A digital image analysis method and device for labial gland biopsy pathological tissue for evaluating the prognosis of patients with Sjögren's syndrome

Through the digital image analysis method, data preprocessing, patch-based transfer learning and multi-instance attention learning are used to solve the subjectivity and efficiency problems in pathological tissue analysis of labial biopsy, and the diagnostic accuracy and clinical value are improved.

CN118762823BActive Publication Date: 2025-03-25UESTC (SHENZHEN) ADVANCED RES INST
2 Cites 0 Cited by

Patent Information

Application Number
CN202410913883.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-09
Publication Date
2025-03-25
Estimated Expiration
2044-07-09

AI Technical Summary

Technical Problem

The prior art has problems of strong subjectivity, low efficiency and difficulty in explaining the image analysis of labial biopsy pathological tissues, and models based on manual features are difficult to promote in clinical practice.

Method used

Digital image analysis methods are adopted, including data preprocessing, patch-based transfer learning and WSI-based multi-instance attention learning, feature extraction and complication prediction are performed through EfficientNet-B5 model and VIT multi-instance model.

Benefits of technology

It improves the diagnostic accuracy of patients with Sjogren's syndrome, reduces the risk of misdiagnosis and missed diagnosis, helps doctors better evaluate patients' complication risks, provides more effective prevention and treatment measures, and improves patients' quality of life.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present invention discloses a digital image analysis method and device for labial gland biopsy pathological tissue for evaluating the prognosis of patients with Sjögren's syndrome. The digital image analysis method includes data preprocessing, patch-based transfer learning, WSI-based multi-instance attention learning, and model prediction of the probability of complications. The above analysis method can improve the diagnostic accuracy of patients with Sjögren's syndrome, reduce the risk of misdiagnosis and missed diagnosis, help doctors better assess the risk of complications in patients with Sjögren's syndrome, provide patients with more effective prevention and treatment measures, improve the prognosis of patients with Sjögren's syndrome, and improve the quality of life of patients. It also has significant technical advancement and clinical value, and can provide new ideas and methods for the diagnosis and treatment of patients with Sjögren's syndrome.
Need to check novelty before this filing date? Find Prior Art