EUVIMEDEUVIMED entdecken
Open Access

Integration of Transformer-Based Architecture and Large Language Models for Optical Coherence Tomography Data Analysis to Improve the Accuracy of Differential Diagnosis of Retinal Diseases

O V Konshina, A D Pershin, M K Kulyabin, V I Borisov

Sovremennye tehnologii v medicine · 30.06.2026 · 18 (3) · 5

DOI: 10.17691/stm2026.18.3.01 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sovremennye tehnologii v medicine, online erschienen am 30.06.2026. Verfasst von O V Konshina und 3 weiteren Autor:innen. Beteiligte Einrichtung: Master Student, Educational and Research Center "Artificial Intelligence"; Ural Federal University named after the First President of Russia B.N. Yel…. Hauptthemen laut MEDLINE: Tomography, Optical Coherence, Retinal Diseases. Der Volltext ist frei in PubMed Central verfügbar.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The aim of the present study was to improve the differential diagnosis accuracy of retinal diseases combining a transformer model to classify the biomarkers on OCT images and the large language model DeepSeek-V3. Materials and Methods. Two datasets were collected and annotated: a training set (3288 central retinal OCT images annotated for 8 biomarkers) and a validation set (50 clinical cases from octcases.com). For biomarker classification, we compared ResNet, DenseNet, EfficientNet, and Vision Transformer (ViT-Tiny-Patch16-224) architectures. The ViT-Tiny model demonstrated the highest performance, its F1 macro — 0.84±0.03. An integration algorithm was developed to combine predicted biomarker labels with clinical history data via the DeepSeek-V3 LLM API.

Results

Combining OCT biomarkers and the medical history significantly improved the diagnostic accuracy: Top-1 Accuracy — 78%, Top-3 Accuracy — 94%, MRR — 84%, representing 10–44% improvement over using either data type alone.

Conclusion

The suggested approach enabled to automate the detection of biomarkers on OCT images and enhance the differential diagnosis accuracy of eye diseases, reducing the image interpretation time and supporting clinical decision-making.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Tomography, Optical CoherenceRetinal DiseasesHumansLarge Language ModelsDiagnosis, DifferentialBiomarkers

Stichwörter der Autor:innen

DeepSeekLLM in ophthalmologyOCT biomarker classificationViTdeep learningmulticlass OCT classificationretina

Substanzen

Biomarkers

Autor:innen und Institutionen

  • Master Student, Educational and Research Center "Artificial Intelligence"; Ural Federal University named after the First President of Russia B.N. Yeltsin, Institute of Radioelectronics and Information Technologies - RTF…
  • PhD Student, Educational and Research Center "Artificial Intelligence"; Ural Federal University named after the First President of Russia B.N. Yeltsin, Institute of Radioelectronics and Information Technologies - RTF, 3…
  • PhD, Technical Director; LLC "VisioMedAI", 8 Golovinskoye Shosse, Moscow, 125212, Russia.
  • PhD, Associate Professor, Educational and Research Center "Artificial Intelligence"; Ural Federal University named after the First President of Russia B.N. Yeltsin, Institute of Radioelectronics and Information Technolo…

Zitieren

APA

Konshina, O. V., Pershin, A. D., Kulyabin, M. K., & Borisov, V. I. (2026). Integration of Transformer-Based Architecture and Large Language Models for Optical Coherence Tomography Data Analysis to Improve the Accuracy of Differential Diagnosis of Retinal Diseases. Sovremennye tehnologii v medicine, 18(3), 5. https://doi.org/10.17691/stm2026.18.3.01

Vancouver

Konshina OV, Pershin AD, Kulyabin MK, Borisov VI. Integration of Transformer-Based Architecture and Large Language Models for Optical Coherence Tomography Data Analysis to Improve the Accuracy of Differential Diagnosis of Retinal Diseases. Sovremennye tehnologii v medicine. 2026;18(3):5. doi:10.17691/stm2026.18.3.01

BibTeX
@article{Konshina2026,
  author = {Konshina, O V and Pershin, A D and Kulyabin, M K and Borisov, V I},
  title = {Integration of Transformer-Based Architecture and Large Language Models for Optical Coherence Tomography Data Analysis to Improve the Accuracy of Differential Diagnosis of Retinal Diseases},
  journal = {Sovremennye tehnologii v medicine},
  year = {2026},
  volume = {18},
  number = {3},
  pages = {5},
  doi = {10.17691/stm2026.18.3.01}
}
Kostenlos starten →