International Journal of Retina and Vitreous最新文献

Background: Chatbots powered by large language models have shown promising results when addressing medical queries, thus transforming access to medical information. Although these chatbots produce detailed and accurate responses, it is unclear how they perform when handling real, unedited patient questions, particularly in non-English languages. This study aimed to assess the readability, accuracy, and comprehensiveness of responses to retinal disease queries provided by four chatbots (ChatGPT 4.0, ConsensusGPT, Gemini, and Claude 3) compared to responses from retina specialists.

Methods: In this cross-sectional, comparative, and blinded study, twenty unedited questions about retinal diseases were randomly selected from a popular online video channel in Portuguese. The questions were submitted to the four selected chatbots and retina specialists with fellowship training. Two independent retinal experts evaluated the responses using standardized Likert scales for accuracy and completeness. Readability was assessed using the Flesch Reading Ease Score and the Flesch-Kincaid Grade Level tests. Additional metrics, including word count and response generation time, were analyzed. Data were compared among groups using non-parametric statistical tests, including the Kruskal-Wallis test with Dunn's pairwise comparisons and chi-squared tests, with a two-sided p-value threshold of 0.05 for statistical significance.

Results: Retinal specialists and the Gemini chatbot produced responses with higher readability, indicating lower educational levels were needed for comprehension. In contrast, ChatGPT 4.0, ConsensusGPT, and Claude 3 delivered more detailed and accurate answers but required a higher reading level. ChatGPT 4.0 and ConsensusGPT achieved superior quality and comprehensiveness ratings compared to human experts and the other chatbots. Additionally, all AI systems generated responses significantly faster than the human specialists. Evaluators could correctly distinguish between human-generated and AI-generated responses in most cases.

Conclusions: Artificial intelligence chatbots demonstrate considerable promise for rapidly disseminating accurate medical information directly to the end-user, not only in English. However, optimizing the simplicity of their language is essential to ensure that detailed responses remain accessible to a broad audience. Future research should aim to replicate our findings with larger datasets of questions with the goal of refining these systems to balance comprehensive content with user-friendly language.

Background: To report the clinical course and outcomes of a surgical approach for progressive severe stellate non-hereditary idiopathic foveomacular retinoschisis (SNIFR) using pars plana vitrectomy (PPV).

Methods: Multi-center, consecutive, interventional case series. Patients with a diagnosis of SNIFR presenting with progressive loss of vision between January 1, 2017 and January 1, 2023. Evaluation of ophthalmologic findings and multimodal ocular imaging at the time of diagnosis, surgical procedure, and of visual and anatomic outcomes postoperatively. The main outcome measures evaluated include best corrected visual acuity (BCVA), central macular thickness (CMT), and findings on optical coherence tomography (OCT).

Results: Seven patients diagnosed with SNIFR were included. Median age in years at the time of diagnosis was 64 (range, 46-77). Four patients were female, and three were male. Genetic testing for mutations in retinoschisin 1 (RS1) and for other inherited conditions associated with foveomacular retinoschisis was negative. All patients demonstrated progressive and severe retinoschisis, as well as worsening vision loss and metamorphopsia when managed conservatively. PPV was performed and revealed anomalously broad and dense adherence of the posterior hyaloid in all eyes. The internal limiting membrane (ILM) was peeled in all but one case. Median BCVA at baseline measured 20/50, and declined to 20/70 at the time of surgery. Median preoperative CMT measured 561 μm, with OCT demonstrating prominent retinoschisis of the outer plexiform and outer nuclear layers. All eyes demonstrated postoperative resolution of retinoschisis and subretinal fluid, with corresponding improvements in both BCVA and subjective central visual distortion up to six months after surgery. BCVA for the entire cohort improved to a median of 20/30, and with a corresponding decrease in CMT to a median of 240 μm.

Conclusion: PPV is an effective surgical intervention resulting in anatomic resolution of retinoschisis and improved functional vision in eyes with progressive and severe SNIFR.

Purpose: To develop a deep learning (DL) model for segmenting retinal hard exudates (HE) from optical coherence tomography (OCT) scans.

