Design and Preclinical Validation of an Eyeglass-Coupled Perilimbal Antenna for Retinal Prostheses With a Biocompatible Graft Technique.

Design and Preclinical Validation of an Eyeglass-Coupled Perilimbal Antenna for Retinal Prostheses With a Biocompatible Graft Technique - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Search Log in Primary site navigation Search Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Transl Vis Sci Technol . 2026 May 22;15(5):20. doi: 10.1167/tvst.15.5.20 Design and Preclinical Validation of an Eyeglass-Coupled Perilimbal Antenna for Retinal Prostheses With a Biocompatible Graft Technique Cheng-Yung Lee 1, 2, 3, 4 , Tso-Ting Lai 2, 3, 4 , Chien-Yu Chiang 5 , Yun-Chen Wang 5 , San-Yuan Lin 5 , Yi-Ju Chiang 5 , Yun-Ta Yang 5 , Chang-Hao Yang 2, 4 , Long-Sheng Fan 5, ✉ Author information Article notes Copyright and License information 1 Department of Ophthalmology, National Taiwan University Hospital Hsin-Chu Branch, Hsinchu City, Taiwan 2 Department of Ophthalmology, National Taiwan University Hospital, Taipei City, Taiwan 3 Graduate Institute of Clinical Medicine, College of Medicine, National Taiwan University, Taipei City, Taiwan 4 Department of Ophthalmology, College of Medicine, National Taiwan University, Taipei City, Taiwan 5 Iridium Medical Technology Company, Hsinchu City, Taiwan * Correspondence: Long-Sheng Fan, Iridium Medical Technology Company, 2F No. 13, R&D Rd. II, Hsinchu Science Park, Hsinchu City 300, Taiwan. e-mail: lsfan@irmedtech.com ✉ Corresponding author. Accepted 2026 Mar 30; Received 2025 Oct 28; Collection date 2026 May. Copyright 2026 The Authors This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. PMC Copyright notice PMCID: PMC13221729  PMID: 42171424 Abstract Purpose To present a novel eyeglass-coupled perilimbal antenna for retinal prostheses and evaluate preclinical biocompatibility and ocular surface effects using a refined, graft-assisted implantation technique. Methods The structural, mechanical, and functional properties of the antenna were assessed, including buffering systems for tissue resilience. Coupling between the eyeglass-rim–mounted emitter and perilimbal receiver was optimized through finite-element electromagnetic simulation and validated in electrolyte solution. A preclinical study was performed in 11 eyes of six Lanyu minipigs, in which antennas were implanted using a refined surgical technique and covered with amniotic membrane, porcine small intestinal submucosa, or pericardium graft. Postoperative evaluations over 6 months included serial external eye photography, anterior segment optical coherence tomography (AS-OCT), tear secretion testing, tear cytokine profiling, and conjunctival impression cytology. Results All eyes showed stable healing without exposure, infection, or granulation. ASOCT revealed no significant change in perilimbal conjunctival thickness (400.5 ± 89.6 µm baseline vs. 399.4 ± 158.2 µm at 6 months; P = 0.612). Coverage materials thinned significantly (360.7 ± 119.5 µm at 1 month vs. 71.4 ± 103.0 µm at 6 months; P = 0.018) without adverse ocular effects. Transient postoperative elevations of tear cytokines (IL-1β, IL-8, IL-6) normalized over time. Tear production, conjunctival redness, and goblet cell density returned to baseline levels. Conclusions The eyeglass-coupled perilimbal antenna enabled efficient transmission with preserved ocular surface integrity. Graft-assisted implantation ensured stability and biocompatibility, serving as a safe interface for retinal prostheses. Translational Relevance A novel perilimbal antenna design enables efficient wireless transmission and biocompatible integration for next-generation retinal prostheses. Keywords: antenna, retinal prosthesis, perilimbal, wireless signal transmission, ocular surface Introduction Retinal prostheses represent an important advance in vision restoration, offering patients with severe visual impairment the possibility of regaining partial function. 1 – 8 These devices typically use an external camera or an integrated image sensor, a signal transmission unit, and a retinal stimulation array to bypass degenerated photoreceptors and, directly or indirectly through bipolar cells, to activate retinal ganglion cells, thereby generating patterned visual perception. Initially designed for advanced retinitis pigmentosa, their use has been extended to other causes of outer retinal degeneration, such as inherited retinal dystrophies and late-stage dry age-related macular degeneration. 4 , 8 Despite these achievements, current prosthetic systems face significant limitations that restrict their clinical utility. High-resolution visual restoration requires not only advances in electrode design but also stable and efficient transmission of external signals and energy for stimulation or signal processing into the eye. 9 – 11 In chip-based devices such as the Argus II (Cortigent, Valencia, CA), orbital placement of the antenna complicates surgery, risks tissue disruption, and hinders long-term stability. Photovoltaic-based systems avoid implanted antennas by using near-infrared light from external glasses projected onto a relatively small intraocular stimulating array, providing central prosthetic vision on top of natural peripheral vision but with a field of view constrained by implant coverage and optical projection geometry. 12 , 13 These antenna-free approaches nonetheless highlight the remaining need for transmission interfaces that combine efficient coupling with long-term ocular biocompatibility. In this study, we present a novel perilimbal antenna system engineered to provide a mechanically compliant interface with the ocular surface while efficiently coupling to external eyewear. Placement in the perilimbal region minimizes orbital tissue disruption and facilitates surgical accessibility. In this study, we evaluated the surgical feasibility, biocompatibility, and functional performance of the system in a longitudinal preclinical model, integrating refined surgical techniques and biological coverage grafts. This work sought to establish a stable, biologically integrated signal transmission platform to support the development of next-generation retinal prostheses. Methods Animals All procedures were approved by the Institutional Animal Care and Use Committee of National Taiwan University and complied with National Institute of Health guidelines and the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Six female Lanyu minipigs (Sus scrofa, >6 months old) contributed 11 eyes: three received perilimbal antenna implantation with double-layer amniotic membrane, two with pericardium graft, two with small intestinal submucosa graft, one underwent sham surgery (peritomy, scleral/muscle exposure, scleral flap, closure), and three served as untreated controls. Perilimbal implantation preserved uveal integrity and visual function. Minipigs were housed individually with limited tactile contact and environmental enrichment. All surgical procedures and ocular examinations, both preoperatively and postoperatively, were performed under general anesthesia. Anesthesia was induced after premedication with atropine (40 µg/kg, intramuscular [IM]), azaperone (0.4 mg/kg, IM), and Zoletil 100 (4.4 mg/kg, IM; Virbac, Carros, France), followed by isoflurane inhalation (2% for induction and 1.5%–3% for maintenance) delivered via endotracheal intubation. Prophylactic cefazolin (20 mg/kg, intravenous [IV]) was administered before surgery. Animals were mechanically