Industrial Internet of Things (IIoT) faces increasing security risks with wide applications. Compared with the Internet, the IIoT has a broader attack surface and unique structural characteristics, posing difficulties in directly transferring the previous vulnerability analysis techniques. This paper proposes a triple-stage vulnerability analysis framework (TS-VulA) for IIoT via attack graphs combining ModernBERT and multi-layer heterogeneous networks, which includes three stages. In the first stage, the SentenceBERT based on ModernBERT and IIoT disruption losses are combined to conduct a single-node vulnerability assessment from the aspects of likelihood and criticality of vulnerability exploitation. In the second stage, an IIoT device importance calculation based on multi-layer heterogeneous network theory is proposed, which can acquire the inherent relationships among various IIoT devices. Stage 3 extends attack graph rules for IIoT, then computes node priorities by integrating vulnerability assessment and device importance, which can guide the mitigation strategy. Extensive experiments demonstrate that the proposed vulnerability assessment method achieves an average accuracy and average precision of 87.86% and 87.33%, both exceeding the existing methods. Simulated case studies illustrate that the proposed TS-VulA outperforms the prevailing vulnerability analysis methods.

The identification of lesions is critical for the diagnostic evaluation of prostate cancer.68Ga-PSMA-11 PET/CT is a specific imaging for prostate cancer. However, this is extremely challenging considering the large number of lesions of varying size and uptake that may be distributed in various anatomical settings with different backgrounds throughout the body. In this paper, we propose a deep learning approach for automatic detection of whole-body prostate cancer lesions on PSMA imaging. We established and evaluated our model on the68Ga-PSMA-11 PET/CT image dataset of 107 patients with metastatic prostate cancer, and finally obtained Precision, Recall and F1-score of 82.9%, 100% and 90.6%, respectively, on the independent test set. Preliminary tests confirmed the potential of our method for disease detection on a systemic scale. Increasing the amount of training data can further improve the performance of the proposed deep learning method.