Machine learning via DARTS-Optimized MobileViT models for pancreatic Cancer diagnosis with graph-based deep learning

• Published in: BMC medical informatics and decision making Vol. 25; no. 1; pp. 81 - 21
• Main Author: Alaca, Yusuf
• Format: Journal Article
• Language: English
• Published: London BioMed Central 15.02.2025; BioMed Central Ltd; Springer Nature B.V; BMC
• Subjects: Accuracy; Algorithms; Architecture; Cancer; Classification; Corner detection; CT imaging; DARTS; Datasets; Deep Learning; Diagnosis; Graph representations; Graphical representations; Health Informatics; Humans; Image processing; Information Systems and Communication Service; Learning algorithms; Machine Learning; Magnetic resonance imaging; Management of Computing and Information Systems; Medical diagnosis; Medical imaging; Medicine; Medicine & Public Health; Methods; MobileViT; Neural networks; Optimization; Pancreatic Cancer; Pancreatic Neoplasms - diagnosis; Pancreatic Neoplasms - diagnostic imaging; Support vector machines; Tomography, X-Ray Computed; Tumors; XGBoost; Turkey; Deep learning; Medical imaging; XGBoost; MobileViT; Machine learning; DARTS; CT images; Pancreatic Cancer; Graph representations; Cancer diagnosis
• ISSN: 1472-6947; 1472-6947
• DOI: 10.1186/s12911-025-02923-x

The diagnosis of pancreatic cancer presents a significant challenge due to the asymptomatic nature of the disease and the fact that it is frequently detected at an advanced stage. This study presents a novel approach combining graph-based data representation with DARTS-optimised MobileViT models, with the objective of enhancing diagnostic accuracy and reliability. The images of the pancreatic CT were transformed into graph structures using the Harris Corner Detection algorithm, which enables the capture of complex spatial relationships. Subsequently, the graph representations were processed using MobileViT models that had been optimised with Differentiable Architecture Search (DARTS), thereby enabling dynamic architectural adaptation. To further enhance classification accuracy, advanced machine learning algorithms, including K-Nearest Neighbours (KNN), Support Vector Machines (SVM), Random Forest (RF), and XGBoost, were applied. The MobileViTv2_150 and MobileViTv2_200 models demonstrated remarkable performance, with an accuracy of 97.33% and an F1 score of 96.25%, surpassing the capabilities of traditional CNN and Vision Transformer models. This innovative integration of graph-based deep learning and machine learning techniques demonstrates the potential of the proposed method to establish a new standard for early pancreatic cancer diagnosis. Furthermore, the study highlights the scalability of this approach for broader applications in medical imaging, which could lead to improved patient outcomes. Graphical Abstract Highlights • The performance of MobileViT models for pancreatic cancer diagnosis was enhanced using DARTS for dynamic optimization. • By converting pancreatic CT images into graph representations, complex structural features can be captured, thereby enhancing diagnostic accuracy. • MobileViT models have been integrated by combining the strengths of CNN architectures and visual transformer (ViT) models in medical imaging. • Classification algorithms have been utilized to improve the classification performance of optimized MobileViT models, achieving superior accuracy and reliability. • Comparative analyses with traditional CNNs and Vision Transformer models highlight the advantages of the proposed approach. • This approach has the potential to improve patient outcomes and survival rates by enhancing early pancreatic cancer diagnosis. • A scalable and generalizable solution for other cancer diagnoses and medical imaging applications was presented. • The practical applicability of the proposed model in clinical settings was demonstrated, paving the way for its integration into routine diagnostic workflows. • The innovative combination of graph-based data representation, MobileViT models, and DARTS optimization has been introduced, thereby setting a new standard for pancreatic cancer diagnosis.