{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,20]],"date-time":"2026-07-20T10:35:42Z","timestamp":1784543742261,"version":"3.55.0"},"reference-count":59,"publisher":"MDPI AG","issue":"9","license":[{"start":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T00:00:00Z","timestamp":1777507200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Diagnostics"],"abstract":"<jats:p>Background\/Objectives: Automated brain tumor classification from MRI is particularly challenging when restricted to single post-contrast axial T1-weighted slices without volumetric or clinical context. Methods: We present a four-class (glioma, meningioma, pituitary tumor, no tumor) slice-level classification framework that combines a fine-tuned Swin-Tiny Transformer with inverse-frequency class-weighted learning and a prototype SMT-based symbolic auditing layer for post hoc logical consistency checks. All architectures were trained and evaluated under identical preprocessing, augmentation, optimization, and evaluation protocols. Results: On an internal clinical dataset from Band\u0131rma Onyedi Eyl\u00fcl University Hospital (n = 8040 slices), Swin-Tiny achieved 97.42% slice-level accuracy (macro-F1 97.42%, macro-AUC 0.994), exceeding matched convolutional baselines by approximately eight percentage points. Five-fold stratified cross-validation confirmed stability (mean accuracy 97.40% \u00b1 0.28%). Zero-shot evaluation on the independent BRISC-2025 dataset (n = 6000 slices) yielded 94.82% accuracy and macro-AUC 0.97, indicating maintained performance under acquisition-related distribution shift. Per-class metrics were consistently high across tumor types, with residual errors dominated by glioma\u2013meningioma confusion, reflecting known radiologic overlap on single contrast-enhanced T1 slices. The symbolic auditing layer flagged 1.2\u20132.9% of predictions as constraint-violating; most such cases were borderline but correctly classified, suggesting sensitivity of heuristic thresholds rather than systematic model failure. Conclusions: These findings support the value of hierarchical shifted-window attention for integrating local texture and broader spatial context in slice-level MRI classification. While patient-wise, multimodal, and prospective validation remain necessary for clinical deployment, this study provides a controlled empirical benchmark and a prototype mechanism for post hoc logical auditing in neuro-oncologic imaging.<\/jats:p>","DOI":"10.3390\/diagnostics16091361","type":"journal-article","created":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T12:20:25Z","timestamp":1777551625000},"page":"1361","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["A Verifiable Framework for Brain Tumor Classification: Combining Vision Transformers, Class-Weighted Learning, and SMT-Based Formal Decision Traces"],"prefix":"10.3390","volume":"16","author":[{"given":"Mehmet Akif","family":"\u00c7if\u00e7i","sequence":"first","affiliation":[{"name":"Institute of Research and Development, Duy Tan University, Da Nang 551111, Vietnam"},{"name":"Department of Computer Engineering, Band\u0131rma Onyedi Eyl\u00fcl University, 10200 Bal\u0131kesir, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8114-3270","authenticated-orcid":false,"given":"Kadir","family":"Karata\u015f","sequence":"additional","affiliation":[{"name":"Department of Computer Engineering, Band\u0131rma Onyedi Eyl\u00fcl University, 10200 Bal\u0131kesir, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8142-0466","authenticated-orcid":false,"given":"Fazli","family":"Y\u0131ld\u0131r\u0131m","sequence":"additional","affiliation":[{"name":"Faculty of Economics, Administrative and Social Sciences, Department of Management Information Systems, Istanbul Topkap\u0131 University, 34087 Istanbul, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ali","family":"Do\u011fan","sequence":"additional","affiliation":[{"name":"Department of Neurology, Faculty of Medicine, Band\u0131rma Onyedi Eyl\u00fcl University, 10200 Bal\u0131kesir, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,4,30]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"188458","DOI":"10.1016\/j.bbcan.2020.188458","article-title":"Pre-clinical tumor models of primary brain tumors: Challenges and opportunities","volume":"1875","author":"Akter","year":"2021","journal-title":"Biochim. Biophys. Acta (BBA)-Rev. Cancer"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"e20","DOI":"10.1017\/erm.2021.20","article-title":"Preclinical models of glioblastoma: Limitations of current models and the promise of new developments","volume":"23","author":"Liu","year":"2021","journal-title":"Expert Rev. Mol. Med."