Integrated image-based deep learning and language models for primary diabetes care

Nature Medicine, Journal Year: 2024, Volume and Issue: 30(10), P. 2886 - 2896

Published: July 19, 2024

Abstract Primary diabetes care and diabetic retinopathy (DR) screening persist as major public health challenges due to a shortage of trained primary physicians (PCPs), particularly in low-resource settings. Here, bridge the gaps, we developed an integrated image–language system (DeepDR-LLM), combining large language model (LLM module) image-based deep learning (DeepDR-Transformer), provide individualized management recommendations PCPs. In retrospective evaluation, LLM module demonstrated comparable performance PCPs endocrinology residents when tested English outperformed had Chinese. For identifying referable DR, average PCP’s accuracy was 81.0% unassisted 92.3% assisted by DeepDR-Transformer. Furthermore, performed single-center real-world prospective study, deploying DeepDR-LLM. We compared adherence patients under PCP arm ( n = 397) with those PCP+DeepDR-LLM 372). Patients newly diagnosed showed better self-management behaviors throughout follow-up P < 0.05). referral were more likely adhere DR referrals 0.01). Additionally, DeepDR-LLM deployment improved quality empathy level recommendations. Given its multifaceted performance, holds promise digital solution for enhancing screening.

Language: Английский

---

Artificial intelligence for diabetes care: current and future prospects

The Lancet Diabetes & Endocrinology, Journal Year: 2024, Volume and Issue: 12(8), P. 569 - 595

Published: July 23, 2024

Language: Английский

---

Large language models illuminate a progressive pathway to artificial intelligent healthcare assistant

Medicine Plus, Journal Year: 2024, Volume and Issue: 1(2), P. 100030 - 100030

Published: May 17, 2024

With the rapid development of artificial intelligence, large language models (LLMs) have shown promising capabilities in mimicking human-level comprehension and reasoning. This has sparked significant interest applying LLMs to enhance various aspects healthcare, ranging from medical education clinical decision support. However, medicine involves multifaceted data modalities nuanced reasoning skills, presenting challenges for integrating LLMs. review introduces fundamental applications general-purpose specialized LLMs, demonstrating their utilities knowledge retrieval, research support, workflow automation, diagnostic assistance. Recognizing inherent multimodality medicine, emphasizes multimodal discusses ability process diverse types like imaging electronic health records augment accuracy. To address LLMs' limitations regarding personalization complex reasoning, further explores emerging LLM-powered autonomous agents healthcare. Moreover, it summarizes evaluation methodologies assessing reliability safety contexts. transformative potential medicine; however, there is a pivotal need continuous optimizations ethical oversight before these can be effectively integrated into practice.

Language: Английский

---

Large language models for diabetes training: a prospective study

Science Bulletin, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 1, 2025

Language: Английский

---

AI-assisted facial analysis in healthcare: From disease detection to comprehensive management

Patterns, Journal Year: 2025, Volume and Issue: 6(2), P. 101175 - 101175

Published: Feb. 1, 2025

Medical conditions and systemic diseases often manifest as distinct facial characteristics, making identification of these unique features crucial for disease screening. However, detecting using photography remains challenging because the wide variability in human conditions. The integration artificial intelligence (AI) into analysis represents a promising frontier offering user-friendly, non-invasive, cost-effective screening approach. This review explores potential AI-assisted identifying subtle phenotypes indicative health disorders. First, we outline technological framework essential effective implementation healthcare settings. Subsequently, focus on role We further expand our examination to include applications monitoring, support treatment decision-making, follow-up, thereby contributing comprehensive management. Despite its promise, adoption this technology faces several challenges, including privacy concerns, model accuracy, issues with interpretability, biases AI algorithms, adherence regulatory standards. Addressing challenges is ensure fair ethical use. By overcoming hurdles, can empower providers, improve patient care outcomes, enhance global health.

