BACKGROUND: To compare the repeatability and reproducibility of corneal and corneal epithelial thickness mapping using anterior segment optical coherence tomography (AS-OCT) according to tear film break-up time (TBUT).
METHODS: The included eyes were divided into three subgroups according to TBUT (group 1: TBUT ≤ 5 s, group 2: 5 s < TBUT ≤ 10 s, and group 3: TBUT > 10 s). All eyes were imaged separately thrice by two operators to obtain the thickness maps (TMs) of the cornea and corneal epithelium based on spatial zones encompassing a 9-mm-diameter area. Each TM consisted of 25 areas. Intraoperator (repeatability) and interoperator (reproducibility) standard deviations (Sws), coefficients of variation (CoVs), and intraclass correlation coefficients (ICCs) among the tests were calculated and compared in all the areas.
RESULTS: Altogether, 132 eyes of 67 subjects were included (50, 47, and 35 eyes in groups 1, 2, and 3; respectively). The ICCs of corneal epithelial thickness and corneal thickness were > 0.75 in most of the areas. Pairwise comparisons showed that AS-OCT exhibited lower repeatability in group 1 than in groups 2 and 3 (P < 0.05). However groups 2 and 3 showed similar results. Sws and CoVs of corneal epithelial thickness exhibited no significant interoperator differences. While no significant differences were observed in corneal thickness in most of the areas.
CONCLUSIONS: TBUT significantly influences the repeatability of corneal and corneal epithelial thickness measurements. Poor tear film stability requires careful evaluation of corneal epithelial thickness.

UNASSIGNED: Optical coherence tomography has great utility for capturing dynamic processes, but such applications are particularly data-intensive. Samples such as biological tissues exhibit temporal features at varying time scales, which makes data reduction challenging.
UNASSIGNED: We propose a method for capturing short- and long-term correlations of a sample in a compressed way using non-uniform temporal sampling to reduce scan time and memory overhead.
UNASSIGNED: The proposed method separates the relative contributions of white noise, fluctuating features, and stationary features. The method is demonstrated on mammary epithelial cell spheroids in three-dimensional culture for capturing intracellular motility without loss of signal integrity.
UNASSIGNED: Results show that the spatial patterns of motility are preserved and that hypothesis tests of spheroids treated with blebbistatin, a motor protein inhibitor, are unchanged with up to eightfold compression.
UNASSIGNED: The ability to measure short- and long-term correlations compressively will enable new applications in (3+1)D imaging and high-throughput screening.

BACKGROUND: acute radiation dermatitis (ARD) is the most widely reported radiotherapy-induced adverse event. Currently, there is no objective or reliable method to measure ARD.
OBJECTIVE: our main objective was to identify and quantify the effects of radiotherapy with a computational model using optical coherence tomography (OCT) skin scanning. Secondary objectives included determining the ARD impact of different radiotherapeutic schemes and adjuvant topical therapies.
METHODS: we conducted a prospective, single-center case series study in a tertiary referral center of patients with breast cancer who were eligible for whole breast radiotherapy (WBRT).
RESULTS: a total of 39 women were included and distributed according to the radiotherapeutic schemes (15, 20, and 25 fractions). A computational model was designed to quantitatively analyze OCT findings. After radiotherapy, OCT scanning was more sensitive revealing vascularization changes in 84.6% of the patients (vs 69.2% of the patients with ARD by clinical examination). OCT quantified an increased vascularization at the end of WBRT (P < .05) and a decrease after 3 months (P = .032). Erythematous skin changes by OCT were more pronounced in the 25-fraction regime.
CONCLUSIONS: an OCT computational model allowed for the identification and quantification of vascularization changes on irradiated skin, even in the absence of clinical ARD. This may allow the design of standardized protocols for ARD beyond the skin color of the patients involved.