ventilated with 30% O 2 at a rate of 10 to 12 breaths/min and a tidal volume of 7 to 8 mL/kg, with continuous physiological monitoring maintained throughout the procedure. At surgery completion, isoflurane was stopped, oxygen raised to 100%, and ventilation adjusted until spontaneous breathing resumed. Animals were extubated, maintained on oxygen until awake, then monitored during recovery. Postoperatively, implanted eyes received topical tetracycline ointment and prednisolone acetate (1%) twice daily (BID) for 7 days. Systemic therapy for 7 days included cephalexin (500 mg BID), diclofenac (25 mg BID), serrapeptase (5 mg/kg BID), and tranexamic acid (250 mg BID). Retinal Prosthesis Manufacture and Preparation The retinal prosthesis used in this study, HARP4K RPS (Iridium Medical Technology Company, Hsinchu City, Taiwan) ( Fig. 1 ), is designed for patients with advanced geographic atrophy. The system has three components: HARP4K implant (with surgical accessories: three carriers and one marking template), HARP4K eyeglasses, and HARP4K fitting console. The implant consists of an elliptical retinal chip (with a packaged size of 4.9 × 4.7 mm) containing photosensors, microelectronic circuitry, and a sputtered iridium oxide film (SIROF) microelectrode array encapsulated in passivation layers. The chip converts incident images into electrical signals to stimulate retinal bipolar cells following subretinal placement. It is connected to a flexible polymer-based substrate (“Flex”) incorporating embedded gold conductors, which also form the secondary inductive coil (paired with a primary coil in the eyeglasses); springs; and fixation suture holes. The Flex is externally encapsulated with a medical-grade silicone elastomer layer to electrically insulate the conductors and to provide long-term biocompatibility with tissues. 14 Device material safety testing is included in the Supplementary Materials ( Supplementary Method , Supplementary Table S1 ). The Flex conforms to the ocular curvature to facilitate stable implantation. Carriers protect the chip and guide insertion into the subretinal space near the fovea. We used a marking template, identical in size and curvature to the implant, preoperatively to verify fit and to mark scleral incision and fixation sites. Figure 1. Open in a new tab Structural design and mechanical properties of the retinal prosthesis. ( a ) A retinal prosthesis designed for advanced geographic atrophy, composed of four primary components: perilimbal antenna system, spring, neck, and subretinal chip. The term “Flex” refers collectively to the perilimbal antenna system, spring, and neck. ( b ) The spring segment incorporates a fixation ear and two mechanical buffers designed to enhance compliance under stretching, compression, and torsional forces. The neck, measuring 0.78 mm in width and composed of two 0.2-mm substalks, provides additional flexibility. ( c ) Schematic comparison illustrating superior multidirectional stress adaptation of the buffer-equipped prosthesis ( pink ) relative to a prosthesis without buffers ( yellow ). ( d ) Surgical implantation via an ab externo approach, with the subretinal chip positioned beneath the retina. Magnified red-free fundus photography demonstrated the chip located beneath the retinal tissue and vasculature. Surgical Procedure for Perilimbal Antenna Implantation Perilimbal antenna implantation was performed via an ab externo approach ( Fig. 2 ). Because this study focused on biocompatibility of the perilimbal antennae system and surgical effects of biomaterial coverage, only antenna placement was carried out. After lateral canthotomy and 360° peritomy, all rectus muscles and the superior oblique muscle were identified and secured with 2-0 silk traction sutures. A scleral flap was created at the excision site to expose the uvea. The retinal prosthesis was transected at the neck, leaving the perilimbal antenna intact. The circular antenna was positioned and secured with 9-0 nylon, and the transected neck was inserted beneath the scleral flap and fixed with 9-0 nylon without uveal penetration. Figure 2. Open in a new tab Implantation of the perilimbal antenna system using a modified double-layer amniotic membrane (AM) coverage technique. ( a ) Preparation of a double-layer AM graft in a square configuration with a central circular excision to expose the cornea. The excised circular AM graft was preserved for subsequent coverage of the scleral flap and the flexible neck of the retinal prosthesis. ( b ) Following prosthesis implantation, the antenna system was secured in the perilimbal region with non-absorbable 9-0 nylon sutures. The AM grafts were positioned to cover the antenna and its extensions, then fixed to the sclera using absorbable 8-0 Vicryl sutures, with peritomy closure performed using the same suture. ( c – e ) Intraoperative images: initial placement and fixation of the perilimbal circular antenna ( c ); positioning of the central-hollowed double-layer AM graft over the antenna and placement of the excised circular graft over the scleral flap, with both secured by sutures ( d ); and final peritomy closure overlying the grafts ( e ). Coverage materials included amniotic membrane (AMNIOAID, HCT Regenerative, New Taipei City, Taiwan), porcine pericardium (ReadiGraft; LifeNet Health, Virginia Beach, VA), and porcine small intestinal submucosa (SIS; BioSIS ECM; Vetrix, Cumming, GA). The amniotic membrane was applied as a double layer for reinforcement. Each material was prepared by excising a central circular portion from square sheets; the circular graft was placed over the scleral flap, and the hollowed square graft was draped over the antenna. Both were secured with interrupted 8-0 Vicryl sutures (Ethicon, Raritan, NJ), followed by peritomy closure with the same material. External Eye Photography, Fluorescein Staining, and Cobalt Blue Light Examination External eye examinations were performed using an operating microscope video system to evaluate for antenna exposure, wound dehiscence, melting, infiltration, perforation, and infection. Findings were documented when present. Fluorescein staining was conducted by instilling 1% fluorescein sodium onto the palpebral conjunctiva which was distributed by blinking. The ocular surface was examined under cobalt blue illumination with a binocular indirect ophthalmoscope (SIGMA 250 LED; HEINE Optotechnik, Gilching, Germany), and photographs were taken to document conjunctival or corneal epithelial defects. Evaluation of Conjunctival Injection Bulbar conjunctival injection was used as a marker of ocular surface inflammation. External eye photographs were obtained at baseline; on postoperative days 3, 6, 10, 14, and 28; and monthly thereafter for 6 months. Images were preprocessed with inverse gamma correction, non-uniform illumination correction, white balance adjustment, and standard gamma correction. 15 Quantification was performed in ImageJ (National Institutes of Health, Bethesda, MD). A region of interest (ROI) was selected in each image, and red, green, and blue (RGB) channels were converted to the CIELAB color space, or L*a*b*, where color L* denotes lightness (0–100), a* the green–red axis (–60 to +60), and b* the blue–yellow axis (–60 to +60). 