THE RAP STUDY, REPORT 5: REDISCOVERING MACULAR NEOVASCULARIZATION TYPE 3: Multimodal Imaging of Fellow Eyes over 24 months.

PURPOSE: To explore the condition of fellow eyes of patients with macular neovascularization Type 3 (MNV3) and to verify whether the retinal-choroidal anastomosis (RCA) develops equally in all MNV types. METHODS: The contralateral eyes of 94 patients with MNV3, 96 patients with MNV1, and 96 patients with MNV2 were included. Multimodal imaging was performed. The MNV3 stage including the development of fibrosis and RCA over 24 months was determined. RESULTS: In the contralateral eyes of patients of the solitary (one lesion) MNV3 group, 32 eyes (42.1%) showed early/intermediate age-related macular degeneration, 25 eyes (33%) showed MNV3, and 11 eyes (14.5%) experienced fibrosis, of which 4 eyes (5.2%) had a RCA, 7 eyes (9.2%) had atrophy after resolved MNV3, and 1 eye (1.3%) developed MNV1. In the multifocal (more than one lesion) MNV3 group, 2 eyes (11.1%) showed early/intermediate age-related macular degeneration, 9 eyes (50%) showed 15 MNV3 lesions, and 4 eyes (22.2%) showed fibrosis, of which 2 eyes (11.1%) manifested with a RCA and 3 eyes (16.7%) showed atrophy after resolved MNV3. The number of eyes with a RCA accounted for 40% of all eyes with fibrosis. The count of simultaneous bilateral multifocal MNV3 was 5 (55.6%). In the MNV1 and MNV2 groups, no eye developed a RCA. The incidence of RCAs in the scarred eyes in MNV3 was significantly higher (P < 0.0001). CONCLUSION: Retinal-choroidal anastomosis is an exclusive clinical feature of MNV3. The development of the multifocal MNV3 is usually bilateral and simultaneous. The occurrence of fibrosis in MNV3 has decreased dramatically after the introduction of the antiangiogenic therapy.

THE IMPACT OF THE VITREOMACULAR INTERFACE ON FUNCTIONAL AND ANATOMICAL OUTCOMES IN DIABETIC MACULAR EDEMA TREATED WITH THREE DIFFERENT ANTI-VEGF AGENTS: Post Hoc Analysis of The Protocol T Study.

PURPOSE: To investigate the impact of baseline vitreomacular interface status on treatment outcomes in patients treated with three different anti-vascular endothelial growth factors for diabetic macular edema. METHODS: Post hoc analysis from patients enrolled in the DRCR.net Protocol T study. Optical coherence tomography images were analyzed at baseline and at the end of follow-up to identify the presence of complete vitreomacular adhesion, partial vitreomacular adhesion, vitreomacular traction syndrome, and complete posterior vitreous detachment. RESULTS: Six hundred and twenty-nine eyes were eligible for the study based on the study criteria. Complete adhesion eyes gained on average +3.7 more ETDRS letters compared with the complete posterior vitreous detachment group at the end of the 12 months follow-up ( P < 0.001). Baseline vitreomacular interface status had no significant influence on central subfield thickness at 12 months ( P = 0.144). There was no difference between the treatment arms based on effect of baseline vitreomacular interface status on best-corrected visual acuity gain. CONCLUSION: This study provides evidence that vitreomacular interface status affects functional outcomes in diabetic macular edema patients treated with anti-vascular endothelial growth factor injections. The presence of complete or partial vitreomacular adhesion at baseline may be associated with a larger treatment benefit than those with complete posterior vitreous detachment.

VALIDATION OF AN AUTOMATED FLUID ALGORITHM ON REAL-WORLD DATA OF NEOVASCULAR AGE-RELATED MACULAR DEGENERATION OVER FIVE YEARS.