Methods: A modified U-Net architecture was trained on manually segmented OCT B-scans of retinal HE. The training set included 1,811 OCT scans from 15 patients with diabetic retinopathy or branch retinal vein occlusion. The model was evaluated using Dice coefficient and accuracy in idependant test set, and its HE area and volume predictions were compared to manually measured HE areas from a previous clinical study. Additionally, a 2D projected image was generated from the 3D structure of the predicted HE.

Results: The DL model achieved a Dice coefficient of 0.721 and an accuracy of 99.9% on the test set. There was a moderate correlation between model-predicted HE volume and area and manually measured HE area from fundus photographs (R = 0.589 and 0.618, respectively; both P < 0.001). The projected 2D image generated from the model accurately visualized HE details, demonstrating better structural information compared to fundus photographs.

Conclusion: The proposed DL model enables accurate segmentation of retinal HE, offering volumetric data with both horizontal and vertical structural information, which enhances visualization and quantification compared to traditional 2D imaging.

Background: The treatment of retinoblastoma (Rb) has undergone significant improvements over the last century. This study aims to assess the trend of enucleation and mortality in retinoblastoma patients during the last 3 decades.

Methods: The study utilized data from the referral center for ocular oncology in Rasool Akram Hospital, Tehran, Iran. It included all patients diagnosed with Rb from August 1991 to December 2018. The study investigated the trend of enucleation and mortality during three-time intervals: before 2001 (T1), during 2001-2007 (T2), and 2008-2018 (T3). Additionally, it assessed the trend of enucleation and mortality based on laterality, age and presentation of Rb (strabismus and leukocoria).

Results: The incidence of enucleation decreased significantly from T1 to T3 (74-41%) during the study period (p-value < 0.001). Pairwise comparisons between T1 and T3 revealed a significant decrease in the incidence of enucleation (74% vs. 41%, p-value < 0.001). The study also demonstrated a significant reduction in the incidence of enucleation when comparing T2 to T3 (60% vs. 41%, p-value < 0.001). Comparing time intervals, there was no significant difference between T2 and T3 regarding the incidence of death (4% vs. 1%), but both intervals had statistically significant lower death rates compared with T1 (26%, both p-values < 0.001).

Conclusion: This study revealed that Introduction of systemic chemotherapy as mainstay of Rb treatment, has led to a significant reduction in mortality and morbidity rates. The incorporation of targeted chemotherapy has further decreased the need for enucleation, but it has not substantially impacted the mortality rate. Unfortunately, in spite of reduction trend in enucleation, systemic and targeted chemotherapy was unable to save the affected globe in nearly half of the patients, even when the malignancy was diagnosed in the earlier stages.

Diabetic macular edema (DME) is one of the leading causes of blindness in diabetic retinopathy. As a domestically developed fusion protein drug in China, the anti-vascular endothelial growth factor (VEGF) agent Conbercept has demonstrated significant efficacy in the treatment of DME in recent years. This systematic review explores the mechanism of action and latest research advancements of Conbercept in DME treatment, integrating clinical trial data and comparisons with other anti-VEGF therapies. It further discusses existing limitations in current research and proposes future research directions.

Background: Age-related macular degeneration (AMD), a leading cause of vision loss in elderly individuals, is a multifactorial disease driven by genetic, environmental, and cellular aging processes. Emerging evidence highlights the critical role of ribonucleic acid (RNA) splicing dysfunction in AMD pathogenesis, with a focus on the U1 small nuclear ribonucleoprotein (U1 snRNP) complex, a key spliceosome component. U1 snRNPs ensure the fidelity of RNA cotranscription and pre-mRNA splicing initiation, and their dysfunction has been implicated in neurodegenerative disorders and other age-related diseases.

Main body: This narrative review explores the impact of U1 snRNP dysregulation on retinal cells, focusing on its role in transcriptomic instability, impaired protein homeostasis, cellular stress, impaired autophagy, and inflammation, which are important features of AMD pathogenesis. Finally, we propose that targeting U1 snRNP dysfunction could provide a novel therapeutic approach to slow, prevent, or restore retinal degeneration, offering insights into broader implications for age-related diseases.

Short conclusion: Understanding the molecular mechanisms underlying U1 snRNP dynamics in retinal health and degeneration is essential for developing innovative and effective treatments for AMD, which may provide ways to delay or reverse the effects of aging and associated diseases.