Age-related neuronal adaptations are known to help maintain function. This study aims to examine gross age-related in vivo retinal functional adaptations (using electroretinography) in young and middle aged C57BL/6J and Thy1-YFPh mice and to relate this to in vivo retinal structure (using optical coherence tomography). Electroretinography responses were generally larger in Thy1-YFPh mice than in C57BL/6J mice, with similar in vivo retinal layer thicknesses except for longer inner/outer photoreceptor segment in Thy1-YFPh mice. Relative to 3-month-old mice, 12-month-old mice showed reduced photoreceptor (C57BL/6J 84.0±2.5 %; Thy1-YFPh 80.2±5.2 %) and bipolar cell (C57BL/6J 75.6±2.3 %; Thy1-YFPh 68.1±5.5 %) function. There was relative preservation of ganglion cell function (C57BL/6J 79.7±3.7 %; Thy1-YFPh 91.7±5.0 %) with age, which was associated with increased b-wave (bipolar cell) sensitivities to light. Ganglion cell function was correlated with both b-wave amplitude and sensitivity. This study shows that there are normal age-related adaptations to preserve functional output. Different mouse strains may have varied age-related adaptation capacity and should be taken into consideration when examining age-related susceptibility to injury.

This article reviews literature on the use of artificial intelligence (AI) for screening, diagnosis, monitoring and treatment of glaucoma. The first part of the review provides information how AI methods improve the effectiveness of glaucoma screening, presents the technologies using deep learning, including neural networks, for the analysis of big data obtained by methods of ocular imaging (fundus imaging, optical coherence tomography of the anterior and posterior eye segments, digital gonioscopy, ultrasound biomicroscopy, etc.), including a multimodal approach. The results found in the reviewed literature are contradictory, indicating that improvement of the AI models requires further research and a standardized approach. The use of neural networks for timely detection of glaucoma based on multimodal imaging will reduce the risk of blindness associated with glaucoma.
В первой части обзора данных литературы, посвященной применению методов искусственного интеллекта (ИИ) для скрининга, диагностики, мониторинга и лечения глаукомы, приводятся сведения о том, как методы ИИ повышают эффективность скрининга глаукомы. Представлены технологии применения глубокого обучения, включая нейронные сети, в анализе больших данных, полученных с помощью методов визуализации глаза (фундус-изображений глазного дна, оптической когерентной томографии переднего и заднего отрезков глаза, цифровой гониоскопии, ультразвуковой биомикроскопии и т.д.), включая мультимодальный подход. Результаты проанализированной литературы носят противоречивый характер, что указывает на необходимость дальнейшего поиска с учетом стандартизированного подхода для повышения эффективности моделей ИИ. Применение нейросетей для своевременного выявления глаукомы, основанное на мультимодальной визуализации, позволит снизить риск слепоты, ассоциированной с глаукомой.

