001). Each 10 mmHg increase in systolic blood pressure (SBP) and pulse pressure (PP) was associated with 5% and 10% greater risk, respectively. Interocular differences in central macular and choroidal thickness partially mediated the association between BPCs and anisometropia. Time-course bulk RNA-seq identified Rock1 as a candidate gene downregulated under hypertensive conditions. Conclusions: Hypertension was associated with anisometropia across two national cohorts. Interocular differences in central macular and choroidal thickness partially mediated this association, suggesting a link between systemic hemodynamic alterations and ocular structural asymmetry. Translational Relevance: These findings suggest that interocular OCT structural asymmetry may serve as a clinically accessible imaging biomarker to identify individuals at increased risk of hypertension-associated anisometropia.

Digital photography remains the gold standard for dermatological assessment, but reliance on standardized stationary image-capture systems limits trial scalability, whereas smartphones combined with computer vision artificial intelligence (AI) could enable hybrid and decentralized clinical trials (DCTs). This study assessed the facial feature grading of an AI-based system against expert panel consensus across three settings with decreasing acquisition control: Standard Studio Imaging (SSI), Clinic Smartphone Imaging (CSI), and Home Smartphone Imaging (HSI). Expert inter-rater reliability was good to excellent (median ICC 0.84-0.90), and the system showed high operator-independent test-retest repeatability across independently captured images (median ICC 0.97-0.98). Agreement with experts was maintained across all settings for macro-geometric features (ptosis {rho} = 0.74-0.81, nasolabial fold {rho} = 0.68-0.79) and for discrete pigmentation and inflammatory lesions (pigmented spot density {rho} = 0.62-0.76), declined for fine-texture features under uncontrolled acquisition (forehead wrinkles {rho} = 0.76 to 0.48), and remained limited for color-dependent features irrespective of setting. These findings support the AI-based system as an objective grading tool for a defined subset of facial endpoints in hybrid and DCTs in dermatology.

Ascertaining the pathogenicity and clinical relevance of genetic mutations is a longstanding challenge across human disease. This is particularly relevant for genes with many variants of unknown significance (VUS), such as PRPH2-retinopathy, where complex variable phenotype patterns and limited genotype correlations hinder application of precision-medicine therapies. Recent AI-based pathogenicity tools, while powerful, may be limited by insufficient mechanistic or phenotypic training data. We established a three-dimensional proteoform-phenotype analysis framework by combining AI-generated protein structural models with clinical genomics data, in which expert consensus adjudicates model outputs through a human-in-the-loop AI approach, to evaluate 46 curated pathogenic missense variants against a background VUS pool. The framework was prospectively validated using an independently evaluated clinical patient cohort. This analysis resolved spatial biophysical signatures corresponding to distinct clinical phenotypes (i.e., retinitis pigmentosa, macular dystrophy, and pattern dystrophy) as well as complex blended presentations. Leveraging these confirmed structural signatures, we prioritized high-risk VUS. This mechanism-resolved framework can be extended to other therapeutically actionable genes to identify novel causal variants and improve diagnosis, prognosis, and better select rare, eligible patients for emerging molecular gene therapy trials.

Background and Objective: High-speed and high-resolution retinal imaging using adaptive optics line confocal ophthalmoscopy (AOLCO) resolves individual red blood cells (erythrocytes) flowing through the smallest retinal capillaries, but quantifying their velocity first requires removing linear and nonlinear distortions caused by eye motion that may result in a displacement of tens to hundreds of pixels between frames. We present a computational pipeline that stabilizes AOLCO capillary video and measures erythrocyte velocity in true physical units, delivered as an interactive 3D Slicer extension and a batch command-line tool sharing one analysis core. Methods: Our video analysis pipeline has four stages: preprocessing, registration, postprocessing, and velocity measurement. During preprocessing, illumination is corrected, and closed-eye frames are identified. During registration, each frame is aligned to a reference frame using affine and, when needed, B-Spline registration, initialized by our robust and fast (GPU-based) translation-registration algorithm to accommodate large eye movements (e.g., microsaccades). During postprocessing, adjacent registered frames are differenced to suppress stationary structures, and a 2D projection of the frame-difference video reveals the capillary network. During velocity measurement, an operator manually identifies capillary segments on the 2D projection image. Spatiotemporal images are then generated along these segments from the frame-difference video, and blood-flow velocity curves are obtained from these images using the Radon transform. Results: We validate the velocimetry of the pipeline against synthetic ground-truth flow data, and the experiments show that the velocimetry recovers the speed within an error of 4% The GPU-based translation registration algorithm aligns the large eye motions (~90 um per frame on average) and runs about 3.6 x (100 Hz) to 38 x (400 Hz) faster than the baseline method. Across the full cohort (2344 videos, 57 subjects, 30 Hz to 400 Hz), our registration approach improves structural similarity (SSIM) of the video frames on every acquisition from 0.73 to 0.86. The recovered blood flow velocities are physiologically plausible, of the same order as published AO measurements. Conclusions: Our computational pipeline enables AOLCO velocimetry on the capillary-level.