BACKGROUND/PURPOSE: To apply an automated deep learning automated fluid algorithm on data from real-world management of patients with neovascular age-related macular degeneration for quantification of intraretinal/subretinal fluid volumes in optical coherence tomography images. METHODS: Data from the Vienna Imaging Biomarker Eye Study (VIBES, 2007-2018) were analyzed. Databases were filtered for treatment-naive neovascular age-related macular degeneration with a baseline optical coherence tomography and at least one follow-up and 1,127 eyes included. Visual acuity and optical coherence tomography at baseline, Months 1 to 3/Years 1 to 5, age, sex, and treatment number were included. Artificial intelligence and certified manual grading were compared in a subanalysis of 20%. Main outcome measures were fluid volumes. RESULTS: Intraretinal/subretinal fluid volumes were maximum at baseline (intraretinal fluid: 21.5/76.6/107.1 nL; subretinal fluid 13.7/86/262.5 nL in the 1/3/6-mm area). Intraretinal fluid decreased to 5 nL at M1-M3 (1-mm) and increased to 11 nL (Y1) and 16 nL (Y5). Subretinal fluid decreased to a mean of 4 nL at M1-M3 (1-mm) and remained stable below 7 nL until Y5. Intraretinal fluid was the only variable that reflected VA change over time. Comparison with human expert readings confirmed an area under the curve of >0.9. CONCLUSION: The Vienna Fluid Monitor can precisely quantify fluid volumes in optical coherence tomography images from clinical routine over 5 years. Automated tools will introduce precision medicine based on fluid guidance into real-world management of exudative disease, improving clinical outcomes while saving resources.

IMAGING OF VITELLIFORM MACULAR LESIONS USING POLARIZATION-SENSITIVE OPTICAL COHERENCE TOMOGRAPHY.

PURPOSE: To examine the involvement of the retinal pigment epithelium (RPE) in the presence of vitelliform macular lesions (VML) in Best vitelliform macular dystrophy (BVMD), autosomal recessive bestrophinopathy, and adult-onset vitelliform macular degeneration using polarization-sensitive optical coherence tomography (PS-OCT). METHODS: A total of 35 eyes of 18 patients were imaged using a PS-OCT system and blue light fundus autofluorescence imaging. Pathogenic mutations in the BEST1 gene, 3 of which were new, were detected in all patients with BVMD and autosomal recessive bestrophinopathy. RESULTS: Polarization-sensitive optical coherence tomography showed a characteristic pattern in all three diseases with nondepolarizing material in the subretinal space consistent with the yellowish VML seen on funduscopy with a visible RPE line below it. A focal RPE thickening was seen in 26 eyes under or at the edge of the VML. Retinal pigment epithelium thickness outside the VML was normal or mildly thinned in patients with BVMD and adult-onset vitelliform macular degeneration but was diffusely thinned or atrophic in patients with autosomal recessive bestrophinopathy. Patients with autosomal recessive bestrophinopathy showed sub-RPE fibrosis alongside the subretinal VML. Polarization-sensitive optical coherence tomography was more reliable in assessing the localization and the integrity of the RPE than spectral domain OCT alone. On spectral domain OCT, identification of the RPE was not possible in 19.4% of eyes. Polarization-sensitive optical coherence tomography allowed for definite identification of the location of VML in respect to the RPE in all eyes, since it provides a tissue-specific contrast. CONCLUSION: Polarization-sensitive optical coherence tomography confirms in vivo the subretinal location of VML and is useful in the assessment of RPE integrity.

Long-Term Results of Combination Treatment with Single-Dose Ranibizumab plus Photodynamic Therapy for Retinal Angiomatous Proliferation.

PURPOSE: To present functional and anatomic outcomes of combination therapy with ranibizumab and photodynamic therapy (PDT) in a series of retinal angiomatous proliferation (RAP) cases. METHODS: A total of 17 eyes of 17 patients with RAP were included. Thirteen eyes were treatment naïve and 4 were already on ranibizumab monotherapy. Combined treatment with single-dose ranibizumab and PDT was performed on all treatment-naïve cases. Follow-up was performed every month for 6-38 months, using optical coherence tomography and indocyanine green angiography when necessary. RESULTS: Overall, 6 out of 7 treatment-naïve cases with stage I or II disease manifested complete angiographic resolution of the hot spot with a single injection of ranibizumab plus PDT. In stage III disease, 7 out of 10 eyes showed persistent leakage immediately after the combined treatment. The rate of hot spot occlusion was found to correlate with the stage of disease (p = 0.05). There was a significant improvement in posttreatment BCVA for all patients regardless of disease stage (p = 0.02), which was more evident in the earlier stages. Improvement was also greater for treatment-naïve patients (p = 0.03). The posttreatment improvement in BCVA was strongly correlated with hot spot occlusion (p < 0.001). CONCLUSIONS: Single-dose modified anti-VEGF treatment in combination with PDT may provide long-term regression at the initial stages of RAP.