},{"key":"ref_3","first-page":"1","article-title":"Advancements in the Management of Meningioma: A Comprehensive Review of Current Treatment Modalities","volume":"7","author":"Dimitrios","year":"2024","journal-title":"Clin. Oncol."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Kumar, A.A., and Kesavadas, C. (2025). Potential of MRI in Clinical Medicine. Multimodal Biomedical Imaging Techniques, Springer Nature.","DOI":"10.1007\/978-981-96-1124-9_12"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"80","DOI":"10.1038\/s41698-024-00575-0","article-title":"Artificial intelligence in neuro-oncology: Advances and challenges in brain tumor diagnosis, prognosis, and precision treatment","volume":"8","author":"Khalighi","year":"2024","journal-title":"npj Precis. Oncol."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"100942","DOI":"10.1016\/j.modpat.2025.100942","article-title":"Diagnostic discordance and error in breast pathology: Causes, classifications, and medicolegal implications","volume":"39","author":"Rakha","year":"2025","journal-title":"Mod. Pathol."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Liew, X.Y., Hameed, N., and Clos, J. (2021). A review of computer-aided expert systems for breast cancer diagnosis. Cancers, 13.","DOI":"10.3390\/cancers13112764"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"113623","DOI":"10.1109\/ACCESS.2023.3313977","article-title":"Recent advancements and future prospects in active deep learning for medical image segmentation and classification","volume":"11","author":"Mahmood","year":"2023","journal-title":"IEEE Access"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"66726","DOI":"10.1109\/ACCESS.2024.3399114","article-title":"Signal to image conversion and convolutional neural networks for physiological signal processing: A review","volume":"12","author":"Vidyasagar","year":"2024","journal-title":"IEEE Access"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"3145","DOI":"10.1007\/s10586-022-03951-2","article-title":"A comprehensive review of the deep learning-based tumor analysis approaches in histopathological images: Segmentation, classification and multi-learning tasks","volume":"26","author":"Ali","year":"2023","journal-title":"Clust. Comput."},{"key":"ref_11","first-page":"857","article-title":"Revolutionizing tumor detection and classification in multimodality imaging based on deep learning approaches: Methods, applications and limitations","volume":"32","author":"Hussain","year":"2024","journal-title":"J. X-Ray Sci. Technol."},{"key":"ref_12","unstructured":"Zhu, Z., Rehman, A., Cao, X., Liao, C., Lee, Y.J., Ohliger, M., Xue, H., and Yang, Y. (2024). Advancing low-field MRI with a universal denoising imaging transformer: Towards fast and high-quality imaging. arXiv."},{"key":"ref_13","first-page":"331","article-title":"Deep Multi-Scale and Attention-Based Architectures for Semantic Segmentation in Biomedical Imaging","volume":"85","author":"Harouni","year":"2025","journal-title":"Comput. Mater. Contin."},{"key":"ref_14","unstructured":"Pereira, G.A., and Hussain, M. (2024). A review of transformer-based models for computer vision tasks: Capturing global context and spatial relationships. arXiv."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"191824","DOI":"10.1109\/ACCESS.2025.3618215","article-title":"A recent survey of vision transformers for medical image segmentation","volume":"13","author":"Khan","year":"2025","journal-title":"IEEE Access"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"84","DOI":"10.1007\/s10916-024-02105-8","article-title":"Comparison of vision transformers and convolutional neural networks in medical image analysis: A systematic review","volume":"48","author":"Takahashi","year":"2024","journal-title":"J. Med. Syst."