16 Mean L* and a* values within the ROI were measured, 17 and the erythema index was calculated as a*/L*. Anterior Segment Spectral-Domain Optical Coherence Tomography Anterior segment imaging was performed preoperatively and monthly using a spectral-domain optical coherence tomography (OCT) system (SPECTRALIS; Heidelberg Engineering, Heidelberg, Germany) with a corneal anterior module. Pachymetry mode was applied to capture limbal-centered images, defined by the first hyporeflective corneal boundary. Raster scans perpendicular to the limbus were acquired in four quadrants (superior, inferior, nasal, and temporal) with gentle ocular manipulation. At least three scans per quadrant were obtained; scans lacking a visible scleral spur were excluded. Measurements were averaged across quadrants. Coverage material thickness and conjunctival thickness overlying the antenna were quantified with the anterior segment OCT (AS-OCT) software (Heidelberg Eye Explorer). Perilimbal conjunctival curvature was measured by a four-step manual method: (1) the limbus point was defined as the first corneal boundary, (2) a 1000-µm circle was centered on this point, (3) the conjunctival surface between the limbus and circle intersection was delineated, and (4) the average radius of curvature was calculated using the ImageJ Fiji Kappa plugin. Non-Anesthetic Schirmer Test for Tear Secretion The Schirmer test was performed without topical anesthesia at baseline; on postoperative days 3, 6, 14, 21, and 28; and at 6 months. A Schirmer strip (Omni Schirmer Tear Test Strips; Omni Lens Pvt. Ltd., Gujarat, India) was folded and inserted into the lower conjunctival fornix between the palpebral and bulbar conjunctiva. The eyelid was gently closed for 5 minutes, after which the length of the moistened area was measured in millimeters using the manufacturer's scale. Strips were immediately frozen on dry ice and stored at −80°C for subsequent protein extraction and analysis. Tear Protein Extraction Tear proteins were extracted from Schirmer strips (5 mm) following a modified protocol. 18 Strips were cut, incubated in 150-µL phosphate-buffered saline (PBS) on ice with shaking for 1.5 hours, and centrifuged at 12,000 rpm at 4°C for 10 minutes. Residual fragments were processed using the “piggyback method” (23-gauge needle–pierced tube nested in a 1.5-mL tube) and centrifuged at 13,000 rpm at 4°C for 3 minutes. Protein concentrations were measured spectrophotometrically (NanoVue Plus; GE HealthCare, Chicago, IL), and samples stored at –80°C. Cytokine Array of Tear Fluid Tear cytokines were measured in experimental and control eyes at baseline and at postoperative days 3, 6, 10, 14, and 28 using a porcine cytokine antibody array (10 targets; Abcam, Cambridge, UK). Arrays were processed per the manufacturer's instructions: blocking, incubation with diluted tear proteins (100 µL), and detection with biotinylated antibodies and Cy3-equivalent streptavidin, followed by washing, drying, and scanning. Cytokine concentrations were quantified against slide-based standard curves. Conjunctival Impression Cytology Impression cytology was performed using cellulose acetate filters (25 mm and 0.22 µm; MilliporeSigma; Burlington, MA) cut into quadrants and marked for orientation. After topical anesthesia and removal of excess fluid, filters were applied to the bulbar conjunctiva across the limbus for 5 to 10 seconds with gentle digital pressure, then removed with blunt forceps and fixed in 75% ethanol for 5 minutes at room temperature. Filters were placed epithelial side up and stained with periodic acid–Schiff (PAS) and hematoxylin (PAS-2-IFU; ScyTek Laboratories, Logan, UT). Samples were treated with periodic acid for 10 minutes, rinsed four times with double-distilled water, incubated with Schiff's reagent for 30 minutes, and washed under running tap water. Sections were then counterstained with Mayer's hematoxylin, dehydrated, cleared, and mounted in xylene and synthetic resin. Goblet cell mucin stained pink (PAS), and nuclei appeared blue (hematoxylin). Statistical Analysis and Graphical Representation Statistical analyses were performed using SPSS Statistics 19 (IBM, Chicago, IL). Graphs were generated in Excel (Office LTSC Professional Plus 2024; Microsoft Corp., Redmond, WA). Data were log-transformed when necessary to achieve normality, assessed by the Shapiro–Wilk test. Depending on distribution, parametric or nonparametric tests were applied. Results are presented as mean ± SD, and P < 0.05 was considered statistically significant. Results Design of Perilimbal Antenna and Radiofrequency Properties of the Coupled System A three-part mechanical buffer was integrated into the neck and springs of the retinal prosthesis to improve compliance and accommodate ocular movement ( Figs. 1 b, 1 c). The perilimbal antenna functioned as the secondary receiving coil, capturing inductive power from the primary transmitting coil embedded in the eyeglass frame ( Figs. 3 a–c). This power, up to 20 mW, was sufficient to drive the complementary metal-oxide-semiconductor (CMOS) subretinal chip, which converts visual input into spatiotemporal electrical patterns to stimulate retinal neurons ( Fig. 1 d). Anatomical constraints required the secondary coil to be implanted circumferentially around the limbus beneath the conjunctiva. The secondary coil was designed to provide strong electromagnetic coupling, biocompatibility, mechanical durability, and surgical feasibility while avoiding bulky components that could irritate ocular tissues. Fabricated on a thin flexible substrate with gold traces embedded in a dielectric layer, the coil had an overall thickness of approximately 40 µm. Figure 3. Open in a new tab Design and schematic representation of the eyeglasses-coupled perilimbal antenna system. ( a ) Schematic illustration of the wireless signal transmission pathway. Eyeglasses-mounted emission antennae (blue ring) couple with the perilimbally implanted receiving antenna ( yellow ring ), which relays the signal to the subretinal chip for retinal stimulation. ( b ) Spatial configuration of the coupled antenna system. The emission antenna is modeled as a 15-mm radius tab spiral coil (r1), and the receiving antenna as a 6-mm radius tab spiral coil (r2). Both coils are arranged coaxially (or nearly coaxially) with a separation distance ( z ) of 12 mm, corresponding to the vertex distance between the eyeglass lens and the corneal surface. ( c ) First-order equivalent circuit model of the coupled system. Parameters of the primary ( pale blue box ) and secondary ( yellow box ) circuits are denoted by subscripts “p” and “s,” respectively. R, L, and C represent resistance, inductance, and capacitance, respectively, and the inductors correspond to the transmission coils. RL denotes the load resistance of the retinal chip, Rac represents additional resistance under alternating current, and M denotes mutual inductance. An input voltage ( V in ) applied to the primary circuit generates a current ( I p ) encoding spatial information, which is transferred via inductive coupling to the secondary circuit, inducing an output voltage ( V out ) and current ( I s ). The mathematical relationship between V in and V out is provided in Equation 1 . ( d ) Experimental setup for impedance measurements of the secondary coil, which was fully immersed in saline and connected to a network analyzer (E5061B). Because image processing was integrated into the subretinal chip, video transmission between the external camera and implant was unnecessary. This allowed use of a low-frequency radiofrequency (RF) link for low-bandwidth, bidirectional communication. A proprietary amplitude-shift keying/load-shift keying (ASK/LSK)–based protocol, rather than standard Wi-Fi