Developing a Fused AI Model That Integrates 3D Radiomic Data From CBCT With Whole-Slide Images of Oral Biopsies
DOI:
https://doi.org/10.61919/5qhm7309Keywords:
Artificial Intelligence; Cone-Beam Computed Tomography; Oral Squamous Cell Carcinoma; Computational Pathology; Radiomics; Whole-Slide Imaging.Abstract
Background: Oral squamous cell carcinoma grading remains clinically important for prognostic stratification and treatment planning, but conventional diagnostic workflows often evaluate radiological and histopathological data separately. Objective: To develop and evaluate a fused artificial intelligence model integrating three-dimensional CBCT radiomic features with whole-slide image–derived computational pathology features for improving OSCC grading accuracy. Methods: This cross-sectional diagnostic model-development study included 36 patients with histopathologically confirmed OSCC who had complete pre-treatment CBCT imaging and analyzable digitized biopsy slides. Radiomic features were extracted from segmented CBCT tumor volumes, while computational pathology features were derived from whole-slide hematoxylin and eosin images. Radiomics-only, whole-slide image–only, and fused feature-level machine learning models were evaluated using grading accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve. Results: The CBCT radiomics–only model achieved 69.4% accuracy and an AUC of 0.74, while the whole-slide image model achieved 77.8% accuracy and an AUC of 0.81. The fused AI model demonstrated the highest performance, with 86.1% accuracy and an AUC of 0.89. Fusion significantly improved accuracy compared with CBCT radiomics alone (p = 0.012) and whole-slide imaging alone (p = 0.031). Conclusion: Multimodal fusion of CBCT radiomics and whole-slide image analysis improved OSCC grading performance and may support more objective diagnostic decision-making.
References
1. Broggi G, Maniaci A, Lentini M, Palicelli A, Zanelli M, Zizzo M, et al. Artificial intelligence in head and neck cancer diagnosis: a comprehensive review with emphasis on radiomics, histopathological, and molecular applications. 2024;16(21):3623.
2. Di Vincenzo M. Review on multi-modal AI models to integrate imaging and omics data. 2024.
3. Song B, Leroy A, Yang K, Dam T, Wang X, Maurya H, et al. Deep learning informed multimodal fusion of radiology and pathology to predict outcomes in HPV-associated oropharyngeal squamous cell carcinoma. 2025;114.
4. Chaudhary N, Muddemanavar P, Singh DK, Rai A, Mishra D, SV S, et al. Dual-stage AI system for pathologist-free tumor detection and subtyping in oral squamous cell carcinoma. 2025.
5. Mahmood H. Artificial intelligence based assessment of oral precancer to aid early detection of oral cancer. Sheffield: University of Sheffield; 2023.
6. İlhan B, Güneri P. Artificial intelligence in oral medicine: from data to diagnosis. 2025;27.
7. Tiwari A, Mishra S, Kuo TR. Current AI technologies in cancer diagnostics and treatment. 2025;24(1):159.
8. Förster F, Attia PM, Severson KA, Witmer JD. Towards a safe and efficient clinical implementation of machine learning in radiation oncology by exploring model interpretability, explainability and data-model dependency.
9. Redlich JP, Feuerhake F, Weis J, Schaadt NS, Teuber-Hanselmann S, Buck C, et al. Applications of artificial intelligence in the analysis of histopathology images of gliomas: a review. 2024;2(1):16.
10. Shi R, Sun J, Zhou Z, Su Q, Shu Y. Deep multimodal fusion of patho-radiomic and clinical data for enhanced survival prediction for colorectal cancer patients. 2025.
11. Cifci D, Veldhuizen GP, Foersch S, Kather JN. AI in computational pathology of cancer: improving diagnostic workflows and clinical outcomes? Annu Rev Cancer Biol. 2023;7(1):57-71.
12. Zhang W, Xu W, Chen T, Sakal C, Li X. Integrating images and genomics for multi-modal cancer survival analysis via mixture of experts. Inf Fusion. 2025;103521.
13. Wang J, Zeng Z, Li Z, Liu G, Zhang S, Luo C, et al. The clinical application of artificial intelligence in cancer precision treatment. 2025;23(1):120.
14. Kour T, Raina JK, Gondhi NK, Sharma A, Banerjee S, Kumar P, et al. Transformative impact of deep learning and machine learning in oncology: a comprehensive review of AI-based approaches for early detection, diagnosis and therapeutics across different cancer types. 2025:1-26.
15. Islam S. From radiology to pathology: deep learning models in diagnostic medicine. J Adv Res. 2025;1(03):1-21.
16. Vojgani Y, Sadeghinia MJ. Application of artificial intelligence in cancer management with a personalized medicine approach. ATJ. 2024;6(20):24-33.
17. Liao J, Li X, Gan Y, Han S, Rong P, Wang W, et al. Artificial intelligence assists precision medicine in cancer treatment. 2023;12:998222.
18. Saednia KS. Machine learning and digital histopathology analysis for tissue characterization and treatment response prediction in breast cancer. 2023.
19. Lococo F, Ghaly G, Chiappetta M, Flamini S, Evangelista J, Bria E, et al. Implementation of artificial intelligence in personalized prognostic assessment of lung cancer: a narrative review. 2024;16(10):1832.
20. Cheng Y, Lama N, Chen M, Amidi E, Ramzanpour M, Rahman MA, et al. Synergistic H&E and IHC image analysis by AI predicts cancer biomarkers and survival outcomes in colorectal and breast cancer. 2025;5(1):328.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Mariam Imdad, Uzma Zareef, Laiba Sohail, Amna Shakeel, Muhammad Hamza, Ahmad Hassan, Laiba Sheikh (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
© The Authors. This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).
.png)
