Department of Pathology and Laboratory Medicine, Taichung Veterans General Hospital, Taichung, Taiwan, R.O.C.
Department of Electrical Engineering, National Cheng Kung University, Tainan, Taiwan, R.O.C.
In Vivo. 2024 Sep-Oct;38(5):2239-2244. doi: 10.21873/invivo.13688.
BACKGROUND/AIM: In this study, we introduce an innovative deep-learning model architecture aimed at enhancing the accuracy of detecting and classifying organizing pneumonia (OP), a condition characterized by the presence of Masson bodies within the alveolar spaces due to lung injury. The variable morphology of Masson bodies and their resemblance to adjacent pulmonary structures pose significant diagnostic challenges, necessitating a model capable of discerning subtle textural and structural differences. Our model incorporates a novel architecture that integrates advancements in three key areas: Semantic segmentation, texture analysis, and structural feature recognition.
We employed a dataset of whole slide imaging from 20 patients, totaling 100 slides of OP, segmented into training, validation, and testing sets to reflect real-world application scenarios. Our approach utilizes a modified multi-head self-attention mechanism combined with ResUNet for semantic segmentation, enhanced by superpixel concepts. This method facilitates the generation of representative token features through iterative super-token blocks, creating high-resolution token maps that leverage local and high-level feature information for improved accuracy.
Benefiting from token features and distribution for enhanced texture alignment with fewer false-positives, the super-token transformer (STT) model achieved a mean intersection over union (mIOU) of 72.42%, with a sensitivity of 47.81%, specificity of 99.83%, positive predictive value of 64.03%, and negative predictive value of 99.94%, highlighting superior efficacy in Masson body segmentation in complex cross-tissue analyses.
Our team developed an iterative learning model based on the STT approach, emphasizing token features of super token, including texture and distribution, that enable enhanced alignment with the unique textures of Masson bodies to improve sensitivity and mIOU, The development of this STT model presents a significant advancement in the field of medical image analysis for OP that offers a promising avenue for improving diagnostic precision and patient outcomes in pulmonary pathology.
背景/目的：本研究引入了一种创新的深度学习模型架构，旨在提高检测和分类特发性间质性肺炎（OP）的准确性。OP 是一种肺泡腔中存在 Masson 小体的疾病，其特征是由于肺损伤导致 Masson 小体的形态多变且与相邻肺结构相似，这给诊断带来了很大的挑战，需要一个能够辨别细微纹理和结构差异的模型。我们的模型采用了一种新的架构，该架构集成了三个关键领域的进展：语义分割、纹理分析和结构特征识别。
我们使用了来自 20 名患者的全切片成像数据集，总共包含 100 张 OP 切片，将其划分为训练集、验证集和测试集，以反映实际应用场景。我们的方法采用了一种改进的多头自注意力机制，结合 ResUNet 进行语义分割，并通过超像素概念进行增强。该方法通过迭代的超像素块生成具有代表性的令牌特征，创建高分辨率的令牌图，利用局部和高级特征信息提高准确性。
受益于令牌特征和分布，该模型在较少出现假阳性的情况下实现了更好的纹理对齐，超像素变换（STT）模型的平均交并比（mIOU）为 72.42%，敏感性为 47.81%，特异性为 99.83%，阳性预测值为 64.03%，阴性预测值为 99.94%，在复杂的交叉组织分析中，对 Masson 小体分割具有卓越的效果。
我们的团队开发了一种基于 STT 方法的迭代学习模型，强调超像素的令牌特征，包括纹理和分布，以改善与 Masson 小体独特纹理的对齐，提高敏感性和 mIOU。STT 模型的开发是 OP 医学图像分析领域的重大进展，为提高肺病理学的诊断精度和患者预后提供了有前途的途径。
