Post

New research · Ophthalmology
NPJ digital medicine · 5d
AI / informaticsNPJ digital medicine · 2026

Development of a multimodal large language model for early warning and diagnosis of chronic ocular GVHD.

Shuwan Liu, Lili Cao, Haoran Wu … Jing Hong
Read paper
OphthalmologyAI / informatics

AI diagnostic tool detects chronic ocular graft-versus-host disease with high accuracy

Development of a multimodal large language model for early warning and diagnosis of chronic ocular GVHD.

Shuwan Liu … Jing Hong
NPJ digital medicine · 2026
Background

Chronic ocular graft-versus-host disease (coGVHD) after allogeneic hematopoietic stem cell transplantation (allo-HSCT) may lead to irreversible ocular surface damage and even vision loss.

Purpose

This study aimed to leverage a multimodal large language model (multimodal large language model) to develop an early warning and diagnostic system for coGVHD.

Methods

A total of 666 post-allo-HSCT patients (early warning model) and 805 post-allo-HSCT patients (1574 eyes, diagnostic model) were enrolled for construction, internal validation, and external validation of the corresponding models.

n = 1574 eyes
Results

how accurately the AI told coGVHD patients apart from those without it

AUROC 93.44 (95% CI 91.85 to 95.03)
null = 091.8595.03
CI excludes the null - significant
More results

Current management of coGVHD faces challenges, with frequent missed or misdiagnosed cases.

In external validation, the early warning AUROC was 83.45%, while diagnostic AUROCs across three external sites were all above 96.0%.

“
Conclusion

The GVHD-MLLM can process rich multi-modal information collected in clinical practice, and is expected to become an effective tool for managing coGVHD.

Read paper
0 comments

No comments yet. Be the first.

Related papers

LatestFoundational
AI / Informatics
0·962 for OCT
MerMED-FM was very accurate at diagnosing diseases using eye scans
AI / Informatics
0.98
Artificial intelligence's eyelid-height measurements matched doctors' manual measurements almost perfectly
AI / Informatics
92.1%
combining eye scans and photos correctly told benign from cancerous lesions apart nearly every time
Cohort Study
28.6%
recurred locally in more than 1 in 4 patients over years of follow-up
Cohort Study
-0.395
excision group's post-op eyelid fullness score was lower - greater improvement
Cohort Study
86.7%
of infants probed after 12 months still had unresolved tear duct blockage
Observational
16%
eyelid tissue in rosacea patients showed less of this key repair-signaling protein inside cell nuclei
Cohort Study
25%
about 1 in 4 treated cases had symptoms return after improving