Managing Pervasive Sensing Campaigns via an Experimentation-as-a-Service Platform for Smart Cities.

The adoption of technologies like the IoT in urban environments, together with the intensive use of smartphones, is driving transformation towards smart cities. Under this perspective, Experimentation-as-a-Service within OrganiCity aims to create an experimental facility with technologies, services, and applications that simplify innovation within urban ecosystems. We discuss here tools that facilitate experimentation, implementing ways to organize, execute, and administer experimentation campaigns in a smart city context. We discuss the benefits of our framework, presenting some preliminary results. This is the first time such tools are paired with large-scale smart city infrastructures, enabling both city-scale experimentation and cross-site experimentation.

An IoT-Based Solution for Monitoring a Fleet of Educational Buildings Focusing on Energy Efficiency.

Raising awareness among young people and changing their behaviour and habits concerning energy usage is key to achieving sustained energy saving. Additionally, young people are very sensitive to environmental protection so raising awareness among children is much easier than with any other group of citizens. This work examines ways to create an innovative Information & Communication Technologies (ICT) ecosystem (including web-based, mobile, social and sensing elements) tailored specifically for school environments, taking into account both the users (faculty, staff, students, parents) and school buildings, thus motivating and supporting young citizens' behavioural change to achieve greater energy efficiency. A mixture of open-source IoT hardware and proprietary platforms on the infrastructure level, are currently being utilized for monitoring a fleet of 18 educational buildings across 3 countries, comprising over 700 IoT monitoring points. Hereon presented is the system's high-level architecture, as well as several aspects of its implementation, related to the application domain of educational building monitoring and energy efficiency. The system is developed based on open-source technologies and services in order to make it capable of providing open IT-infrastructure and support from different commercial hardware/sensor vendors as well as open-source solutions. The system presented can be used to develop and offer new app-based solutions that can be used either for educational purposes or for managing the energy efficiency of the building. The system is replicable and adaptable to settings that may be different than the scenarios envisioned here (e.g., targeting different climate zones), different IT infrastructures and can be easily extended to accommodate integration with other systems. The overall performance of the system is evaluated in real-world environment in terms of scalability, responsiveness and simplicity.

Co-Creating the Cities of the Future.

In recent years, the evolution of urban environments, jointly with the progress of the Information and Communication sector, have enabled the rapid adoption of new solutions that contribute to the growth in popularity of Smart Cities. Currently, the majority of the world population lives in cities encouraging different stakeholders within these innovative ecosystems to seek new solutions guaranteeing the sustainability and efficiency of such complex environments. In this work, it is discussed how the experimentation with IoT technologies and other data sources form the cities can be utilized to co-create in the OrganiCity project, where key actors like citizens, researchers and other stakeholders shape smart city services and applications in a collaborative fashion. Furthermore, a novel architecture is proposed that enables this organic growth of the future cities, facilitating the experimentation that tailors the adoption of new technologies and services for a better quality of life, as well as agile and dynamic mechanisms for managing cities. In this work, the different components and enablers of the OrganiCity platform are presented and discussed in detail and include, among others, a portal to manage the experiment life cycle, an Urban Data Observatory to explore data assets, and an annotations component to indicate quality of data, with a particular focus on the city-scale opportunistic data collection service operating as an alternative to traditional communications.

Refractive changes after pharmacologic resolution of diabetic macular edema.

