Christopher Bowd

Jalil Jalili, Hossein Taghizad, Anuwat Jiravarnsirikul et al.

Large language models (LLMs) show promise in medical image interpretation but suffer from hallucination, limited accuracy, and run-to-run inconsistency. We developed and validated an agentic AI framework integrating LLMs with specialized deep learning tools for glaucoma detection from fundus photography. The workflow had three steps: (1) LLM initial assessment; (2) function calling to invoke specialized tools for image quality (QAModel, FundaQ-8), glaucoma classification (SwinV2-Tiny), and optic disc/cup segmentation (SegFormer-B0); and (3) LLM reflection integrating the initial impression with tool outputs. Two LLMs (Gemini 2.5 Flash, GPT-5.4 mini) were evaluated on two public datasets (ORIGA, n=100; RIM-ONE-v3, n=100) under uncropped and cropped fields of view; all images were independently graded by a masked fellowship-trained glaucoma specialist. The agentic workflow improved classification accuracy by 16 to 47 percentage points across all conditions, reaching within 6 points of the specialist; on RIM-ONE-v3 the best configurations matched the specialist accuracy of 88%. LLM-alone approaches failed in two ways: GPT-5.4 mini showed positive bias (sensitivity 95-100%, specificity 0-5%), while Gemini 2.5 Flash varied stochastically between runs; the agentic workflow corrected both. Cup-to-disc ratio error fell 15-50% (MAE 0.156-0.228 to 0.104-0.132), and correlation with specialist grading rose from weak (r=0.12-0.39) to moderate-strong (r=0.59-0.84). Run-to-run consistency rose from near-random (kappa as low as -0.01) to near-perfect (kappa up to 0.96). Integrating LLMs with specialized tools addressed key limitations of LLM-alone approaches, including over-diagnosis and run-to-run variability. Gains held for both LLMs, suggesting generalizability across backbones, and may signal a shift from monolithic models toward orchestrated multi-agent systems in medical AI.

Jalil Jalili, Yashraj Gavhane, Evan Walker et al.

Objective: To develop an explainable multimodal large language model (MM-LLM) that (1) screens optic nerve head (ONH) OCT circle scans for quality and (2) generates structured clinical reports that include glaucoma diagnosis and sector-wise retinal nerve fiber layer (RNFL) thinning assessments. Design: Retrospective cohort study of 1,310 subjects contributing 43,849 Spectralis ONH OCT circle scans (1,331 glaucomatous and 867 healthy eyes) from the DIGS and ADAGES cohorts. Methods: A MM-LLM (Llama 3.2 Vision-Instruct model) was fine-tuned to generate clinical descriptions of OCT imaging data. Training data included paired OCT images and automatically generated, structured clinical reports that described global and sectoral RNFL thinning. Poor-quality scans were labeled as unusable and paired with a fixed refusal statement. The model was evaluated on a held-out test set for three tasks: quality assessment, glaucoma detection, and RNFL thinning classification across seven anatomical sectors. Evaluation metrics included accuracy, sensitivity, specificity, precision, and F1-score. Model description quality was also evaluated using standard text evaluation metrics. Results: The model achieved 0.90 accuracy and 0.98 specificity for quality triage. For glaucoma detection, accuracy was 0.86 (sensitivity 0.91, specificity 0.73, F1-score 0.91). RNFL thinning prediction accuracy ranged from 0.83 to 0.94, with highest performance in global and temporal sectors. Text generation scores showed strong alignment with reference reports (BLEU: 0.82; ROUGE-1: 0.94; ROUGE-2: 0.87; ROUGE-L: 0.92; BERTScore-F1: 0.99). Conclusions: The fine-tuned MM-LLM generated accurate clinical descriptions based on OCT imaging. The model achieved high accuracy in identifying image quality issues and detecting glaucoma. The model also provided sectoral descriptions of RNFL thinning to help support clinical OCT evaluation.

Rui Fan, Christopher Bowd, Nicole Brye et al.

Convolutional neural networks (CNNs) are a promising technique for automated glaucoma diagnosis from images of the fundus, and these images are routinely acquired as part of an ophthalmic exam. Nevertheless, CNNs typically require a large amount of well-labeled data for training, which may not be available in many biomedical image classification applications, especially when diseases are rare and where labeling by experts is costly. This article makes two contributions to address this issue: (1) It extends the conventional Siamese network and introduces a training method for low-shot learning when labeled data are limited and imbalanced, and (2) it introduces a novel semi-supervised learning strategy that uses additional unlabeled training data to achieve greater accuracy. Our proposed multi-task Siamese network (MTSN) can employ any backbone CNN, and we demonstrate with four backbone CNNs that its accuracy with limited training data approaches the accuracy of backbone CNNs trained with a dataset that is 50 times larger. We also introduce One-Vote Veto (OVV) self-training, a semi-supervised learning strategy that is designed specifically for MTSNs. By taking both self-predictions and contrastive predictions of the unlabeled training data into account, OVV self-training provides additional pseudo labels for fine-tuning a pre-trained MTSN. Using a large (imbalanced) dataset with 66,715 fundus photographs acquired over 15 years, extensive experimental results demonstrate the effectiveness of low-shot learning with MTSN and semi-supervised learning with OVV self-training. Three additional, smaller clinical datasets of fundus images acquired under different conditions (cameras, instruments, locations, populations) are used to demonstrate the generalizability of the proposed methods.