Energy-Efficient Hybrid Impulsive Model for Joint Classification and Segmentation on CT Images

Highly flexible foundation models like artificial neural networks are imperative in medical practice, enabling diverse tasks with little or no task-specific labeled data. The crucial problem remains as how to link latent features and a priori knowledge within multitask decision outputs, particularly in joint classification and segmentation tasks on images. This article develops a hybrid encoder-decoding model substantiating hybrid computations of continuous convolution variables and discrete nerve impulses, where impulsive neurons are adopted to boost nonlinear activations. By presenting a flexible network architecture with regularized multiloss training, this hybrid model can learn shared features of classification and segmentation. The joint decoder does not only provide classification results, but also predicts intelligible task-specific outputs from input images. Applied to the COVID-19 lung CT and the Synapse multiorgan CT datasets, experimental results and ablation studies demonstrate the effectiveness and flexibility of this hybrid model, which outperforms convolution models and human experts. Comparative studies further highlight the high energy-efficient attribute and the decision-output visibility of the hybrid impulsive model, indicating a potential for edge healthcare and biomedical applications.