or Bluetooth bands, was adopted to enhance security and minimize interference. Finite element method (FEM) simulations were used to optimize the spiral secondary coil and its coupling with the irregularly shaped primary coil. A first-order equivalent circuit model was developed ( Fig. 3 ), and the voltage transfer function between primary coils ( V in ) and secondary coils ( V out ) was expressed as V o u t V i n = ω 2 ω 0 2 k L s / L p k 2 + 1 Q p Q s - 1 - ω 2 ω 0 2 + j 1 - ω 2 ω o 2 1 Q p - 1 Q s (1) where subscripts p and s denote the primary and secondary antenna, respectively; L is the inductance; Q is the quality factor ( Q P = ω L p / R p , Q s = ω L s /[ R s + R L ]), where R represents the resistance, R p and R s are the general resistance of primary antennae circuit and secondary antennae circuit, respectively, and R L is the load (or the resistance) of the retinal chip; ω is the operating frequency; ω 0 is the resonant angular frequency; k is the coupling coefficient; and j denotes the imaginary unit ( - 1 ). At resonance, the impedance becomes purely resistive, and maximum power transfer is achieved when k c = 1 Q p Q s . Physiological factors were incorporated into simulations. Immersion of the secondary coil in electrolytic fluids reduced the resonant frequency and quality factor due to parasitic capacitances. Additional corrections accounted for proximity effects, RF skin effects, and radiation losses. To validate these results, network analyzer testing (E5061B; Keysight Technologies, Santa Rosa, CA) was performed with the secondary coil immersed in saline ( Fig. 3 d). With coils separated by >2 cm and tuned below 10 MHz, the inductive link achieved approximately 60% coupling efficiency. Detailed engineering aspects of the device will be reported separately. Preclinical Study of Perilimbal Antenna Implantation With Coverage Materials In the preclinical study, none of the eyes that underwent implantation surgery exhibited device exposure, conjunctival melting, infectious complications, or granulation tissue related to impaired conjunctival healing during the 6-month postoperative follow-up. The ocular surface remained stable and intact in all animals ( Table 1 ). In eyes receiving perilimbal antenna implantation with coverage material, mean perilimbal conjunctival thickness measured 444.7 ± 91.4 µm at postoperative month 1 (POM1) and 399.4 ± 158.2 µm at POM6, with no significant difference from the preoperative baseline of 400.5 ± 89.6 µm ( P = 0.398 and P = 0.612, respectively) ( Table 1 ). Longitudinal analysis likewise showed no significant postoperative change ( P = 0.196) ( Table 2 ). Differences among coverage materials were minimal ( Table 2 ; Supplementary Table S2 ). Table 1. Baseline and Postoperative Changes in Ocular Surface Measurements After Perilimbal Antenna Implantation With Coverage Materials Graft Material Ocular Surface Measurement Cases With Coverage Materials Sham † Control Case number, n 7 1 3 Perilimbal conjunctiva thickness (µm), mean ± SD Baseline 400.5 ± 89.6 430.1 342.4 ± 67.9 POM1 444.7 ± 91.4 274.1 — POM6 399.4 ± 158.2 264.3 — Statistical comparison of POM1 to baseline a P = 0.398 — — Statistical comparison of POM6 to POM1 a P = 0.310 — — Statistical comparison of POM6 to baseline a P = 0.612 — — Coverage material thickness (µm), mean ± SD POM1 360.7 ± 119.5 — — POM6 71.4 ± 103.0 — — Statistical comparison of final to POM1 a P = 0.018 * — — Perilimbal conjunctival radius of curvature (mm), mean ± SD Baseline 9.29 ± 3.83 8.87 16.01 ± 10.06 POM1 2.36 ± 0.81 8.51 — POM3 6.07 ± 3.53 8.91 — POM6 8.02 ± 3.73 9.23 — Statistical comparison of POM1 to baseline a P = 0.018 * — — Statistical comparison of POM6 to POM1 a P = 0.018 * — — Statistical comparison of POM6 to baseline P = 0.237 — — Conjunctiva redness, mean ± SD Baseline 0.16 ± 0.12 0.17 0.24 ± 0.11 POD3 0.43 ± 0.09 0.41 — POM1 0.24 ± 0.07 0.17 — POM3 0.29 ± 0.13 0.16 — POM6 0.23 (0.08) 0.15 — Statistical comparison of POD3 to baseline a P = 0.018 * — — Statistical comparison of POM6 to POD3 a P = 0.018 * — — Statistical comparison of POM6 to baseline a P = 0.176 — — Non-anesthetic Schirmer's test (mm/5 min), mean ± SD Baseline 1.68 ± 0.64 0.94 1.23 ± 0.51 POD3 2.75 ± 0.36 2.33 — POD14 1.49 ± 0.46 1.83 — POM1 0.93 ± 0.55 0.91 — POM6 0.94 (0.53) 2.17 — Statistical comparison of POD3 to baseline a P = 0.028 * — — Statistical comparison of POM6 to POD3 a P = 0.017 * — — Statistical comparison of POM6 to baseline a P = 0.128 — — Goblet cell density from impression cytology at or after b POM6 (goblet cells/100 epithelial cells), mean ± SD 6.04 ± 1.53 6.39 6.30 ± 1.30 External eye conditions at POM6, n Prosthesis exposure 0 — — Corneal or conjunctival epithelial defect 0 0 0 Corneal melting or infiltration 0 0 0 Any other clinically significant complications c 0 0 0 Open in a new tab To avoid overwhelming with unnecessary data, only selected time points for each measurement factor are listed in this table to concisely illustrate the longitudinal trend. In contrast, the longitudinal analysis in Table 2 incorporates data from all time points. Similarly, Figure 5 presents the longitudinal data from all time points using line charts. † Only one case underwent sham operation without prosthesis and coverage material implantation. The statistical analysis was thus not conducted due its single case condition. a The pair test for each factor was conducted using non-parametric, Wilcoxon signed-rank test. b Goblet cell density was considered stabilized before POM6. Impression cytology tests were performed at or after POM6. c Any other clinically significant complications are defined as any visual or structural compromising conditions including but not limited to infectious endophthalmitis, scleral melting, and graft exposure. * P < 0.05. Table 2. Longitudinal Analyses Using Linear Mixed Effect Model on Postoperative Measurements, and Subgroup Analyses Among Groups With Different Coverage Materials 95% CI b Postoperative Measurement Comparisons a Estimate b , c SD b P b Lower Upper P cm b P Time b Conjunctiva thickness (µm) All cases — — — — — — 0.196 Subgroup comparison: SIS to AM 29.27 21.05 0.207 −20.52 79.06 0.212 0.266 Subgroup comparison: pericardium to AM −22.44 22.48 0.359 −78.36 33.47 — — Coverage material thickness (µm) All cases — — — — — — 0.002 * Subgroup comparison: SIS to AM 61.85 31.20 0.060 −2.92 126.62 0.153 0.004 * Subgroup comparison: pericardium to AM 31.73 29.58 0.299 −30.79 94.26 — — Perilimbal conjunctival radius of curvature (mm) All cases — — — — — — 0.011 * Subgroup comparison: SIS to AM −3.86 0.92 <0.001 ** −5.72 −1.99 <0.001 ** 0.001 * Subgroup comparison: pericardium to AM 0.19 0.92 0.837 −1.67 2.05 — — Conjunctiva redness All cases — — — — — — <0.001 ** Subgroup comparison: SIS to AM 0.04 0.04 0.233 −0.03 0.12 0.460 <0.001 ** Subgroup comparison: pericardium to AM 0.03 0.04 0.427 −0.04 0.10 — — Non-anesthetic Schirmer's test All cases — — — — — — <0.001 ** Subgroup comparison: SIS to AM −0.16 0.17 0.342 −0.50 0.18 0.608 <0.001 ** Subgroup comparison: pericardium to AM −0.02 0.17 0.902 −0.36 0.32 — — Open in a new tab AM, amniotic membrane; CI, confidence interval; SD, standard deviation; SIS, small intestinal submucosa. a Seven cases were included in the analyses of this table; three cases received AM coverage, two cases received SIS coverage, and two cases received pericardium coverage. b Longitudinal changes of several postoperative measurements were analyzed using linear mixed-effects models