Multiple sclerosis (MS) is a chronic autoimmune-inflammatory and neurodegenerative disease.
OBJECTIVE: This study explores the main structural changes in patients with MS and their relationships with the activity and type of disease course.
METHODS: This prospective study included 159 patients (318 eyes) with an established diagnosis of MS: group (44 eyes; 13.84%) - relapsing-remitting type MS (RRMS) lasting up to 1 year without a history of optic neuritis (ON); group 2 (30 eyes; 9.43%) - RRMS up to 1 year with ON; group 3 (56 eyes; 17.61%) - RRMS lasting from 1 to 10 years without ON; group 4 (38 eyes; 11.95%) - RRMS from 1 to 10 years with ON; group 5 (49 eyes; 15.41%) - RRMS >10 years without ON; group 6 (37 eyes; 11.63%) - RRMS >10 years with ON; group 7 (34 eyes; 10.69%) - secondary progressive multiple sclerosis (SPMS) without ON; group 8 (30 eyes; 9.43%) - SPMS with ON. Patients underwent standard ophthalmological examinations, including optical coherence tomography.
RESULTS: A decrease in structural parameters was diagnosed, progressing with the duration of the disease and the presence of ON: the minimum values of mGCL+IPL (65.83±9.14 μm) and mSNFL (76.37±14.77 μm) were detected in the group with SPMS with ON. High inverse correlations of EDSS with mGCL+IPL and mRNFL were demonstrated, with maximum in the group with the longest duration of MS without ON (-0.48 and -0.52 (p=0.01), respectively).
CONCLUSIONS: Changes in the thickness of the structural parameters of the retina, measured by OCT, can be considered as a predictor of the course of MS.
Рассеянный склероз (РС) — хроническое аутоиммунно-воспалительное и нейродегенеративное заболевание.
UNASSIGNED: Изучение основных структурных изменений сетчатки у пациентов с РС и их взаимосвязи с активностью и типом течения заболевания.
UNASSIGNED: В проспективное исследование включено 159 пациентов (318 глаз) с установленным диагнозом РС: 1-я группа (44 глаза; 13,84%) — ремиттирующий РС (РРС) длительностью до 1 года без оптического неврита (ОН) в анамнезе; 2-я группа (30 глаз; 9,43%) — РРС длительностью до 1 года с ОН; 3-я группа (56 глаз; 17,61%) — РРС длительностью от 1 года до 10 лет без ОН; 4-я группа (38 глаз; 11,95%) — РРС длительностью от 1 года до 10 лет с ОН; 5-я группа (49 глаз; 15,41%) — РРС длительностью > 10 лет без ОН; 6-я группа (37 глаз; 11,63%) — РРС длительностью >10 лет с ОН; 7-я группа (34 глаза; 10,69%) — РС с вторично-прогрессирующим течением (ВПРС) без ОН; 8-я группа (30 глаз; 9,43%) — ВПРС с ОН. Пациентам выполняли стандартное офтальмологическое обследование, включающее проведение оптической когерентной томографии (ОКТ).
UNASSIGNED: Диагностировано снижение структурных показателей, прогрессирующее в соответствии с длительностью заболевания и наличием ОН: минимальные значения срGCL+IPL (65,83±9,14 мкм) и срRNFL (76,37±14,77 мкм) выявлены в группе с ВПРС с ОН. Продемонстрированы высокие обратные корреляционные связи EDSS с срGCL+IPL и срRNFL, максимальные в группе с наибольшей длительностью РС без ОН (–0,48 и –0,52 соответственно; p=0,01).
UNASSIGNED: Изменения толщины структурных параметров сетчатки, оцениваемые при ОКТ, могут быть рассмотрены как предиктор течения РС.

UNASSIGNED: Intraocular metastases of systemic cancer are most frequently located in the choroid, followed by the iris and ciliary body, while retinal metastases are extremely rare. Here we present a case of retinal metastasis and analyze multimodal imaging.
UNASSIGNED: A 66-year-old woman with a medical history of breast cancer 5 years earlier was referred to our Department struggling with blurry vision in her right eye. At initial examination, her best-corrected visual acuity (BCVA) was 1.0 oculus dexter (OD). Fundus examination revealed a yellowish elevated lesion with irregular surface, measuring 2 papillary diameters, along with serous retinal detachment (SRD) on the temporal side of the optic disc. Optical coherence tomography showed SRD with an isointense nodule extending across all retinal layers. Fluorescein angiography showed hyperfluorescence and vigorous fluorescence leakage inside the tumor in the early and late phases, respectively. Indocyanine green angiography depicted feeder and drainage vessels within the mass. Laser speckle flowgraphy (LSFG) showed a cold signal inside the tumor. Based on these clinical findings, the mass was diagnosed as a retinal metastasis. Eight days after the initial visit, the patient underwent external beam radiation to the right eye. One month after the initial diagnosis, her BCVA was 0.7 OD, the tumor was localized, and SRD had decreased. LSFG indicated vascular remodeling with marginally warmer signals in the tumor.
UNASSIGNED: LSFG of the retinal metastasis showed a cold signal, suggesting low tumor blood flow velocity and that the tumor may have grown slowly. LSFG findings are likely to play a supportive role in clinical diagnosis and contribute to better understanding of pathogenesis in juxtapapillary tumors.

