Deep Convolutional Neural Network for Accurate Classification of Myofibroblastic Lesions on Patch-Based Images.

Autor: Giraldo-Roldán D; Faculdade de Odontologia de Piracicaba, Universidade de Campinas (FOP-UNICAMP), Piracicaba, São Paulo, Brazil. Danygir.2@gmail.com.; Department of Oral Diagnosis, Oral Pathology Area Piracicaba Dental School, University of Campinas (UNICAMP), Av. Limeira, 901, 13.414-903, Piracicaba, São Paulo, Brazil. Danygir.2@gmail.com., Dos Santos GC; Institute of Science and Technology, Federal University of São Paulo (ICT-Unifesp), São José dos Campos, São Paulo, Brazil., Araújo ALD; Head and Neck Surgery Department, University of São Paulo Medical School (FMUSP), São Paulo, Brazil., Nakamura TCR; Institute of Science and Technology, Federal University of São Paulo (ICT-Unifesp), São José dos Campos, São Paulo, Brazil., Pulido-Díaz K; Health Care Department, Oral Pathology and Medicine Master, Autonomous Metropolitan University, Mexico City, Mexico., Lopes MA; Faculdade de Odontologia de Piracicaba, Universidade de Campinas (FOP-UNICAMP), Piracicaba, São Paulo, Brazil., Santos-Silva AR; Faculdade de Odontologia de Piracicaba, Universidade de Campinas (FOP-UNICAMP), Piracicaba, São Paulo, Brazil., Kowalski LP; Head and Neck Surgery Department, University of São Paulo Medical School (FMUSP), São Paulo, Brazil.; Department of Head and Neck Surgery and Otorhinolaryngology, A.C. Camargo Cancer Center, São Paulo, Brazil., Moraes MC; Institute of Science and Technology, Federal University of São Paulo (ICT-Unifesp), São José dos Campos, São Paulo, Brazil., Vargas PA; Faculdade de Odontologia de Piracicaba, Universidade de Campinas (FOP-UNICAMP), Piracicaba, São Paulo, Brazil.
Jazyk: angličtina
Zdroj: Head and neck pathology [Head Neck Pathol] 2024 Oct 28; Vol. 18 (1), pp. 117. Date of Electronic Publication: 2024 Oct 28.
DOI: 10.1007/s12105-024-01723-5
Abstrakt: Objective: This study aimed to implement and evaluate a Deep Convolutional Neural Network for classifying myofibroblastic lesions into benign and malignant categories based on patch-based images.
Methods: A Residual Neural Network (ResNet50) model, pre-trained with weights from ImageNet, was fine-tuned to classify a cohort of 20 patients (11 benign and 9 malignant cases). Following annotation of tumor regions, the whole-slide images (WSIs) were fragmented into smaller patches (224 × 224 pixels). These patches were non-randomly divided into training (308,843 patches), validation (43,268 patches), and test (42,061 patches) subsets, maintaining a 78:11:11 ratio. The CNN training was caried out for 75 epochs utilizing a batch size of 4, the Adam optimizer, and a learning rate of 0.00001.
Results: ResNet50 achieved an accuracy of 98.97%, precision of 99.91%, sensitivity of 97.98%, specificity of 99.91%, F1 score of 98.94%, and AUC of 0.99.
Conclusions: The ResNet50 model developed exhibited high accuracy during training and robust generalization capabilities in unseen data, indicating nearly flawless performance in distinguishing between benign and malignant myofibroblastic tumors, despite the small sample size. The excellent performance of the AI model in separating such histologically similar classes could be attributed to its ability to identify hidden discriminative features, as well as to use a wide range of features and benefit from proper data preprocessing.
(© 2024. The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.)
Databáze: MEDLINE
Externí odkaz: