A feasibility study on the adoption of a generative denoising diffusion model for the synthesis of fundus photographs using a small dataset

Deleted Journal, Journal Year: 2024, Volume and Issue: 6(4)

Published: April 3, 2024

Abstract The generative diffusion model has been highlighted as a state-of-the-art artificial intelligence technique for image synthesis. Here, we show that denoising probabilistic (DDPM) can be used domain-specific task generating fundus photographs based on limited training dataset in an unconditional manner. We trained the DDPM U-Net backbone architecture, which is most popular form of model. After training, serial multiple U-Nets generate FPs using random noise seeds. A thousand healthy retinal images were to train input size was set pixel resolution 128 × 128. successfully generated synthetic with pixels our small dataset. failed 256-by-256-pixel due computation capacity personal cloud platform. In comparative analysis, progressive growing adversarial network (PGGAN) synthesized more sharpened than vessels and optic discs. PGGAN (Frechet inception distance [FID] score: 41.761) achieved better FID score (FID 65.605). synthesize relatively Because disadvantages dataset, including difficulty low quality compared networks such PGGAN, further studies are needed improve models medical tasks numbers samples.

Language: Английский

---

Retina Fundus Photograph-Based Artificial Intelligence Algorithms in Medicine: A Systematic Review

Ophthalmology and Therapy, Journal Year: 2024, Volume and Issue: 13(8), P. 2125 - 2149

Published: June 24, 2024

We conducted a systematic review of research in artificial intelligence (AI) for retinal fundus photographic images. highlighted the use various AI algorithms, including deep learning (DL) models, application ophthalmic and non-ophthalmic (i.e., systemic) disorders. found that algorithms interpretation images, compared to clinical data physician experts, represents an innovative solution with demonstrated superior accuracy identifying many (e.g., diabetic retinopathy (DR), age-related macular degeneration (AMD), optic nerve disorders), disorders dementia, cardiovascular disease). There has been significant amount imaging this research, leading potential incorporation DL automated analysis. transform healthcare by improving accuracy, speed, workflow, lowering cost, increasing access, reducing mistakes, transforming worker education training.

Language: Английский

---

Exploring large language model for next generation of artificial intelligence in ophthalmology

Frontiers in Medicine, Journal Year: 2023, Volume and Issue: 10

Published: Nov. 23, 2023

In recent years, ophthalmology has advanced significantly, thanks to rapid progress in artificial intelligence (AI) technologies. Large language models (LLMs) like ChatGPT have emerged as powerful tools for natural processing. This paper finally includes 108 studies, and explores LLMs’ potential the next generation of AI ophthalmology. The results encompass a diverse range studies field ophthalmology, highlighting versatile applications LLMs. Subfields general retinal diseases, anterior segment glaucoma, ophthalmic plastics. Results show competence generating informative contextually relevant responses, potentially reducing diagnostic errors improving patient outcomes. Overall, this study highlights promising role shaping AI’s future By leveraging AI, ophthalmologists can access wealth information, enhance accuracy, provide better care. Despite challenges, continued advancements ongoing research will pave way AI-assisted practices.

Language: Английский

---

Latest developments of generative artificial intelligence and applications in ophthalmology

Asia-Pacific Journal of Ophthalmology, Journal Year: 2024, Volume and Issue: 13(4), P. 100090 - 100090

Published: July 1, 2024

The emergence of generative artificial intelligence (AI) has revolutionized various fields. In ophthalmology, AI the potential to enhance efficiency, accuracy, personalization and innovation in clinical practice medical research, through processing data, streamlining documentation, facilitating patient-doctor communication, aiding decision-making, simulating trials. This review focuses on development integration models into workflows scientific research ophthalmology. It outlines need for a standard framework comprehensive assessments, robust evidence, exploration multimodal capabilities intelligent agents. Additionally, addresses risks model application service including data privacy, bias, adaptation friction, over interdependence, job replacement, based which we summarized risk management mitigate these concerns. highlights transformative enhancing patient care, improving operational efficiency also advocates balanced approach its adoption.

Language: Английский

---

Application of Artificial Intelligence in Ophthalmology: An Updated Comprehensive Review

Journal of Ophthalmic and Vision Research, Journal Year: 2024, Volume and Issue: unknown

Published: Sept. 16, 2024

Artificial intelligence (AI) holds immense promise for transforming ophthalmic care through automated screening, precision diagnostics, and optimized treatment planning. This paper reviews recent advances challenges in applying AI techniques such as machine learning deep to major eye diseases. In diabetic retinopathy, algorithms analyze retinal images accurately identify lesions, which helps clinicians ophthalmology practice. Systems like IDx-DR (IDx Technologies Inc, USA) are FDA-approved autonomous detection of referable retinopathy. For glaucoma, models assess optic nerve head morphology fundus photographs detect damage. age-related macular degeneration, can quantify drusen diagnose disease severity from both color optical coherence tomography images. has also been used screening retinopathy prematurity, keratoconus, dry disease. Beyond aid decisions by forecasting progression anti-VEGF response. However, potential limitations the quality diversity training data, lack rigorous clinical validation, regulatory approval clinician trust must be addressed widespread adoption AI. Two other significant hurdles include integration into existing workflows ensuring transparency decision-making processes. With continued research address these limitations, promises enable earlier diagnosis, resource allocation, personalized treatment, improved patient outcomes. Besides, synergistic human-AI systems could set a new standard evidence-based, precise care.