Gene therapy can restore retinal signals absent since birth in animal models, yet the translation to humans remains challenging. An unresolved possibility is that meaningful retinal rescue already occurs in humans, but its impact is masked by downstream neural bottlenecks shaped by long-term deprivation. Tracing restored signals through the visual system is therefore essential to distinguish rescue and constraints. We address this challenge in CNGA3-associated achromatopsia, in which retinal cone photoreceptors are present but non-functional from birth. We used tests specifically targeting cone-mediated processing to trace signals from genetically modified cones from the retina to visual cortex and perception. After treatment in late childhood, we find clear evidence of new cone signals reaching visual cortex in 4 out of 7 patients. These signals were replicable over years, retinotopically organised, available for conscious perception and showed higher temporal resolution. Recovery of foveal acuity and canonical colour vision remained limited, consistent with neuroplasticity constraints on specialised neurocircuitry beyond critical developmental windows. Together, these findings show that the human brain can incorporate a new sensory signal after years of complete congenital deprivation, while revealing selective bottlenecks that limit full visual recovery. They provide a framework for optimising sensory restoration: measure rescued input, trace downstream integration of signal, and tailor treatment timing and rehabilitation to maximise functional gain.

While lysosomal lipid dysregulation is implicated in Parkinsons disease (PD), identifying the precise enzymatic drivers remains limited by the lack of subcellularly resolved metabolic mapping. Leveraging endosome-origin plasma extracellular vesicle (PsEV) lipidomics, we identified a profound bidirectional disruption of the ceramide-sphingomyelin (SM) axis in PD, driven by asymmetric upregulation of sphingomyelin phosphodiesterase 1 (SMPD1) and sphingomyelin synthase 1 (SGMS1). BODIPY-C5 SM tracking revealed that SMPD1 perturbation severely impairs lysosomal SM turnover. Crucially, this SMPD1-driven hydrolysis forces a salvage pathway collapse, bottlenecked by SPHK1/SGPL1 downregulation, which precipitates oxidative membrane damage and PINK1/Parkin-dependent mitophagy failure. We robustly validated this multi-pathway dysfunction across orthogonal in vitro, in vivo mouse, and postmortem human brain models. Furthermore, SMPD1/SGMS1 expression distinctly stratifies clinical phenotypes, marking the aggressive Postural Instability/Gait Difficulty (PIGD) subtype. Our findings establish SMPD1-mediated lysosomal impairment and ceramide-SM imbalance as a core biochemical axis in PD pathophysiology.

In eukaryotic cells, the positive transcription elongation factor b (P-TEFb) plays a critical role in the transition of RNAPII from the paused state to actively transcribing mode. Although three different cyclin (Cyc) T (CycT1, T2a, and T2b) interact with CDK9 to form functional P-TEFb, a majority of P-TEFb complexes are comprised of the CycT1 and CDK9. To analyze the role of the CycT1 subunit of P-TEFb on cellular transcription, we established HEK293T cells lacking CycT1 proteins (CycT1-KO cells). While no apparent growth defects were observed with CYcT1-KO cells, the level of CDK9 was decreased and no compensatory over-expression of CycT2 was observed, suggesting that the reduced level of functional P-TEFb complexes is sufficient for the normal cell growth. Reporter gene assays indicate that NF-kappaB-dependent transcription induced by PMA was impaired in CycT1-KO cells although the function of NF-kappaB (P65) per se was not affected. On the other hand, AP1-dependent transactivation induced by PMA was unaffected in CycT1-KO cells. Moreover, transcription stimulated by JQ1, a strong P-TEFb inducer, was reduced in CycT1-KO cells. Data of biochemical analysis indicate that P-TEFb with CycT1 (P-TEFb (CDK9:CycT1)) was efficiently released from 7SKsnRNP by JQ1 while P-TEFb (CDK9:CycT2) was not affected by JQ1. Transcriptome analysis indicated that the lack of CycT1 had a minor effect on the steady-state transcription. However, expression of JQ1-dependent genes was severely reduced in CycT1-KO cells. Interestingly, referring to gene expression profiles of 33 different types of cancer revealed that 11 out of 13 top JQ1-dependent genes were also highly upregulated in Pancreatic Adenocarcinoma. From these results, we conclude that although many genes can be regulated by both P-TEFb (CDK9:CycT1) and P-TEFb (CDK9:CycT2), CycT1 plays a critical role in regulating highly inducible genes, which are also aberrantly regulated in a particular type of cancer.