Idioma:

A novel transformer-based aggregation model for predicting gene mutations in lung adenocarcinoma

Sun, Kai ; Zheng, Yuanjie ; Yang, Xinbo ; Jia, Weikuan

Medical & biological engineering & computing, 2024-05, Vol.62 (5), p.1427-1440 [Periódico revisado por pares]

Berlin/Heidelberg: Springer Berlin Heidelberg

Texto completo disponível

Citações Citado por

Enviar para

Título:
A novel transformer-based aggregation model for predicting gene mutations in lung adenocarcinoma
Autor: Sun, Kai ; Zheng, Yuanjie ; Yang, Xinbo ; Jia, Weikuan
Assuntos: Adenocarcinoma ; Adenocarcinoma - genetics ; Adenocarcinoma of Lung - genetics ; Aggregation ; Biomedical and Life Sciences ; Biomedical Engineering and Bioengineering ; Biomedicine ; Cancer ; Computer Applications ; Computer networks ; Electric Power Supplies ; Genetic screening ; Genetic testing ; Genomes ; Human Physiology ; Humans ; Imaging ; Learning ; Lung cancer ; Lung Neoplasms - diagnostic imaging ; Lung Neoplasms - genetics ; Lungs ; Medical imaging ; Mutation ; Mutation - genetics ; Neoplasms ; Original Article ; p53 Protein ; Pathology ; Patients ; Radiology ; Semantics ; Training ; Transformers ; Tumors ; Vectors
É parte de: Medical & biological engineering & computing, 2024-05, Vol.62 (5), p.1427-1440
Notas: ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
Descrição: In recent years, predicting gene mutations on whole slide imaging (WSI) has gained prominence. The primary challenge is extracting global information and achieving unbiased semantic aggregation. To address this challenge, we propose a novel Transformer-based aggregation model, employing a self-learning weight aggregation mechanism to mitigate semantic bias caused by the abundance of features in WSI. Additionally, we adopt a random patch training method, which enhances model learning richness by randomly extracting feature vectors from WSI, thus addressing the issue of limited data. To demonstrate the model’s effectiveness in predicting gene mutations, we leverage the lung adenocarcinoma dataset from Shandong Provincial Hospital for prior knowledge learning. Subsequently, we assess TP53, CSMD3, LRP1B, and TTN gene mutations using lung adenocarcinoma tissue pathology images and clinical data from The Cancer Genome Atlas (TCGA). The results indicate a notable increase in the AUC (Area Under the ROC Curve) value, averaging 4%, attesting to the model’s performance improvement. Our research offers an efficient model to explore the correlation between pathological image features and molecular characteristics in lung adenocarcinoma patients. This model introduces a novel approach to clinical genetic testing, expected to enhance the efficiency of identifying molecular features and genetic testing in lung adenocarcinoma patients, ultimately providing more accurate and reliable results for related studies. Graphical abstract Novel aggregation model based on transformer. The WSI selects tumor patches via the tumor classification model and then inputs them into the Swin Transformer backbone network to extract tumor vectors. The extracted vectors are trained using the random patch training method, and then processed through a transformer encoder and weighted pooling layer to converge into a feature vector that represents global information. The predicted results are then outputted.
Editor: Berlin/Heidelberg: Springer Berlin Heidelberg
Idioma: Inglês

Links

View this record in MEDLINE/PubMed

Voltar para lista de resultados

Anterior Resultado 6 Avançar Ir para próxima página

Realização: Logos de Redes Sociais:

A novel transformer-based aggregation model for predicting gene mutations in lung adenocarcinoma

Sun, Kai ; Zheng, Yuanjie ; Yang, Xinbo ; Jia, Weikuan

Berlin/Heidelberg: Springer Berlin Heidelberg

Buscando em bases de dados remotas. Favor aguardar.