Language: Английский

---

Evaluation of Convolutional Neural Networks (CNNs) in Identifying Retinal Conditions Through Classification of Optical Coherence Tomography (OCT) Images

Cureus, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 7, 2025

Introduction Diabetic retinopathy (DR) is a leading cause of blindness globally, emphasizing the urgent need for efficient diagnostic tools. Machine learning, particularly convolutional neural networks (CNNs), has shown promise in automating diagnosis retinal conditions with high accuracy. This study evaluates two CNN models, VGG16 and InceptionV3, classifying optical coherence tomography (OCT) images into four categories: normal, choroidal neovascularization, diabetic macular edema (DME), drusen. Methods Using 83,000 OCT across categories, CNNs were trained tested via Python-based libraries, including TensorFlow Keras. Metrics such as accuracy, sensitivity, specificity analyzed confusion matrices performance graphs. Comparisons dataset sizes evaluated impact on model accuracy tools deployed JupyterLab. Results InceptionV3 achieved between 85% 95%, peaking at 94% outperforming (92%). Larger datasets improved sensitivity by 7% all highest normal drusen classifications. like positively correlated size. Conclusions The confirms CNNs' potential diagnostics, achieving classification Limitations included reliance grayscale computational intensity, which hindered finer distinctions. Future work should integrate data augmentation, patient-specific variables, lightweight architectures to optimize clinical use, reducing costs improving outcomes.

Language: Английский

---

Comparison of ChatGPT-4o, Google Gemini 1.5 Pro, Microsoft Copilot Pro, and Ophthalmologists in the management of uveitis and ocular inflammation: A comparative study of large language models

Journal Français d Ophtalmologie, Journal Year: 2025, Volume and Issue: 48(4), P. 104468 - 104468

Published: March 13, 2025

Language: Английский

---

Lights and Shadows on Artificial Intelligence in Glaucoma: Transforming Screening, Monitoring, and Prognosis

Journal of Clinical Medicine, Journal Year: 2025, Volume and Issue: 14(7), P. 2139 - 2139

Published: March 21, 2025

Background/Objectives: Artificial intelligence (AI) is increasingly being integrated into medicine, including ophthalmology, owing to its strong capabilities in image recognition. Methods: This review focuses on the most recent key applications of AI diagnosis and management of, as well research on, glaucoma by performing a systematic latest papers literature. Results: In glaucoma, can help analyze large amounts data from diagnostic tools, such fundus images, optical coherence tomography scans, visual field tests. Conclusions: technologies enhance accuracy diagnoses could provide significant economic benefits automating routine tasks, improving accuracy, enhancing access care, especially underserved areas. However, despite these promising results, challenges persist, limited dataset size diversity, class imbalance, need optimize models for early detection, integration multimodal clinical practice. Currently, ophthalmologists are expected continue playing leading role managing glaucomatous eyes overseeing development validation tools.

Language: Английский

---

GRAPE: A multi-modal dataset of longitudinal follow-up visual field and fundus images for glaucoma management

Scientific Data, Journal Year: 2023, Volume and Issue: 10(1)

Published: Aug. 5, 2023

As one of the leading causes irreversible blindness worldwide, glaucoma is characterized by structural damage and functional loss. Glaucoma patients often have a long follow-up prognosis prediction an important part in treatment. However, existing public datasets are almost cross-sectional, concentrating on segmentation optic disc (OD) diagnosis. With development artificial intelligence (AI), deep learning model can already provide accurate future visual field (VF) its progression with support longitudinal datasets. Here, we proposed real-world appraisal ensemble (GRAPE) dataset. The GRAPE dataset contains 1115 records from 263 eyes, VFs, fundus images, OCT measurements clinical information, OD VF annotated. Two baseline models demonstrated feasibility progression. This will advance AI research management.

Language: Английский

---

Integration of smartphone technology and artificial intelligence for advanced ophthalmic care: A systematic review

Advances in Ophthalmology Practice and Research, Journal Year: 2024, Volume and Issue: 4(3), P. 120 - 127

Published: March 25, 2024

The convergence of smartphone technology and artificial intelligence (AI) has revolutionized the landscape ophthalmic care, offering unprecedented opportunities for diagnosis, monitoring, management ocular conditions. Nevertheless, there is a lack systematic studies on discussing integration AI in this field. This review includes 52 studies, explores smartphones ophthalmology, delineating its collective impact screening methodologies, disease detection, telemedicine initiatives, patient management. findings from curated indicate promising performance smartphone-based various diseases which encompass major retinal diseases, glaucoma, cataract, visual impairment children surface diseases. Moreover, utilization imaging modalities, coupled with algorithms, able to provide timely, efficient cost-effective pathologies. modality can also facilitate self-monitoring, remote monitoring enhancing accessibility eye care services, particularly underserved regions. Challenges involving data privacy, algorithm validation, regulatory frameworks issues trust are still need be addressed. Furthermore, evaluation real-world implementation imperative as well, prospective currently lacking. Smartphone merged enables earlier, precise diagnoses, personalized treatments, enhanced service care. Collaboration crucial navigate ethical security challenges while responsibly leveraging these innovations, potential revolution access global health equity.

Language: Английский

---