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"20655","DOI":"10.1038\/s41598-025-07561-x","article-title":"Transformer attention fusion for fine grained medical image classification","volume":"15","author":"Badar","year":"2025","journal-title":"Sci. Rep."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"22964","DOI":"10.1038\/s41598-024-74753-2","article-title":"A fine-grained image classification algorithm based on self-supervised learning and multi-feature fusion of blood cells","volume":"14","author":"Jia","year":"2024","journal-title":"Sci. Rep."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"3174","DOI":"10.1007\/s10278-024-01140-8","article-title":"Enhancing skin cancer diagnosis using swin transformer with hybrid shifted window-based multi-head self-attention and SwiGLU-based MLP","volume":"37","author":"Pacal","year":"2024","journal-title":"J. Imaging Inform. Med."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Mienye, I.D., and Viriri, S. (2025). Graph Neural Networks in Medical Imaging: Methods, Applications and Future Directions. Information, 16.","DOI":"10.3390\/info16121051"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"3579","DOI":"10.1007\/s13042-024-02110-w","article-title":"A novel Swin transformer approach utilizing residual multi-layer perceptron for diagnosing brain tumors in MRI images","volume":"15","author":"Pacal","year":"2024","journal-title":"Int. J. Mach. Learn. Cybern."},{"key":"ref_22","unstructured":"Model, A.I.R. (2022). Imaging and 18F-FDG PET Imaging. Structural, Metabolic and Physiologic MR Imaging to Study Glioblastomas, Frontiers Media SA."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"118","DOI":"10.1109\/RBME.2023.3269776","article-title":"Radiomics and deep learning in nasopharyngeal carcinoma: A review","volume":"17","author":"Wang","year":"2023","journal-title":"IEEE Rev. Biomed. Eng."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Wang, H., Sun, Y., Cao, Y., Ouyang, G., Wang, X., Wu, S., and Tian, M. (2022). Classification for diabetic retinopathy by using staged convolutional neural network. Proceedings of the 2022 Asia Conference on Algorithms, Computing and Machine Learning (CACML), Hangzhou, China, 25\u201327 March 2022, IEEE.","DOI":"10.1109\/CACML55074.2022.00045"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"e1557","DOI":"10.1002\/ecm.1557","article-title":"Cross validation for model selection: A review with examples from ecology","volume":"93","author":"Yates","year":"2023","journal-title":"Ecol. Monogr."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Yilmaz Ba\u015fer, H., Evran, T., and Cifci, M.A. (2025). Machine learning-augmented triage for sepsis: Real-time ICU mortality prediction using SHAP-explained meta-ensemble models. Biomedicines, 13.","DOI":"10.3390\/biomedicines13061449"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"1607","DOI":"10.1007\/s12530-024-09581-w","article-title":"Self-supervised learning for medical image analysis: A comprehensive review","volume":"15","author":"Rani","year":"2024","journal-title":"Evol. Syst."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"C\u00e8, M., Chiriac, M.D., Cozzi, A., Macr\u00ec, L., Rabaiotti, F.L., Irmici, G., Fazzini, D., Carrafiello, G., and Cellina, M. (2024). Decoding Radiomics: A Step-by-Step Guide to Machine Learning Workflow in Hand-Crafted and Deep Learning Radiomics Studies. Diagnostics, 14.","DOI":"10.3390\/diagnostics14222473"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"20","DOI":"10.1016\/j.ymeth.2020.05.022","article-title":"Radiomics for precision medicine: Current challenges, future prospects, and the proposal of a new framework","volume":"188","author":"Ibrahim","year":"2021","journal-title":"Methods"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"1525","DOI":"10.1007\/s11831-024-10188-2","article-title":"A critical review on segmentation of glioma brain tumor and prediction of overall survival","volume":"32","author":"Rasool","year":"2025","journal-title":"Arch. Comput. Methods Eng."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"1199","DOI":"10.1109\/TIP.2024.3359815","article-title":"Multimodal brain tumor segmentation boosted by monomodal normal brain images","volume":"33","author":"Liu","year":"2024","journal-title":"IEEE Trans. Image Process."