PURPOSE: To determine precisely the mean change in refractive power induced by treatment in patients with diabetic macular edema (DME). DESIGN: Prospective, randomized study. PARTICIPANTS: Fifty eyes of 50 consecutive patients with clinically significant macular edema receiving all 3 types of current state-of-the-art treatment with intravitreal antiedematous substances (ranibizumab, bevacizumab, or triamcinolone). METHODS: Patients were followed up at monthly intervals and were treated following a standardized pro re nata regimen according to protocol. Best-corrected visual acuity (BCVA) was determined by certified visual acuity examiners. The refractive power of the treated eyes was determined using a push-plus technique. The change in refraction between baseline and the visit when the macula was completely dry or when the central subfield thickness (CST) measured by optical coherence tomography had reached the thinnest level was analyzed. MAIN OUTCOME MEASURES: Spherical equivalent refraction (SER) and CST. RESULTS: Fifty eyes of 50 patients received intravitreal therapy using ranibizumab (n = 11), bevacizumab (n = 20), or triamcinolone (n = 19). Mean BCVA was 0.33±0.23 logarithm of the minimum angle of resolution (logMAR) and mean CST was 492±130 µm. The mean SER was 0.41±2.06 diopters (D) at baseline. The BCVA at the time of optimal retinal morphologic features was 0.24±0.2 logMAR, mean CST was 300±78 µm, and mean change in SER was -0.01±0.46 D. Changes is BCVA and CST were statistically significant (P < 0.0001), but the SER change was not (P = 0.824). CONCLUSIONS: Appropriate spectacle correction can be prescribed to patients with DME any time during ongoing therapy using antiedematous substances because resolution of retinal thickening is not associated with an increased risk of a myopic shift.

Exploring the transformative power of AI in art through a circular economy lens. A systematic literature review.

Today, technology and sustainability are two strategic axes for the development of any industry. Art is no exception and embodies both principles. Artificial intelligence (AI) is driving the art world forwards with its applications and algorithms. Additionally, the circular economy (CE) is concerned with resources and the environment in this context. The objective of the present work is to provide an overview of the current state of research on the application of AI in the art world and an analysis of how CE principles are being incorporated, considering the interactions between AI and the CE. To this end, a systematic review of the literature is carried out in which 60 articles related to the subject are selected, analysed, and classified, highlighting the lines of research addressed. The assessment of the current state of research on the subject concludes with the four main axes of classification of works. The first line is related to AI generative content in art, addressing issues of content creation, image and painting, video, and theatre. The second line is related to AI applications for art industry production, considering the sustainability of the supply chain. The third line focuses on how the CE is being applied to art, while the fourth line focuses on other relevant aspects analysed, such as training and design. The topic is still incipient, mandating further research to study the full potential of AI and the CE in the world of art.

Surgically induced astigmatism in the scleral fixated 'Carlevale' IOL technique.

PURPOSE: To evaluate the surgically induced astigmatism over a 6-month follow-up period in patients who underwent scleral IOL fixation using an acrylic single-piece IOL with special haptics designed for sutureless scleral fixation. METHODS: We conducted a prospective longitudinal study at a single site with a single surgeon. We included patients who received transscleral IOL implantation following the Carlevale technique and were followed up post-operatively for 24 weeks. We measured the patient's refraction at baseline, week 12 and week 24 using the best corrected visual acuity at 4 m (EDTRS chart). We performed corneal tomography at every visit using an anterior segment optical coherence tomography (AS-OCT). We evaluated surgically induced astigmatism (SIA) and refraction during each follow-up visit and compared them to baseline. We then assessed changes in SIA over time. RESULTS: In total, 27 eyes of 27 patients consisting of 16 female and 11 male individuals were evaluated. The mean patient age was 71 ± 11.7 years, mean axial length was 24.30 ± 1.47 mm (range: 21.4-27.23) and mean white-to-white distance was 12.07 ± 0.40 mm (range: 11.4-12.7). The mean SIA decreased from 1.78 ± 0.96D at week 1 significantly to 0.80 ± 0.55D at week 12 (p < 0.001) and then stayed unchanged around 0.82 ± 0.72D at week 24 (p = 1.0). CONCLUSIONS: The scleral fixated Carlevale IOL and its implantation procedure were found to result in a predictable SIA of <1D after 24 weeks. However, the axis orientation of the SIA appeared to be random, making it unsuitable for implementation in toric IOL calculations.

Structure-Function Correlation of Retinal Fibrosis in Eyes with Neovascular Age-Related Macular Degeneration.