with unstructured method. The repeat measurements were conducted monthly postoperatively until the 6th month. Postoperative thicknesses were collected through POD3 to POM6. The preoperative baseline data were not included in this analysis. Each fixed-effect variable was assessed individually using the model Y = β 0 + β 1 × ( fixed – effect variable ) + β 2 × time + ε, where Y is the dependent variable (thickness of either graft or conjunctiva), β 0 is the intercept, β 1 is the estimate of the fixed-effect variable (different coverage materials), β 2 is the estimate for time, and ε is the residual error. Interactions with time were not modeled, assuming independence. The primary outcomes reported are the P value and 95% CI for β 1 . The P value for β 2 is shown as P T ime and reflects the time effect. The P value of overall effect of β 1 is indicated as P cm . c For subgroup analysis, the group using amniotic membrane was used as the reference group. The estimates were the estimated differences of measurement between cases with other coverage materials and that of cases using amniotic membrane (reference group). * P < 0.05. ** P < 0.001. In contrast, all implanted coverage materials exhibited progressive thinning, decreasing from 360.7 ± 119.5 µm at POM1 to 71.4 ± 103.0 µm at POM6. Linear mixed-effects analysis confirmed a significant reduction over time ( P = 0.002) ( Table 2 ). Subgroup analyses by material type did not reach statistical significance ( Supplementary Table S2 ). As an indicator of perilimbal conjunctival bulging, the mean radius of curvature significantly decreased at POM1 (2.36 ± 0.81 mm) compared with baseline (9.29 ± 3.83 mm; P = 0.018) ( Table 1 ). The curvature gradually normalized over time, with the mean radius increasing to 8.02 ± 3.73 mm by POM6 ( P = 0.018; longitudinal analysis P = 0.011) ( Tables 1, 2 ). Tear cytokine analysis revealed a transient postoperative inflammatory response ( Table 3 ). Among the evaluated cytokines, IL-1β and IL-8 demonstrated the most pronounced early postoperative elevations, with subsequent normalization within approximately 2 weeks. IL-1β levels rose significantly from baseline to postoperative day 3 (POD3) ( P = 0.005; 95% confidence interval [CI], 2.8–7.2) and decreased from POD3 to POD28 ( P = 0.013; 95% CI, −4.0 to −0.99). Similarly, IL-8 levels increased sharply from baseline to POD3 ( P < 0.001; 95% CI, 6.2–7.4) and declined from POD3 to POD28 ( P < 0.001; 95% CI, −8.5 to −5.4). IL-6 also exhibited a significant postoperative rise (POD3 to baseline: P = 0.036; 95% CI, 0.37–5.6) with normalization by POD14 ( P = 0.028; 95% CI, −3.0 to −0.35); however, the reduction by POD28 was not statistically significant compared with POD3 ( P = 0.369; 95% CI, −6.7 to 3.4). Table 3. Analyses on Baseline and Postoperative Changes of Various Ocular Surface Cytokines d , e ELISA (SD) P (95% CI) Cytokine Baseline POD3 POD14 POD28 POD3–Baseline a , b POD14–POD3 a , b POD28–POD3 a , b POD28–Baseline a , b Postoperative Longitudinal Analysis c IL-1β 7.21 (9.88) 662.58 (388.66) 42.72 (8.81) 53.26 (31.23) 0.005 * (2.8, 7.2) 0.009 * (−3.9, −1.2) 0.013 * (−4.0, −0.99) 0.077 (−0.51, 5.6) 0.031 * IL-4 40.00 (51.65) 70.74 (73.07) 46.27 (81.71) 32.07 (32.50) 0.462 (−3.2, 5.4) 0.285 (−4.5, 1.9) 0.651 (−5.0, 3.7) 0.827 (−5.6, 6.5) 0.896 IL-6 0.00 (0.00) 49.87 (75.75) 15.98 (30.69) 9.60 (16.29) 0.036 * (0.37, 5.6) 0.028 * (−3.0, −0.35) 0.369 (−6.7, 3.4) 0.218 (−1.4, 4.0) 0.518 IL-8 0.82 (0.83) 1590.34 (699.49) 32.16 (50.90) 0.77 (1.53) <0.001 ** (6.2, 7.4) 0.004 * (−6.6, −2.8) 0.001 * (−8.5, −5.4) 0.668 (−1.3, 0.98) <0.001 ** IL-10 0.00 (0.00) 108.55 (169.12) 5.17 (10.34) 50.97 (58.88) 0.189 (−2.2, 7.4) 1.97 (−5.2, 1.7) 0.917 (−7.1, 6.6) 0.182 (−1.9, 6.6) 0.381 IL-12 1693.22 (735.23) 5475.49 (5165.18) 553.41 (421.48) 2584.56 (1533.94) 0.474 (−2.0, 3.4) 0.208 (−9.3, 3.1) 0.667 (−2.6, 2.0) 0.372 (−0.72, 1.4) 0.101 GM-CSF 14.84 (23.62) 10.45 (12.23) 72.85 (88.63) 152.01 (197.77) 0.977 (−3.6, 3.5) 0.615 (−4.4, 6.3) 0.646 (−6.8, 9.3) 0.585 (−5.3, 7.8) 0.733 IFN-γ 0.00 (0.00) 1.80 (3.61) 7.03 (14.07) 19.48 (32.52) 0.391 (−1.1, 2.2) 0.391 (−0.69, 1.3) 0.473 (−3.3, 5.5) 0.205 (−1.6, 4.9) 0.364 TGF-β1 414.22 (319.41) 2378.78 (3305.52) 1627.48 (2608.56) 797.22 (718.14) 0.762 (−8.8, 11) 0.618 (−12, 8.8) 0.866 (−11, 9.5) 0.304 (−0.71, 1.6) 0.894 TNF-α 221.99 (269.90) 186.91 (373.81) 34.95 (69.90) 141.90 (51.81) 0.679 (−11, 8.3) 0.871 (−8.0, 7.1) 0.133 (−1.8, 8.3) 0.403 (−4.2, 7.9) 0.012 * Open in a new tab POD, post operative day. a All ELISA values were log-transformed prior to statistical analysis to normalize the dataset, using the formula ln( X + 1), where X represents the original ELISA value for each cytokine and Y is the transformed value. Normality was assessed using the Shapiro–Wilk test for each dataset, particularly for those involved in statistically significant comparisons; datasets with P < 0.1 were considered not normally distributed. For comparisons that achieved statistical significance but did not meet normality assumptions, non-parametric tests were performed and the results reported. b The pair tests for each cytokine between two given postoperative days were conducted using paired t -tests. c Longitudinal changes in cytokine levels were analyzed using linear mixed-effects models with a compound symmetry correlation structure. Repeated measurements for each cytokine were taken on postoperative days 3, 6, 10, 14, and 28. Postoperative cytokine measurements were collected from POD3 through POM6. Preoperative baseline data were not included in this analysis. Each fixed-effect variable was evaluated individually using the model: Y = β 0 + β 1 × time + ε, where Y is the dependent variable (thickness of either graft or conjunctiva), β 0 is the intercept, β 1 is the estimate for time, and ε is the residual error. Interactions with time were not modeled, assuming independence. The primary outcomes reported are the P values that indicate whether cytokines changes over time achieved statistical significance. d Four cases underwent cytokine sampling: Two cases received amniotic membrane coverage, one case received pericardium coverage, and one case received small intestine submucosa. e To avoid overwhelming with unnecessary data, only selected time points for each cytokine are presented in this table to concisely illustrate the longitudinal trend. In contrast, the longitudinal analysis in the last column incorporates data from all time points. Similarly, Figure 5 presents longitudinal data from all time points using line charts. * P < 0.05. ** P < 0.001. Conjunctival redness, used as a proxy for ocular surface irritation, increased significantly from 0.16 ± 0.12 preoperatively to 0.43 ± 0.09 at POD3 ( P = 0.018) ( Table 1 ). Redness then gradually subsided at POM6 (longitudinal analysis P = 0.018) ( Table 1 ). Intergroup differences were minimal ( Table 2 ). Non-anesthetic Schirmer's test values, representing functional irritation, rose from 1.68 ± 0.64 mm/5 min preoperatively to 2.75 ± 0.36 mm/5 min at POD3 ( P = 0.028), followed by a decline to baseline levels at POM6 (0.94 ± 0.53 mm/5 min; P = 0.017). Eyes implanted with SIS grafts demonstrated consistently higher tear production trends, although without statistical significance ( Supplementary Table S2 ; Fig. 5 ). Figure 5. Open in a new tab Longitudinal postoperative changes in ocular surface parameters and tear cytokine expression. ( a – e ) Line graphs depicting five ocular surface parameters. Dashed horizontal lines