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Owusu-Adjei, M., Ben Hayfron-Acquah, J., Frimpong, T., and Abdul-Salaam, G. (2023). Imbalanced class distribution and performance evaluation metrics: A systematic review of prediction accuracy for determining model performance in healthcare systems. PLoS Digit. Health, 2.","DOI":"10.1371\/journal.pdig.0000290"},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Berroukham, A., Housni, K., and Lahraichi, M. (2023). Vision transformers: A review of architecture, applications, and future directions. Proceedings of the 2023 7th IEEE Congress on Information Science and Technology (CiSt), Agadir-Essaouira, Morocco, 16\u201322 December 2023, IEEE.","DOI":"10.1109\/CiSt56084.2023.10410015"},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Liu, Z., Hu, H., Lin, Y., Yao, Z., Xie, Z., Wei, Y., Ning, J., Cao, Y., Zhang, Z., and Guo, B. (2022, January 18\u201324). Swin transformer v2: Scaling up capacity and resolution. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA.","DOI":"10.1109\/CVPR52688.2022.01170"},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Saraei, M., and Liu, S. (2023). Attention-based deep learning approaches in brain tumor image analysis: A mini review. Front. Health Inform., 12.","DOI":"10.30699\/fhi.v12i0.493"},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Aliferis, C., and Simon, G. (2024). Lessons learned from historical failures, limitations and successes of AI\/ML in healthcare and the health sciences. Enduring problems, and the role of best practices. Artificial Intelligence and Machine Learning in Health Care and Medical Sciences: Best Practices and Pitfalls, Springer.","DOI":"10.1007\/978-3-031-39355-6_12"},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"e210064","DOI":"10.1148\/ryai.210064","article-title":"External validation of deep learning algorithms for radiologic diagnosis: A systematic review","volume":"4","author":"Yu","year":"2022","journal-title":"Radiol. Artif. Intell."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"14657","DOI":"10.1038\/s41598-024-64941-5","article-title":"A retrospective study of deep learning generalization across two centers and multiple models of X-ray devices using COVID-19 chest-X rays","volume":"14","author":"Fraguela","year":"2024","journal-title":"Sci. Rep."},{"key":"ref_39","unstructured":"Rahman, M.M., Tutul, A.A., Nath, A., Laishram, L., Jung, S.K., and Hammond, T. (2024). Mamba in vision: A comprehensive survey of techniques and applications. arXiv."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"95496","DOI":"10.1109\/ACCESS.2025.3571735","article-title":"Transformers for vision: A survey on innovative methods for computer vision","volume":"13","author":"Hassija","year":"2025","journal-title":"IEEE Access"},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"181071","DOI":"10.1109\/ACCESS.2024.3509372","article-title":"Transformers: A security perspective","volume":"12","author":"Latibari","year":"2024","journal-title":"IEEE Access"},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Kvande, M.A., Jacobsen, S.L., Goodwin, M., and Gupta, R. (2024). Using AI to Empower Norwegian Agriculture: Attention-Based Multiple-Instance Learning Implementation. Agronomy, 14.","DOI":"10.3390\/agronomy14061089"},{"key":"ref_43","unstructured":"Duarte, M., and de la Paz, M. (2025). A Roadmap to Predicting Associations Between Ectomycorrhizal Fungi and Host Plants. [Doctoral Dissertation, University of British Columbia]."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"669","DOI":"10.1016\/S1064-9689(21)00359-7","article-title":"Benign and malignant masses: MR imaging differentiation","volume":"3","author":"Weatherall","year":"1995","journal-title":"Magn. Reson. Imaging Clin. N. Am."},{"key":"ref_45","doi-asserted-by":"crossref","unstructured":"Mienye, I.D., Swart, T.G., Obaido, G., Jordan, M., and Ilono, P. (2025). Deep convolutional neural networks in medical image analysis: A review. Information, 16.","DOI":"10.3390\/info16030195"},{"key":"ref_46","doi-asserted-by":"crossref","first-page":"139367","DOI":"10.1109\/ACCESS.2020.3004766","article-title":"A comparison of transfer learning performance versus health experts in disease diagnosis from medical imaging","volume":"8","author":"Malik","year":"2020","journal-title":"IEEE Access"},{"key":"ref_47","doi-asserted-by":"crossref","first-page":"191","DOI":"10.1186\/s13195-021-00924-2","article-title":"Improving 3D convolutional neural network comprehensibility via interactive visualization of relevance maps: Evaluation in Alzheimer\u2019s disease","volume":"13","author":"Dyrba","year":"2021","journal-title":"Alzheimer\u2019s Res. Ther."