Wang Minghan, Li Guangfei, Yang Yimin, Yang Yongxiu, Feng Yongkang, Li Yashuang, Liu Guoli, Hao Dongmei
Department of Biomedical Engineering, College of Chemistry and Life Science, Beijing University of Technology, 219 Life Sciences Building, 100 Pingleyuan, Beijing, 100124 China.
Beijing International Science and Technology Cooperation Base for Intelligent Physiological Measurement and Clinical Transformation, Beijing, China.
Biomed Eng Lett. 2024 May 9;14(5):1037-1048. doi: 10.1007/s13534-024-00388-x. eCollection 2024 Sep.
In clinical practice, obstetricians use visual interpretation of fetal heart rate (FHR) to diagnose fetal conditions, but inconsistencies among interpretations can hinder accuracy. This study introduces MTU-Net3+, a deep learning model designed for automated, multi-task FHR analysis, aiming to improve diagnostic accuracy and efficiency. The proposed MTU-Net3 + was built upon the UNet3 + architecture, incorporating an encoder, a decoder, full-scale skip connections, and a deep supervision module, and further integrates a self-attention mechanism and bidirectional Long Short-Term Memory layers to enhance its performance. The MTU-Net3 + model accepts the preprocessed 20-minute FHR signals as input, outputting categorical probabilities and baseline values for each time point. The proposed MTU-Net3 + model was trained on a subset of a public database, and was tested on the remaining data of the public database and a private database. In the remaining public datasets, this model achieved F1 scores of 84.21% for deceleration (F1.Dec) and 61.33% for acceleration (F1.Acc), with a Root Mean Square Baseline Difference (RMSD.BL) of 3.46 bpm, 0% of points with an absolute difference exceeding 15 bpm(D15bpm), a Synthetic Inconsistency Coefficient (SI) of 44.82%, and a Morphological Analysis Discordance Index (MADI) of 7.00%. On the private dataset, the model recorded an RMSD.BL of 1.37 bpm, 0% D15bpm, F1.Dec of 100%, F1.Acc of 87.50%, an SI of 12.20% and a MADI of 2.79%. The MTU-Net3 + model proposed in this study performed well in automated FHR analysis, demonstrating its potential as an effective tool in the field of fetal health assessment.
在临床实践中，产科医生通过对胎儿心率（FHR）的视觉解读来诊断胎儿状况，但解读之间的不一致会影响准确性。本研究引入了MTU-Net3+，这是一种为自动化多任务FHR分析设计的深度学习模型，旨在提高诊断准确性和效率。所提出的MTU-Net3+基于UNet3+架构构建，包含一个编码器、一个解码器、全尺度跳跃连接和一个深度监督模块，并进一步集成了自注意力机制和双向长短期记忆层以提升其性能。MTU-Net3+模型接受预处理后的20分钟FHR信号作为输入，输出每个时间点的分类概率和基线值。所提出的MTU-Net3+模型在一个公共数据库的子集上进行训练，并在公共数据库的其余数据和一个私有数据库上进行测试。在其余的公共数据集中，该模型减速的F1分数（F1.Dec）为84.21%，加速的F1分数（F1.Acc）为61.33%，均方根基线差异（RMSD.BL）为3.46次/分钟，绝对差异超过15次/分钟的点数比例（D15bpm）为0%，综合不一致系数（SI）为44.82%，形态分析不一致指数（MADI）为7.00%。在私有数据集上，该模型的RMSD.BL为1.37次/分钟，D15bpm为0%，F1.Dec为100%，F1.Acc为87.50%，SI为12.20%，MADI为2.79%。本研究中提出的MTU-Net3+模型在自动化FHR分析中表现良好，证明了其作为胎儿健康评估领域有效工具的潜力。