Patients with acute coronary syndrome (ACS), frequently caused by plaque rupture (PR), often have vulnerable plaques in residual lesions as well as in culprit lesions. However, whether this occurs in patients with plaque erosion (PE) as well is unknown. We retrospectively analyzed the data of 88 patients with ACS who underwent both optimal coherence tomography (OCT) and intravascular ultrasound (IVUS). Based on plaque morphology of the culprit lesions identified using OCT, patients were classified into PE (n=23) and PR (n=35) groups. The tissue characteristics of residual lesions evaluated using integrated backscatter IVUS were compared between both groups after percutaneous coronary intervention. The PE group had a significantly lower percent lipid volume and a higher percent fibrous volume than the PR group (35.0±17.8% vs 49.2±13.4%, p<0.001; 63.2±17.1% vs 50.3±13.1%, p=0.002, respectively). Receiver operating characteristic curve analysis revealed that percent lipid volume in the residual lesions was a significant discriminant factor in estimating the plaque morphology of the culprit lesion (optimal cut-off value, <43.5%; sensitivity and specificity values were 73.9% and 68.6%, respectively). In conclusion, patients with PE had a significantly lower percent lipid volume and a significantly higher percent fibrous volume in the residual lesions than those with PR, suggesting that the nature of coronary plaques in patients with PE is different from that of those with PR.

BACKGROUND: Inflammatory skin diseases, such as psoriasis, atopic eczema, and contact dermatitis pose diagnostic challenges due to their diverse clinical presentations and the need for rapid and precise diagnostic assessment.
OBJECTIVE: While recent studies described non-invasive imaging devices such as Optical coherence tomography and Line-field confocal OCT (LC-OCT) as possible techniques to enable real-time visualization of pathological features, a standardized analysis and validation has not yet been performed.
METHODS: One hundred forty lesions from patients diagnosed with atopic eczema (57), psoriasis (50), and contact dermatitis (33) were imaged using OCT and LC-OCT. Statistical analysis was employed to assess the significance of their characteristic morphologic features. Additionally, a decision tree algorithm based on Gini\'s coefficient calculations was developed to identify key attributes and criteria for accurately classifying the disease groups.
RESULTS: Descriptive statistics revealed distinct morphologic features in eczema, psoriasis, and contact dermatitis lesions. Multivariate logistic regression demonstrated the significance of these features, providing a robust differentiation between the three inflammatory conditions. The decision tree algorithm further enhanced classification accuracy by identifying optimal attributes for disease discrimination, highlighting specific morphologic criteria as crucial for rapid diagnosis in the clinical setting.
CONCLUSIONS: The combined approach of descriptive statistics, multivariate logistic regression, and a decision tree algorithm provides a thorough understanding of the unique aspects associated with each inflammatory skin disease. This research offers a practical framework for lesion classification, enhancing the interpretability of imaging results for clinicians.

Deep learning classification models for medical image analysis often perform well on data from scanners that were used to acquire the training data. However, when these models are applied to data from different vendors, their performance tends to drop substantially. Artifacts that only occur within scans from specific scanners are major causes of this poor generalizability. We aimed to enhance the reliability of deep learning classification models using a novel method called Uncertainty-Based Instance eXclusion (UBIX). UBIX is an inference-time module that can be employed in multiple-instance learning (MIL) settings. MIL is a paradigm in which instances (generally crops or slices) of a bag (generally an image) contribute towards a bag-level output. Instead of assuming equal contribution of all instances to the bag-level output, UBIX detects instances corrupted due to local artifacts on-the-fly using uncertainty estimation, reducing or fully ignoring their contributions before MIL pooling. In our experiments, instances are 2D slices and bags are volumetric images, but alternative definitions are also possible. Although UBIX is generally applicable to diverse classification tasks, we focused on the staging of age-related macular degeneration in optical coherence tomography. Our models were trained on data from a single scanner and tested on external datasets from different vendors, which included vendor-specific artifacts. UBIX showed reliable behavior, with a slight decrease in performance (a decrease of the quadratic weighted kappa (κw) from 0.861 to 0.708), when applied to images from different vendors containing artifacts; while a state-of-the-art 3D neural network without UBIX suffered from a significant detriment of performance (κw from 0.852 to 0.084) on the same test set. We showed that instances with unseen artifacts can be identified with OOD detection. UBIX can reduce their contribution to the bag-level predictions, improving reliability without retraining on new data. This potentially increases the applicability of artificial intelligence models to data from other scanners than the ones for which they were developed. The source code for UBIX, including trained model weights, is publicly available through https://github.com/qurAI-amsterdam/ubix-for-reliable-classification.