Purpose: To assess retinal function in areas of presumed fibrosis due to neovascular age-related macular degeneration (nAMD), using multimodal imaging and structure-function correlation. Design: Cross-sectional observational study. Methods: 30 eyes of 30 consecutive patients with nAMD with a minimum history of one year of anti-vascular endothelial growth factor therapy were included. Each patient underwent microperimetry (MP), color fundus photography (CFP), standard spectral-domain-based OCT (SD-OCT), and polarization sensitive-OCT (PS-OCT) imaging. PS-OCT technology can depict retinal fibrosis based on its birefringence. CFP, SD-OCT, and PS-OCT were evaluated independently for the presence of fibrosis at the corresponding MP stimuli locations. MP results and morphologic findings in CFP, SD-OCT, and PS-OCT were co-registered and analyzed using mixed linear models. Results: In total, 1350 MP locations were evaluated to assess the functional impact of fibrosis according to a standardized protocol. The estimated means of retinal areas with signs of fibrosis were 12.60 db (95% confidence interval: 10.44-14.76) in CFP, 11.60 db (95% COI: 8.84-14.36) in OCT, and 11.02 db (95% COI 8.10-13.94) in PS-OCT. Areas evaluated as subretinal fibrosis in three (7.2 db) or two (10.1 db) modalities were significantly correlated with a lower retinal sensitivity than a subretinal fibrosis observed in only one (15.3 db) or none (23.3 db) modality (p < 0.001). Conclusions: CFP, SD-OCT and PS-OCT are all suited to detect areas of reduced retinal sensitivity related to fibrosis, however, a multimodal imaging approach provides higher accuracy in the identification of areas with low sensitivity in MP (i.e., impaired retinal function), and thereby improves the detection rate of subretinal fibrosis in nAMD.

Retinal architecture recovery after grid photocoagulation in diabetic macular edema observed in vivo by spectral domain optical coherence tomography.

PURPOSE: To identify the morphologic changes secondary to macular grid photocoagulation in diabetic macular edema in vivo using spectral domain optical coherence tomography. METHODS: In this prospective cohort study, 13 consecutive patients with vision loss because of clinically significant macular edema associated with diabetes mellitus Type 2 underwent grid laser treatment (PASCAL). Best-corrected visual acuity, Spectralis optical coherence tomography, infrared fundus imaging, and biomicroscopy were performed at baseline, Day 1, Week 1, and Months 1, 2, and 3 after treatment. Fluorescein angiography was performed at baseline and at 3 months. RESULTS: Mean central 1-mm thickness decreased significantly from 438 ± 123 µm (mean ± SD) at baseline to 391 ± 111 µm (P < 0.05) at 3 months with a nonsignificant trend of best-corrected visual acuity improvement. A wipeout of the photoreceptor layer and the inner segment/outer segment line together with an alteration of the overlaying outer nuclear layer and external limiting membrane was seen at Day 1. The lesion was reduced to a focal hyperreflective deposit on the retinal pigment epithelium boundary. In 55% of lesions, the external limiting membrane as well as the previously interrupted inner segment/outer segment line revealed intact continuity at Month 3. In some areas, repair was incomplete indicated by a focal condensation interrupting the inner segment/outer segment line in the lesion center. CONCLUSION: In vivo imaging of morphologic lesion repair in human eyes after PASCAL grid laser of diabetic macular edema demonstrates progressive restoration of the external limiting membrane and inner segment/outer segment integrity as previously described in animal models.

Response of retinal sensitivity to ranibizumab treatment of macular edema after acute branch retinal vein occlusion.