},{"key":"ref_48","doi-asserted-by":"crossref","unstructured":"Matsoukas, C., Haslum, J.F., Sorkhei, M., S\u00f6derberg, M., and Smith, K. (2022, January 18\u201324). What makes transfer learning work for medical images: Feature reuse & other factors. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA.","DOI":"10.1109\/CVPR52688.2022.00901"},{"key":"ref_49","first-page":"1085","article-title":"An extensive review of state-of-the-art transfer learning techniques used in medical imaging: Open issues and challenges","volume":"31","author":"Mukhlif","year":"2022","journal-title":"J. Intell. Syst."},{"key":"ref_50","doi-asserted-by":"crossref","first-page":"3642","DOI":"10.21105\/joss.03642","article-title":"Feature-engine: A Python package for feature engineering for machine learning","volume":"6","author":"Galli","year":"2021","journal-title":"J. Open Source Softw."},{"key":"ref_51","doi-asserted-by":"crossref","first-page":"42","DOI":"10.1186\/s40537-018-0151-6","article-title":"A survey on addressing high-class imbalance in big data","volume":"5","author":"Leevy","year":"2018","journal-title":"J. Big Data"},{"key":"ref_52","unstructured":"Azar, S.G., Tronchin, L., Simk\u00f3, A., Nyholm, T., and L\u00f6fstedt, T. (2025). From Promise to Practice: A Study of Common Pitfalls Behind the Generalization Gap in Machine Learning. Trans. Mach. Learn. Res., Available online: https:\/\/openreview.net\/forum?id=DqWvxSQ1TK."},{"key":"ref_53","first-page":"2490","article-title":"A generic improvement to deep residual networks based on gradient flow","volume":"31","author":"Santhanam","year":"2019","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"ref_54","doi-asserted-by":"crossref","first-page":"9393","DOI":"10.1109\/TPAMI.2023.3241201","article-title":"Attention spiking neural networks","volume":"45","author":"Yao","year":"2023","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_55","doi-asserted-by":"crossref","unstructured":"Cifci, M.A., Hussain, S., and Canatalay, P.J. (2023). Hybrid deep learning approach for accurate tumor detection in medical imaging data. Diagnostics, 13.","DOI":"10.3390\/diagnostics13061025"},{"key":"ref_56","doi-asserted-by":"crossref","first-page":"64717","DOI":"10.52202\/079017-2065","article-title":"Faster neighborhood attention: Reducing the o (n^2) cost of self attention at the threadblock level","volume":"37","author":"Hassani","year":"2024","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_57","doi-asserted-by":"crossref","unstructured":"Cifci, M.A., \u00d6ney, B., Yildirim, F., Yilmaz Ba\u015fer, H., and Zontul, M. (2025). Interpretable Adaptive Graph Fusion Network for Mortality and Complication Prediction in ICUs. Diagnostics, 15.","DOI":"10.3390\/diagnostics15222825"},{"key":"ref_58","unstructured":"Bonal, E. (2025). Framework for a Testable Metaphysical Science: Type-Theoretic System and Computational Experimentation Using Z3 SMT Solver. PhilArchive."},{"key":"ref_59","doi-asserted-by":"crossref","first-page":"273","DOI":"10.1007\/s11102-019-01026-x","article-title":"Machine learning applications in imaging analysis for patients with pituitary tumors: A review of the current literature and future directions","volume":"23","author":"Saha","year":"2020","journal-title":"Pituitary"}],"container-title":["Diagnostics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/9\/1361\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T12:25:49Z","timestamp":1777551949000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/9\/1361"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,30]]},"references-count":59,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2026,5]]}},"alternative-id":["diagnostics16091361"],"URL":"https:\/\/doi.org\/10.3390\/diagnostics16091361","relation":{},"ISSN":["2075-4418"],"issn-type":[{"value":"2075-4418","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,30]]}}}