Ukwuoma Chiagoziem C, Hossain Md Altab, Jackson Jehoiada K, Nneji Grace U, Monday Happy N, Qin Zhiguang
School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu 610054, China.
School of Management and Economics, University of Electronic Science and Technology of China, Chengdu 610054, China.
Diagnostics (Basel). 2022 May 5;12(5):1152. doi: 10.3390/diagnostics12051152.
INTRODUCTION AND BACKGROUND: Despite fast developments in the medical field, histological diagnosis is still regarded as the benchmark in cancer diagnosis. However, the input image feature extraction that is used to determine the severity of cancer at various magnifications is harrowing since manual procedures are biased, time consuming, labor intensive, and error-prone. Current state-of-the-art deep learning approaches for breast histopathology image classification take features from entire images (generic features). Thus, they are likely to overlook the essential image features for the unnecessary features, resulting in an incorrect diagnosis of breast histopathology imaging and leading to mortality. METHODS: This discrepancy prompted us to develop DEEP_Pachi for classifying breast histopathology images at various magnifications. The suggested DEEP_Pachi collects global and regional features that are essential for effective breast histopathology image classification. The proposed model backbone is an ensemble of DenseNet201 and VGG16 architecture. The ensemble model extracts global features (generic image information), whereas DEEP_Pachi extracts spatial information (regions of interest). Statistically, the evaluation of the proposed model was performed on publicly available dataset: BreakHis and ICIAR 2018 Challenge datasets. RESULTS: A detailed evaluation of the proposed model's accuracy, sensitivity, precision, specificity, and f1-score metrics revealed the usefulness of the backbone model and the DEEP_Pachi model for image classifying. The suggested technique outperformed state-of-the-art classifiers, achieving an accuracy of 1.0 for the benign class and 0.99 for the malignant class in all magnifications of BreakHis datasets and an accuracy of 1.0 on the ICIAR 2018 Challenge dataset. CONCLUSIONS: The acquired findings were significantly resilient and proved helpful for the suggested system to assist experts at big medical institutions, resulting in early breast cancer diagnosis and a reduction in the death rate.
引言与背景：尽管医学领域发展迅速，但组织学诊断仍是癌症诊断的金标准。然而，用于在不同放大倍数下确定癌症严重程度的输入图像特征提取过程十分棘手，因为手动操作存在偏差、耗时、劳动强度大且容易出错。当前用于乳腺组织病理学图像分类的最先进深度学习方法从整个图像中提取特征（通用特征）。因此，它们很可能会忽略重要的图像特征而关注不必要的特征，从而导致乳腺组织病理学成像的错误诊断并导致死亡。 方法：这种差异促使我们开发了DEEP_Pachi，用于对不同放大倍数的乳腺组织病理学图像进行分类。所提出的DEEP_Pachi收集有效乳腺组织病理学图像分类所需的全局和局部特征。所提出的模型主干是DenseNet201和VGG16架构的集成。集成模型提取全局特征（通用图像信息），而DEEP_Pachi提取空间信息（感兴趣区域）。从统计学角度，在所公开的数据集BreakHis和ICIAR 2018挑战赛数据集上对所提出的模型进行了评估。 结果：对所提出模型的准确率、灵敏度、精确率、特异性和F1分数指标的详细评估揭示了主干模型和DEEP_Pachi模型在图像分类方面的有效性。所提出的技术优于最先进的分类器，在BreakHis数据集的所有放大倍数下，良性类别的准确率达到1.0，恶性类别的准确率达到0.99，在ICIAR 2018挑战赛数据集上的准确率为1.0。 结论：所获得的结果具有显著的稳定性，并被证明有助于所建议的系统辅助大型医疗机构的专家，从而实现早期乳腺癌诊断并降低死亡率。
Artif Intell Med. 2021-11
Diagnostics (Basel). 2024-7-1
Diagnostics (Basel). 2022-12-30
Med Biol Eng Comput. 2024-5
J King Saud Univ Comput Inf Sci. 2023-7
Diagnostics (Basel). 2023-2-11
Diagnostics (Basel). 2022-12-29
Artif Intell Med. 2021-11
Nucl Med Rev Cent East Eur. 2021
Health Inf Sci Syst. 2020-11-4
Curr Med Imaging. 2020
Zotero 插件