indicate mean preoperative baseline values. Pairwise comparisons within the “all cases with coverage material” group are shown with horizontal bars . ( Upper left ) Baseline versus first postoperative time point (POD3 or POM1). ( Upper right ) First postoperative time point versus POM6; second row shows start versus end of line. * P < 0.05, ** P < 0.001; ns, not significant. Standard deviations are shown only for the “all cases with coverage material” group. ( a ) Perilimbal conjunctival thickness remained stable with no significant postoperative change. ( b ) Coverage material thickness decreased progressively, reflecting graft resorption. ( c ) Perilimbal conjunctival curvature radius decreased transiently after implantation (bulging) and gradually returned to baseline. ( d ) Conjunctival redness increased significantly in the early postoperative period but resolved to baseline by 6 months. ( e ) Non-anesthetic Schirmer's test demonstrated a transient postoperative increase in tear secretion that normalized by 6 months. ( f – o ) Tear cytokine profiles. IL-1β, IL-6, and IL-8 exhibited significant early postoperative surges followed by a gradual return to baseline levels. Corneal evaluation using fluorescein staining demonstrated an intact corneal epithelium without defects across all groups beyond POM6 ( Fig. 4 d). Similarly, conjunctival impression cytology performed at or beyond POM6 revealed similar goblet cell densities across all groups (6.04 ± 1.53 goblet cells/100 epithelial cells; P = 0.928) ( Table 1 ), with no significant difference compared to controls (6.30 ± 1.30 goblet cells/100 epithelial cells; P = 0.881) ( Table 1 ). Figure 4. Open in a new tab Postoperative ocular surface evaluation following perilimbal antenna implantation in minipigs. ( a ) Serial external eye photographs showing progressive resolution of conjunctival edema and hyperemia after implantation with double-layer amniotic membrane coverage. By the final observation point, bulbar conjunctival injection had resolved, and the subconjunctival perilimbal antennae remained stable without exposure, displacement, or excessive tissue reaction. ( b ) Serial AS-OCT images demonstrate the reduction of conjunctival edema and gradual resorption of the amniotic membrane graft, with restoration of normal conjunctival curvature. White double-headed arrows denote conjunctival thickness overlying the antennae; yellow double-headed arrows represent graft thickness; asterisks mark the antennae; and white arrowheads outline graft borders. ( c ) External eye comparisons at POM6 across eyes with different coverage materials relative to baseline, showing no evidence of device exposure or major ocular surface complications. ( d ) Fluorescein staining at ≥6 months postoperatively demonstrating intact corneal epithelium without epithelial defects in all eyes. Discussion This study introduces a novel, accessible, and efficient antenna model for retinal prosthesis and demonstrated its biocompatibility in a minipig preclinical study using a refined surgical technique with coverage graft materials. Existing approaches highlight the challenges of antenna placement. The Argus II chip–based system placed its antenna in the orbital region, but maintaining stability in delicate ocular tissue has proved difficult. 6 Photovoltaic-based devices bypass antennas by transmitting signals via light stimulation. 4 , 11 Here, our proposed perilimbal antenna offers an alternative: its external placement simplifies implantation, improves accessibility for maintenance, and minimizes orbital disruption. Mechanical compliance is critical for neural prosthetics, as rigid devices risk chronic inflammation or microtrauma. 19 , 20 Because retinal prostheses mechanically couple extraocular structures to the neuroretina, incorporating buffers is essential to reduce injury during ocular movements and to prevent migration of the stimulating pad ( Figs. 1 b, 1 c). Subconjunctival implants, however, are vulnerable to conjunctival thinning, erosion, and exposure, as seen with anti-glaucoma devices. 21 , 22 To address this, coverage grafts are often used. 23 – 25 Early experiments showed that perilimbal antenna implantation without grafts increased wound dehiscence and exposure risk. Refinement of the surgical technique, including coverage grafts, careful bare sclera preparation, advancement of Tenon's capsule, and secure circumferential suturing, eliminated exposure and associated complications. Coverage materials have distinct characteristics for ocular surface applications. 24 , 26 Amniotic membrane promotes wound healing, 27 , 28 porcine small intestinal submucosa has been used in animal corneal repair, 29 , 30 and human pericardium has shown safety in Argus II implantation. 31 In this study, amniotic membrane, SIS, and pericardium all maintained surface integrity and prevented device exposure ( Table 1 ; Supplementary Table S2 ). The observed reduction in graft thickness over 6 months suggests that their role is less structural reinforcement than scaffolding for adhesion and conjunctival wound healing. Consistent with this, endpoint histology showed preserved conjunctival architecture adjacent to the implanted antenna ( Supplementary Fig. S1 ), and anterior segment indocyanine green angiography demonstrated perilimbal vascular features comparable between implanted and healthy eyes ( Supplementary Fig. S2 ), supporting long-term tissue compatibility. A potential concern with perilimbal elevation is tear film disturbance leading to corneal dellen, as described after trabeculectomy or subconjunctival silicone oil. 32 – 34 Although rare, such complications may cause perforation 35 and are difficult to manage. 36 In this study, no dellen, epithelial defects, or infections were detected, and imaging confirmed stable conjunctival curvature ( Figs. 4 b, 5 c), indicating that the antenna conforms to the ocular surface without compromising tear dynamics. Tear cytokines influence conjunctival wound healing and serve as biomarkers of ocular surface health. 37 , 38 Proinflammatory mediators such as TNF-α, IL-1β, IL-6, IL-8, IL-12, and IFN-γ are associated with injury, 39 – 44 whereas TGF-β exerts anti-inflammatory effects. 45 Postoperative analysis revealed transient increases in TGF-β and IL-8 with rapid normalization ( Fig. 5 ), consistent with a controlled wound-healing response rather than chronic inflammation. The Schirmer I test assesses both basal and reflex aqueous tear secretion and was influenced by multiple factors. 