PURPOSE: To evaluate microperimetry changes in patients with acute macular edema secondary to branch retinal vein occlusion during a follow-up period of 12 months with intravitreal ranibizumab treatment (Lucentis; Novartis). METHODS: Patients with macular edema secondary to branch retinal vein occlusion received an intravitreous injection of 0.5 mg of ranibizumab (0.05 mL). Best-corrected visual acuity, Spectralis OCT (Heidelberg Engineering), and color fundus photography were performed at monthly intervals over a follow-up period of 1 year. Macular function was documented by microperimetry (Nidek, MP-1) at baseline, 3, and 12 months. RESULTS: Data of 20 patients without lack of microperimetry results were included to the statistical analyses. The size of the area of absolute scotoma was reduced from 16% at baseline to 11.7% at Month 3 and remained stable in the entire study duration (P > 0.05). Mean differential light threshold improved significantly under therapy from 9.47 dB at baseline to 12.53 dB at 12 months (P < 0.001). Best-corrected visual acuity correlated significantly with central millimeter thickness and mean retinal sensitivity at baseline and at 12-month follow-up visits. CONCLUSION: In addition to anatomical restoration and increased visual acuity, intravitreal ranibizumab also improved the central macular function in patients with acute macular edema after branch retinal vein occlusion.

Segmentation of macular neovascularization and leakage in fluorescein angiography images in neovascular age-related macular degeneration using deep learning.

BACKGROUND/OBJECTIVES: We aim to develop an objective fully automated Artificial intelligence (AI) algorithm for MNV lesion size and leakage area segmentation on fluorescein angiography (FA) in patients with neovascular age-related macular degeneration (nAMD). SUBJECTS/METHODS: Two FA image datasets collected form large prospective multicentre trials consisting of 4710 images from 513 patients and 4558 images from 514 patients were used to develop and evaluate a deep learning-based algorithm to detect CNV lesion size and leakage area automatically. Manual segmentation of was performed by certified FA graders of the Vienna Reading Center. Precision, Recall and F1 score between AI predictions and manual annotations were computed. In addition, two masked retina experts conducted a clinical-applicability evaluation, comparing the quality of AI based and manual segmentations. RESULTS: For CNV lesion size and leakage area segmentation, we obtained F1 scores of 0.73 and 0.65, respectively. Expert review resulted in a slight preference for the automated segmentations in both datasets. The quality of automated segmentations was slightly more often judged as good compared to manual annotations. CONCLUSIONS: CNV lesion size and leakage area can be segmented by our automated model at human-level performance, its output being well-accepted during clinical applicability testing. The results provide proof-of-concept that an automated deep learning approach can improve efficacy of objective biomarker analysis in FA images and will be well-suited for clinical application.

Association of microvascular biomarkers in fluorescein angiography with macrovascular-related mortality in clinical routine data.

PURPOSE: Early detection of microvascular changes in the retina may be important for the risk assessment of cardiovascular health. Therefore, the purpose of this study was to investigate imaging biomarkers in fluorescein angiography (FA) as potential predictors for cardiovascular mortality. METHODS: In this retrospective, matched case-control study, we included FA images from clinical routine data between 2007 and 2018 of 100 patients who died of macrovascular events (Group 1) and 100 age- and sex-matched controls (Group 2). All patients were under treatment for different, mostly retinal, ocular diseases. FA images were used for the measurement of the foveal avascular zone (FAZ) and the arteriolar and venular caliber. RESULTS: Patients mean age on examination day was 69.5 ± 8.3 years with a 1:1 female:male subject ratio. Mean FAZ area of our sample was 0.340 ± 0.135 mm2 for Group 1 and 0.264 ± 0.137 mm2 for Group 2 (P < 0.001), showing a larger FAZ area in patients who subsequently died of macrovascular-related systemic diseases. CONCLUSIONS: Individuals effected by a macrovascular-related disease show a larger FAZ on FA examinations before the event compared to patients which are unaffected. Our results highlight a possible role of the FAZ as additional biomarker for the cardiovascular condition.

Quality assessment of colour fundus and fluorescein angiography images using deep learning.