Language: Английский

---

Understanding natural language: Potential application of large language models to ophthalmology

Asia-Pacific Journal of Ophthalmology, Journal Year: 2024, Volume and Issue: 13(4), P. 100085 - 100085

Published: July 1, 2024

Large language models (LLMs), a natural processing technology based on deep learning, are currently in the spotlight. These closely mimic comprehension and generation. Their evolution has undergone several waves of innovation similar to convolutional neural networks. The transformer architecture advancement generative artificial intelligence marks monumental leap beyond early-stage pattern recognition via supervised learning. With expansion parameters training data (terabytes), LLMs unveil remarkable human interactivity, encompassing capabilities such as memory retention comprehension. advances make particularly well-suited for roles healthcare communication between medical practitioners patients. In this comprehensive review, we discuss trajectory their potential implications clinicians For clinicians, can be used automated documentation, given better inputs extensive validation, may able autonomously diagnose treat future. patient care, triage suggestions, summarization documents, explanation patient's condition, customizing education materials tailored level. limitations possible solutions real-world use also presented. Given rapid advancements area, review attempts briefly cover many that play ophthalmic space, with focus improving quality delivery.

Language: Английский

---

Based on Medicine, The Now and Future of Large Language Models

Cellular and Molecular Bioengineering, Journal Year: 2024, Volume and Issue: 17(4), P. 263 - 277

Published: Aug. 1, 2024

Language: Английский

---

Current Research and Future Strategies for the Management of Vision-Threatening Diabetic Retinopathy

Asia-Pacific Journal of Ophthalmology, Journal Year: 2024, Volume and Issue: 13(5), P. 100109 - 100109

Published: Sept. 1, 2024

Diabetic retinopathy (DR) is a major ocular complication of diabetes and the leading cause blindness visual impairment, particularly among adults working-age adults. Although medical economic burden DR significant its global prevalence expected to increase, in low- middle-income countries, large portion vision loss caused by remains preventable through early detection timely intervention. This perspective reviewed latest developments research innovation three areas, first novel biomarkers (including advanced imaging modalities, serum biomarkers, artificial intelligence technology) predict incidence progression DR, second, screening referable vision-threatening (VTDR), finally, therapeutic strategies for VTDR, including diabetic macular oedema (DME), with goal reducing blindness.

Language: Английский

---

A case study on using a large language model to analyze continuous glucose monitoring data

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Jan. 7, 2025

Abstract Continuous glucose monitors (CGM) provide valuable insights about glycemic control that aid in diabetes management. However, interpreting metrics and charts synthesizing them into linguistic summaries is often non-trivial for patients providers. The advent of large language models (LLMs) has enabled real-time text generation summarization medical data. objective this study was to assess the strengths limitations using an LLM analyze raw CGM data produce 14 days with type 1 diabetes. We first evaluated ability GPT-4 compute quantitative specific found Ambulatory Glucose Profile (AGP). Then, two independent clinician graders, we accuracy, completeness, safety, suitability qualitative descriptions produced by across five different analysis tasks. performed 9 out 10 tasks perfect accuracy all cases. clinician-evaluated had good performance measures [lowest task mean score 8/10, highest 10/10], completeness 7.5/10, safety 9.5/10, 10/10]. Our work serves as a preliminary on how generative can be integrated care through and, more broadly, potential leverage LLMs streamlined time series analysis.

Language: Английский

---

Multi-Stage Learning for Intuitive Visualization of Microcystic Macular Edema in OCT Images

Journal of Medical and Biological Engineering, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 4, 2025

Abstract Purpose Detecting and monitoring Microcystic Macular Edema (MME) in Optical Coherence Tomography (OCT) images is vital for early diagnosis of Diabetic (DME), a leading cause blindness developed countries. However, detecting MME remains challenging due to its fuzzy boundaries diffuse nature. In this work, we propose novel fully-automatic methodology based on multi-stage regional learning successfully detect visualize OCT images. Methods Our work divided into two main stages: the first stage coarsely identifies general DME accumulations innermost retinal layers. On other hand, second precisely detects within reduced search space. These detections are then used generate intuitive confidence maps. Results approach achieves mean 0.9618 ± 0.0518 per pixel, demonstrating consistent strong detections. This robust facilitates MME, independent clinicians’ subjectivity, has potential significantly impact quality life patients. Conclusion represents significant advancement automatic analysis complex pathologies. Source code available at: https://github.com/PlacidoFranciscoLizancosVidal/Microcysts_paper_code .

Language: Английский

---