46 , 47 Despite lower baseline tear production compared with humans, postoperative increases likely reflected transient irritation, and normalization suggested preserved ocular surface integrity and intact sensory pathways. Importantly, long-term dry eye did not develop. Goblet cell density, another marker of ocular surface health, 48 – 50 remained stable at 6 months ( Table 1 ; Fig. 4 e), further indicating an absence of chronic inflammation. As neural prosthetics evolve toward more precise and efficient integration, 51 , 52 a stable and biocompatible interface for retinal stimulation remains a critical challenge. This interface consisted of neuroretina stimulation and signal transmission. The perilimbal antenna system, validated here with a refined surgical approach and graft coverage, offers advantages over orbital placement of antennae systems: reduced surgical trauma, proximity for efficient coupling, and easier revision if needed. Its compatibility with the ocular surface also suggests potential utility in other biomedical implants as devices continue to miniaturize. This study has several limitations. The sample size was relatively small, as the study was designed as a pilot feasibility and safety evaluation; therefore, definitive conclusions regarding differences among graft materials could not be drawn. Future studies with larger cohorts will be required to validate these findings and enable formal comparative analyses. Nevertheless, this proof-of-concept study provides preliminary evidence supporting the feasibility of this approach for retinal prosthesis development. The porcine ocular anatomy may not fully mirror human physiology, particularly given the greater robustness of the porcine corneal epithelium; nonetheless, the absence of adverse events supports the feasibility of translation to human studies. Finally, a perilimbal implant could potentially complicate future ocular surgeries, although appropriate design refinements may help mitigate this risk. In conclusion, this preclinical study demonstrated that the perilimbal antenna system provides a stable, biocompatible, and efficient interface for retinal prostheses. By reducing surgical complexity while preserving ocular surface health, this approach offers a promising foundation for future retinal prosthetic systems and potentially for broader applications in biomedical implantation. Supplementary Material Supplement 1 tvst-15-5-20_s001.docx (3.8MB, docx) Acknowledgments Supported by the 2024 Cross-Disciplinary Advancement Program for Precision Health, funded by the National Science and Technology Council and the Hsinchu Science Park Bureau, Taiwan. Part of this work was previously presented as a poster at the 2025 ARVO Annual Meeting. The Figure 4 a schematics were originally created on September 9, 2025, using materials assisted by ChatGPT-5 (OpenAI, San Francisco, CA) to illustrate the front and oblique eye views with eyeglasses, and they were subsequently refined in PowerPoint (Office LTSC Professional Plus 2024; Microsoft Corp., Redmond, WA) to depict the perilimbal antenna and eyeglass-mounted emitter. Disclosure: C.-Y. Lee , None; T.-T. Lai , None; C.-Y. Chiang , Iridium Medical Technology Company (E); Y.-C. Wang , Iridium Medical Technology Company (E); S.-Y. Lin , Iridium Medical Technology Company (E); Y.-J. Chiang , Iridium Medical Technology Company (E); Y.-T. Yang , Iridium Medical Technology Company (E); C.-H. Yang , Iridium Medical Technology Company (O); L.-S. Fan , Iridium Medical Technology Company (O) References 1. Ayton LN, Barnes N, Dagnelie G, et al.. An update on retinal prostheses. Clin Neurophysiol . 2020; 131(6): 1383–1398. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 2. Hallum LE, Dakin SC. Retinal implantation of electronic vision prostheses to treat retinitis pigmentosa: a systematic review. Transl Vis Sci Technol . 2021; 10(10): 8. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 3. Kuehn BM. Retinal implant. JAMA . 2013; 309(12): 1220. [ Google Scholar ] 4. Muqit MMK, Le Mer Y, Olmos de Koo L, Holz FG, Sahel JA, Palanker D. Prosthetic visual acuity with the PRIMA subretinal microchip in patients with atrophic age-related macular degeneration at 4 years follow-up. Ophthalmol Sci . 2024; 4(5): 100510. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 5. Vagni P, Airaghi Leccardi MJI, Vila C-H, et al.. POLYRETINA restores light responses in vivo in blind Göttingen minipigs. Nat Commun . 2022; 13(1): 3678. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 6. da Cruz L, Dorn JD, Humayun MS, et al.. Five-year safety and performance results from the Argus II retinal prosthesis system clinical trial. Ophthalmology . 2016; 123(10): 2248–2254. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 7. Edwards TL, Cottriall CL, Xue K, et al.. Assessment of the electronic retinal implant Alpha AMS in restoring vision to blind patients with end-stage retinitis pigmentosa. Ophthalmology . 2018; 125(3): 432–443. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 8. Holz FG, Mer YL, Muqit MMK, et al.. Subretinal photovoltaic implant to restore vision in geographic atrophy due to AMD. N Engl J Med . 2026; 394(3): 232–242. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 9. Eckhorn R, Wilms M, Schanze T, et al.. Visual resolution with retinal implants estimated from recordings in cat visual cortex. Vision Res . 2006; 46(17): 2675–2690. [ DOI ] [ PubMed ] [ Google Scholar ] 10. Oh Y, Hong J, Kim J. Retinal prosthesis edge detection (RPED) algorithm: low-power and improved visual acuity strategy for artificial retinal implants. PLoS One . 2024; 19(6): e0305132. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 11. Modak S, Kaim V, Zaidi AM, Kanaujia BK, Rambabu K. Design of electrically small intraocular antenna for retinal prosthesis system and its validation. IEEE Trans Biomed Eng . 2024; 71(12): 3402–3412. [ DOI ] [ PubMed ] [ Google Scholar ] 12. Mathieson K, Loudin J, Goetz G, et al.. Photovoltaic retinal prosthesis with high pixel density. Nat Photonics . 2012; 6(6): 391–397. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 13. Bhuckory MB, Monkongpitukkul N, Shin A, et al.. Enhancing prosthetic vision by upgrade of a subretinal photovoltaic implant in situ. Nat Commun . 2025; 16(1): 2820. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 14. Colas A, Curtis J. Silicone biomaterials: history and chemistry. In: Ratner BD, Hoffman AS, Schoen FJ, Lemons JE, eds. Biomaterials Science—An Introduction to Materials in Medicine . 2nd ed. London: Elsevier; 2004: 80–86. [ Google Scholar ] 15. Bala S, Sirazitdinova E, Deserno T. Colorimetric calibration of wound photography with off-the-shelf devices. In: Proceedings of SPIE 10136, Medical Imaging 2017: Image Perception, Observer Performance, and Technology Assessment. Bellingham, WA: SPIE; 2017. [ Google Scholar ] 16. Stamatas GN, Zmudzka BZ, Kollias N, Beer JZ. Non-invasive measurements of skin pigmentation in situ. Pigment Cell Res . 2004; 17(6): 618–626. [ DOI ] [ PubMed ] [ Google Scholar ] 17. Cugmas B, Viškere D, Štruc E, Olivry T. Evaluation of erythema severity in dermatoscopic images of canine skin: erythema index assessment and image sampling reliability. Sensors (Basel) . 2021; 21(4): 1285. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 18. Gijs M, Arumugam S, van de Sande N, et al.. Pre-analytical sample handling effects on tear fluid protein levels. Sci Rep . 2023; 13(1): 1317. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 19. Zhang H, Patel PR, Xie Z, Swanson SD, Wang X, Kotov NA. Tissue-compliant neural implants from microfabricated carbon nanotube multilayer composite. ACS Nano . 2013; 7(9): 7619–7629. [ DOI ] [ PubMed ] [ Google Scholar ] 20. Nguyen JK, Park DJ, Skousen JL, et al.. Mechanically-compliant intracortical implants reduce the neuroinflammatory response. J Neural Eng . 2014; 11(5): 056014. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Buffault J, Baudouin C, Labbé A. Is the Xen Gel Stent really minimally invasive? Am J Ophthalmol Case Rep . 2020; 19: 100850. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Nardi M, Maglionico MN, Posarelli C, Figus M. Managing Ahmed Glaucoma Valve tube exposure: surgical technique. Eur J Ophthalmol . 2021; 31(2): 778–781. [ DOI ] [ PubMed ] [ Google Scholar ] 23. Riva I, Roberti G, Oddone F, Konstas AG, Quaranta L. Ahmed glaucoma valve implant: surgical technique and complications. Clin Ophthalmol . 2017; 11: 357–367. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. Thakur S, Ichhpujani P, Kumar S. Grafts in glaucoma surgery: a review of the literature. Asia Pac J Ophthalmol (Phila) . 2017; 6(5): 469–476. [ DOI ] [ PubMed ] [ Google Scholar ] 25. Ebert JJ, Shildkrot YE, Menezes A, Miller DM. Modified split-thickness corneal patch graft for conjunctival erosions in patients with the port delivery system implant: surgical technique and long-term outcomes. J Vitreoretin Dis . 2023; 7(5): 420–423. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 26. Feng Y, Borrelli M, Reichl S, Schrader S, Geerling G. Review of alternative carrier materials for ocular surface reconstruction. Curr Eye Res . 2014; 39(6): 541–552. [ DOI ] [ PubMed ] [ Google Scholar ] 27. Hopkinson A, Figueiredo FC. A narrative review of amniotic membrane transplantation in ocular surface repair: unveiling the immunoregulatory pathways for timely intervention. Ophthalmol Ther . 2025; 14(7): 1385–1409. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 28. Finger PT, Jain P, Mukkamala SK. Super-thick amniotic membrane graft for ocular surface reconstruction. Am J Ophthalmol . 2019; 198: 45–53. [ DOI ] [ PubMed ] [ Google Scholar ] 29. Barachetti L, Zanni M, Stefanello D, Rampazzo A. Use of four-layer porcine small intestinal submucosa alone as a scaffold for the treatment of deep corneal defects in dogs and cats: preliminary results. Vet Rec . 2020; 186(19): e28. [ DOI ] [ PubMed ] [ Google Scholar ] 30. Featherstone HJ, Sansom J, Heinrich CL. The use of porcine small intestinal submucosa in ten cases of feline corneal disease. Vet Ophthalmol . 2001; 4(2): 147–153. [ DOI ] [ PubMed ] [ Google Scholar ] 31. Matet A, Amar N, Mohand-Said S, Sahel JA, Barale PO. Argus II retinal prosthesis implantation with scleral flap and autogenous temporalis fascia as alternative patch graft material: a 4-year follow-up. Clin Ophthalmol . 2016; 10: 1565–1571. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 32. Baum JL, Mishima S, Boruchoff SA. On the nature of dellen. Arch Ophthalmol . 1968; 79(6): 657–662. [ DOI ] [ PubMed ] [ Google Scholar ] 33. Jampel HD, Musch DC, Gillespie BW, Lichter PR, Wright MM, Guire KE. Perioperative complications of trabeculectomy in the collaborative initial glaucoma treatment study (CIGTS). Am J Ophthalmol . 2005; 140(1): 16–22. [ DOI ] [ PubMed ] [ Google Scholar ] 34. Mahgoub MM, Roshdy MM, Wahba SS. Dellen formation as a complication of subconjunctival silicone oil following microincision vitrectomy. Clin Ophthalmol . 2017; 11: 2215–2219. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 35. González Gomez A, González de Gor Crooke JL, García-Ben A, García-Campos JM. Dellen and corneal perforation after bilateral pterygium excision in a patient with no risk factors. BMJ Case Rep . 2015; 2015: bcr2015213319. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 36. Kymionis GD, Plaka A, Kontadakis GA, Astyrakakis N. Treatment of corneal dellen with a large diameter soft contact lens. Cont Lens Anterior Eye . 2011; 34(6): 290–292. [ DOI ] [ PubMed ] [ Google Scholar ] 37. Barton K, Nava A, Monroy DC, Pflugfelder SC. Cytokines and tear function in ocular surface disease. Adv Exp Med Biol . 1998; 438: 461–469. [ DOI ] [ PubMed ] [ Google Scholar ] 38. Gurumurthy S, Bhambhani V, Agarwal S, Srinivasan B, Iyer G. Tear cytokines and their relevance as biomarkers in ocular surface inflammatory diseases. J Cornea Ocul Surf . 2023; 1(2): 120–129. [ Google Scholar ] 39. Hoover L, Bochicchio GV, Napolitano LM, et al.. Systemic inflammatory response syndrome and nosocomial infection in trauma. J Trauma . 2006; 61(2): 310–316; discussion 316–317. [ DOI ] [ PubMed ] [ Google Scholar ] 40. Hsing CH, Wang JJ. Clinical implication of perioperative inflammatory cytokine alteration. Acta Anaesthesiol Taiwan . 2015; 53(1): 23–28. [ DOI ] [ PubMed ] [ Google Scholar ] 41. Jansen NJ, van Oeveren W, van den Broek L, et al.. Inhibition by dexamethasone of the reperfusion phenomena in cardiopulmonary bypass. J Thorac Cardiovasc Surg . 1991; 102(4): 515–525. [ PubMed ] [ Google Scholar ] 42. Menasché P, Haydar S, Peynet J, et al.. A potential mechanism of vasodilation after warm heart surgery. The temperature-dependent release of cytokines. J Thorac Cardiovasc Surg . 1994; 107(1): 293–299. [ PubMed ] [ Google Scholar ] 43. Finn A, Naik S, Klein N, Levinsky RJ, Strobel S, Elliott M. Interleukin-8 release and neutrophil degranulation after pediatric cardiopulmonary bypass. J Thorac Cardiovasc Surg . 1993; 105(2): 234–241. [ PubMed ] [ Google Scholar ] 44. Kawamura T, Wakusawa R, Okada K, Inada S. Elevation of cytokines during open heart surgery with cardiopulmonary bypass: participation of interleukin 8 and 6 in reperfusion injury. Can J Anaesth . 1993; 40(11): 1016–1021. [ DOI ] [ PubMed ] [ Google Scholar ] 45. Mahmoud NN, Hamad K, Al Shibitini A, et al.. Investigating inflammatory markers in wound healing: understanding implications and identifying artifacts. ACS Pharmacol Transl Sci . 2024; 7(1): 18–27. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 46. Macri A, Pflugfelder S. Correlation of the Schirmer 1 and fluorescein clearance tests with the severity of corneal epithelial and eyelid disease. Arch Ophthalmol . 2000; 118(12): 1632–1638. [ DOI ] [ PubMed ] [ Google Scholar ] 47. Krupin T, Cross DA, Becker B. Decreased basal tear production associated with general anesthesia. Arch Ophthalmol . 1977; 95(1): 107–108. [ DOI ] [ PubMed ] [ Google Scholar ] 48. Calonge M, Diebold Y, Sáez V, et al.. Impression cytology of the ocular surface: a review. Exp Eye Res . 2004; 78(3): 457–472. [ DOI ] [ PubMed ] [ Google Scholar ] 49. Nelson JD, Wright JC. Conjunctival goblet cell densities in ocular surface disease. Arch Ophthalmol . 1984; 102(7): 1049–1051. [ DOI ] [ PubMed ] [ Google Scholar ] 50. Tong L, Hou AH, Wong TT. Altered expression level of inflammation-related genes and long-term changes in ocular surface after trabeculectomy, a prospective cohort study. Ocul Surf . 2018; 16(4): 441–447. [ DOI ] [ PubMed ] [ Google Scholar ] 51. Yoo J, Shoaran M. Neural interface systems with on-device computing: machine learning and neuromorphic architectures. Curr Opin Biotechnol . 2021; 72: 95–101. [ DOI ] [ PubMed ] [ Google Scholar ] 52. Musk E. An integrated brain–machine interface platform with thousands of channels. J Med Internet Res . 2019; 21(10): e16194. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Supplement 1 tvst-15-5-20_s001.docx (3.8MB, docx) Articles from Translational Vision Science & Technology are provided here courtesy of Association for Research in Vision and Ophthalmology ACTIONS View on publisher site PDF (23.1 MB) Cite Collections Permalink RESOURCES Similar articles Cited by other articles Links to NCBI Databases Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• Ultra-thin porous calcium phosphate coatings show extensive remodeling and high osseointegration after 24 weeks in a rabbit femoral condyle model#1106520
• Calcium Phosphates and Bioactive Glasses for Bone Implant Applications#7743
• Optische Signal- und Energieübertragungssysteme für epiretinale Netzhautimplantate : Schlussbericht zum Teilvorhaben 1.4. ; Arbeitsgruppe 1: EPI-RET: Ophthalmologie und Technik ; BMBF-Forschungsverbund Retina-Implantat ; Berichtszeitraum: 01. April 2000 - bis 31. Oktober 2003#129819
• Clinical Evaluation of a New Viscoelastic for Cataract Surgery#903576
• Phenylephrine Versus Eyelid Taping for Muller's Muscle-Conjunctival Resection (MMCR) Evaluation#886440