BACKGROUND/AIMS: Image quality assessment (IQA) is crucial for both reading centres in clinical studies and routine practice, as only adequate quality allows clinicians to correctly identify diseases and treat patients accordingly. Here we aim to develop a neural network for automated real-time IQA in colour fundus (CF) and fluorescein angiography (FA) images. METHODS: Training and evaluation of two neural networks were conducted using 2272 CF and 2492 FA images, with binary labels in four (contrast, focus, illumination, shadow and reflection) and three (contrast, focus, noise) modality specific categories plus an overall quality ranking. Performance was compared with a second human grader, evaluated on an external public dataset and in a clinical trial use-case. RESULTS: The networks achieved a F1-score/area under the receiving operator characteristic/precision recall curve of 0.907/0.963/0.966 for CF and 0.822/0.918/0.889 for FA in overall quality prediction with similar results in most categories. A clear relation between model uncertainty and prediction error was observed. In the clinical trial use-case evaluation, the networks achieved an accuracy of 0.930 for CF and 0.895 for FA. CONCLUSION: The presented method allows automated IQA in real time, demonstrating human-level performance for CF as well as FA. Such models can help to overcome the problem of human intergrader and intragrader variability by providing objective and reproducible IQA results. It has particular relevance for real-time feedback in multicentre clinical studies, when images are uploaded to central reading centre portals. Moreover, automated IQA as preprocessing step can support integrating automated approaches into clinical practice.

Linking Function and Structure with ReSensNet: Predicting Retinal Sensitivity from OCT using Deep Learning.

PURPOSE: The currently used measures of retinal function are limited by being subjective, nonlocalized, or taxing for patients. To address these limitations, we sought to develop and evaluate a deep learning (DL) method to automatically predict the functional end point (retinal sensitivity) based on structural OCT images. DESIGN: Retrospective, cross-sectional study. SUBJECTS: In total, 714 volumes of 289 patients were used in this study. METHODS: A DL algorithm was developed to automatically predict a comprehensive retinal sensitivity map from an OCT volume. Four hundred sixty-three spectral-domain OCT volumes from 174 patients and their corresponding microperimetry examinations (Nidek MP-1) were used for development and internal validation, with a total of 15 563 retinal sensitivity measurements. The patients presented with a healthy macula, early or intermediate age-related macular degeneration, choroidal neovascularization, or geographic atrophy. In addition, an external validation was performed using 251 volumes of 115 patients, comprising 3 different patient populations: those with diabetic macular edema, retinal vein occlusion, or epiretinal membrane. MAIN OUTCOME MEASURES: We evaluated the performance of the algorithm using the mean absolute error (MAE), limits of agreement (LoA), and correlation coefficients of point-wise sensitivity (PWS) and mean sensitivity (MS). RESULTS: The algorithm achieved an MAE of 2.34 dB and 1.30 dB, an LoA of 5.70 and 3.07, a Pearson correlation coefficient of 0.66 and 0.84, and a Spearman correlation coefficient of 0.68 and 0.83 for PWS and MS, respectively. In the external test set, the method achieved an MAE of 2.73 dB and 1.66 dB for PWS and MS, respectively. CONCLUSIONS: The proposed approach allows the prediction of retinal function at each measured location directly based on an OCT scan, demonstrating how structural imaging can serve as a surrogate of visual function. Prospectively, the approach may help to complement retinal function measures, explore the association between image-based information and retinal functionality, improve disease progression monitoring, and provide objective surrogate measures for future clinical trials.

Using gamification and IoT-based educational tools towards energy savings - some experiences from two schools in Italy and Greece.

Climate change and the need for sustainable development have become part of our daily lives. In this context, it is crucial to involve the educational community to the discussion, both students and teachers; by increasing awareness about these issues and the ways school communities can contribute to energy savings, we can kick-start a change towards more sustainable practices in our societies. The Green Awareness in Action (GAIA) H2020 research project implemented an IoT-based approach in several European schools for sustainability awareness and energy efficiency, while at the same time aiming for increasing students' digital skills. By using gamification, competitions and IoT-based educational activities, GAIA engaged directly with teachers and students in order to realize energy-saving activities in their environment. We report here on the use of gamification and competition among schools in this context, and how they helped together with IoT-based lab activities to engage students and educators to participate in the project more actively. We provide details on the implementation of GAIA's intervention in specific school settings to showcase our approach. Our findings, backed up by evaluation data and answers to a survey by 30 educators in Greece and Italy, confirm that the inclusion of competition and gamification aspects can significantly increase students' engagement, especially when having groups/schools competing with each other. Moreover, IoT-based educational activities can supplement existing educational activities in interesting ways, with students evaluating positively the experience and educators reporting increased overall student engagement in their class during the intervention period, and, on average, better class performance compared to previous periods.