Maternal and child health during pregnancy is an important issue in global public health, and the classification accuracy of fetal cardiotocography (CTG), as a key tool for monitoring fetal health during pregnancy, is directly related to the effectiveness of early diagnosis and intervention. Due to the serious category imbalance problem of CTG data, traditional models find it challenging to take into account a small number of categories of samples, increasing the risk of leakage and misdiagnosis. To solve this problem, this paper proposes a two-step innovation: firstly, we design a method of adaptive adjustment of misclassification loss function weights (MAAL), which dynamically identifies and increases the focus on misclassified samples based on misclassification rates. Secondly, a primary and secondary correction network model (MAC-NET) is constructed to carry out secondary correction for the misclassified samples of the primary model. Experimental results show that the method proposed in this paper achieves 99.39% accuracy on the UCI publicly available fetal health dataset, and also obtains excellent performance on other domain imbalance datasets. This demonstrates that the model is not only effective in alleviating the problem of category imbalance, but also has very high clinical utility.
Fetal cardiotocography (CTG) is widely used to assess fetal health during labor and to screen for fetal acidosis. However, CTG interpretation relies heavily on clinicians’ experience and is affected by subjectivity and inconsistency, which limit diagnostic reliability. Most existing artificial intelligence approaches simplify fetal acid–base assessment into a binary classification, making it difficult to distinguish acidosis severity and restricting information for refined clinical decision-making. To address these limitations, this study formulates a three-class classification task—normal, moderate acidosis, and severe acidosis—based on the CTU-CHB dataset, using umbilical artery blood pH as the reference standard. A signal-first, conditionally enhanced, three-phase training and meta-feature fusion framework is proposed. In stage A, a CNN-BiLSTM-attention network performs end-to-end modeling of fetal heart rate signals, while a recall feedback-driven dynamic weighted loss alleviates class imbalance and identifies difficult samples. Stage B incorporates relevant clinical detection information for these difficult samples and applies multimodal feature fusion to enhance discrimination. Stage C constructs meta-features from the outputs of the first two stages to adaptively fuse classification preferences and uncertainty. Experimental results demonstrate that the proposed framework achieves an accuracy of 82.80 ± 2.82% and an F1 score of 78.84 ± 2.96%, effectively mitigating class imbalance and difficult sample classification, and providing reliable support for clinical decision-making in fetal acidosis.