Non-small cell lung cancer (NSCLC) remains one of the leading causes of cancer-related mortality worldwide, with therapeutic progress often limited by late-stage diagnosis and the variable effectiveness of existing treatments. Immune checkpoint inhibitors (ICIs) have emerged as a promising therapeutic approach, but their clinical success depends critically on accurate biomarker-based patient stratification. Current strategies for integrating biomarkers, however, remain fragmented and lack the robustness needed for reliable prediction of treatment outcomes. To address this gap, we propose a novel multimodal deep learning framework that integrates CT images, PET/CT scans, and curated biomarker profiles to improve prediction of ICI treatment response in NSCLC patients. Our model employs a pre-trained Xception encoder for advanced image feature extraction, an Encoder Attention Network for semantic representation learning, and a DenseNet-inspired metadata processor for structured biomarker data. Multimodal features are fused through a Block Dense Attention Convolutional Module with Self-Attention Multi-Head (BDAC-SAMH) mechanism, enabling richer interactions across modalities. Experimental evaluation demonstrates that our framework achieves 94.3% accuracy, 92.1% sensitivity, and 93.5% specificity, significantly outperforming conventional CNN-based unimodal methods. Importantly, the proposed system improves prediction of key ICI biomarkers, including PD-L1, TMB, and MSI, by 15.6% compared to unimodal baselines while revealing novel biomarker interactions. This highlights its potential to guide personalized immunotherapy strategies in NSCLC.
This paper presents a ground-breaking triple-bandpass filter design utilizing a modified T-shape rectangular coupled with a stepped impedance resonator (MTSR-CSIR) for smart portable communication device applications. The MTSR-CSIR filter operates at (2.2, 3.62 and 4.6) GHz, providing wide passbands in three operating modes. To achieve the optimal performance, the filter design is executed on a multi-layered liquid-crystal polymer (LCP) substrate with a thickness of 50 µm, dielectric constant of 2.9 and loss tangent of 0.002. Simulation results for the MTSR-CSIR filter demonstrate a high level of accuracy and consistency with the measurement results for the fabricated filter. The filter exhibits an excellent stopband rejection, low loss and compact size while maintaining high-performance levels. Its performance parameters, such as insertion loss, return loss and group delay, are considered to evaluate the filter’s performance. The results highlight the applicability of the MTSR-CSIR filter for smart portable communication devices.
The monolithic network architectures are used in Bitcoin and other big cryptos. Different hierarchical ways to protect the decentralized blockchain have been discussed to improve its usability. However, no debate in the literature on how an appropriate shard size can be calculated to optimize the output and how deceptive or unreliable nodes can affect the optimum shard size range. This study includes an exchange and authentication scheme for hierarchical blockchain architecture in order to fix these issues. The hierarchy splits the network nodes into many disjoint fragments, with most transfers unregulated between these shards. Taking into account the performance and precision of the transaction validation in the case of malicious or defective nodes, the optimum shard size is calculated. It ensures that most of the good shards are available with the right scale, according to the highest acceptable threshold of defective nodes. It also offers a method to distinguish damaged shards by overlapping a tiny proportion of purchases across all shields and discarding all transactions. The theoretical efficiency analysis and the influence of network sharing on performance were based on experimental findings conducted on Amazon EC-2 nodes.