Lee Gi Pyo, Kim Young Jae, Park Dong Kyun, Kim Yoon Jae, Han Su Kyeong, Kim Kwang Gi
Department of Health Sciences and Technology, Gachon Advanced Institute for Health Sciences and Technology (GAIHST), Gachon University, Incheon 21565, Republic of Korea.
Department of Biomedical Engineering, Gachon University Gil Medical Center, College of Medicine, Gachon University, Incheon 21565, Republic of Korea.
Diagnostics (Basel). 2023 Dec 28;14(1):75. doi: 10.3390/diagnostics14010075.
Most of the development of gastric disease prediction models has utilized pre-trained models from natural data, such as ImageNet, which lack knowledge of medical domains. This study proposes Gastro-BaseNet, a classification model trained using gastroscopic image data for abnormal gastric lesions. To prove performance, we compared transfer-learning based on two pre-trained models (Gastro-BaseNet and ImageNet) and two training methods (freeze and fine-tune modes). The effectiveness was verified in terms of classification at the image-level and patient-level, as well as the localization performance of lesions. The development of Gastro-BaseNet had demonstrated superior transfer learning performance compared to random weight settings in ImageNet. When developing a model for predicting the diagnosis of gastric cancer and gastric ulcers, the transfer-learned model based on Gastro-BaseNet outperformed that based on ImageNet. Furthermore, the model's performance was highest when fine-tuning the entire layer in the fine-tune mode. Additionally, the trained model was based on Gastro-BaseNet, which showed higher localization performance, which confirmed its accurate detection and classification of lesions in specific locations. This study represents a notable advancement in the development of image analysis models within the medical field, resulting in improved diagnostic predictive accuracy and aiding in making more informed clinical decisions in gastrointestinal endoscopy.
大多数胃病预测模型的开发都利用了来自自然数据（如图像网）的预训练模型，而这些模型缺乏医学领域的知识。本研究提出了胃基础网络（Gastro - BaseNet），这是一种使用胃镜图像数据训练的用于异常胃部病变的分类模型。为了证明其性能，我们基于两个预训练模型（胃基础网络和图像网）以及两种训练方法（冻结和微调模式）进行了迁移学习比较。在图像级别和患者级别分类以及病变定位性能方面验证了其有效性。胃基础网络的开发已证明与图像网中的随机权重设置相比具有卓越的迁移学习性能。在开发用于预测胃癌和胃溃疡诊断的模型时，基于胃基础网络的迁移学习模型优于基于图像网的模型。此外，在微调模式下对整个层进行微调时，该模型的性能最高。此外，训练后的模型基于胃基础网络，显示出更高的定位性能，这证实了其对特定位置病变的准确检测和分类。这项研究代表了医学领域图像分析模型开发的一项显著进展，提高了诊断预测准确性，并有助于在胃肠内镜检查中做出更明智的临床决策。
