Since the emergence of convolutional neural networks (CNNs) and, later, vision transformers (ViTs), deep learning architectures have predominantly relied on identical block types with varying hyperparameters. We propose a novel block alternation strategy to leverage the complementary strengths of different architectural designs, assembling structurally distinct components similar to Lego blocks. We introduce LegoNet, a deep learning framework that alternates CNN-based and SwinViT-based blocks to enhance feature learning for medical image segmentation. We investigate three variations of LegoNet and apply this concept to a previously unexplored clinical problem: the segmentation of the internal mammary artery (IMA), aorta, and perivascular adipose tissue (PVAT) from computed tomography angiography (CTA) scans. These PVAT regions have been shown to possess prognostic value in assessing cardiovascular risk and primary clinical outcomes. We evaluate LegoNet on large datasets, achieving superior performance to other leading architectures. Furthermore, we assess the model's generalizability on external testing cohorts, where an expert clinician corrects the model's segmentations, achieving DSC > 0.90 across various external, international, and public cohorts. To further validate the model's clinical reliability, we perform intra- and inter-observer variability analysis, demonstrating strong agreement with human annotations. The proposed methodology has significant implications for diagnostic cardiovascular management and early prognosis, offering a robust, automated solution for vascular and perivascular segmentation and risk assessment in clinical practice, paving the way for personalised medicine.

View on arXiv

@article{sobirov2025_2306.03494,
  title={ Structurally Different Neural Network Blocks for the Segmentation of Atrial and Aortic Perivascular Adipose Tissue in Multi-centre CT Angiography Scans },
  author={ Ikboljon Sobirov and Cheng Xie and Muhammad Siddique and Parijat Patel and Kenneth Chan and Thomas Halborg and Christos P. Kotanidis and Zarqaish Fatima and Henry West and Sheena Thomas and Maria Lyasheva and Donna Alexander and David Adlam and Praveen Rao and Das Indrajeet and Aparna Deshpande and Amrita Bajaj and Jonathan C L Rodrigues and Benjamin J Hudson and Vivek Srivastava and George Krasopoulos and Rana Sayeed and Qiang Zhang and Pete Tomlins and Cheerag Shirodaria and Keith M. Channon and Stefan Neubauer and Charalambos Antoniades and Mohammad Yaqub },
  journal={arXiv preprint arXiv:2306.03494},
  year={ 2025 }
}

Comments on this paper