{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,4]],"date-time":"2026-08-04T13:43:45Z","timestamp":1785851025665,"version":"3.56.0"},"reference-count":38,"publisher":"MDPI AG","issue":"15","license":[{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"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: Extracting radiomic features from volumetric imaging data for lung cancer classification is limited by high dimensionality and the black-box nature of deep learning models. Traditional methods may overlook dependencies between slices, leading to the loss of sequential spatial information. Even when classification is correct, the model\u2019s explanatory power may remain insufficient. Objective: This study proposes SpecAtt-Net, which processes spectral representations of slice-level radiomic data from computed tomography (CT) scans of lung cancer patients. A slice-based explainability approach is also presented to identify the slice groups used in model decision-making. Methods: Slice-level radiomic data were preprocessed, and slice-position\u2013frequency representations were generated using the Short-Time Fourier Transform (STFT). This enables the model to capture spectral textures and spatial transitions from 720 \u00d7 288 feature maps. SpecAtt-Net employs a dual-attention mechanism to focus on distinctive radiomic features. To address the lack of slice-level labels, Slice-Wise Activation Mapping (SWAM), a post hoc interpretability technique, was developed. SWAM converts two-dimensional (2D) activation maps into one-dimensional (1D) importance vectors and provides a weakly supervised indication of the slice groups most relevant to the model\u2019s decision. Results: SpecAtt-Net achieved competitive performance in accuracy, precision, recall, and F1-score compared with the evaluated reference architectures. SWAM highlighted slice-sequence regions contributing strongly to model predictions, providing preliminary evidence for interpretability. Conclusions: SpecAtt-Net provides a compact and interpretable approach for lung cancer subtype classification using attention-based feature modeling. SWAM offers an exploratory visualization of decision-relevant sequence regions and may support future radiological interpretation and validation studies.<\/jats:p>","DOI":"10.3390\/diagnostics16152436","type":"journal-article","created":{"date-parts":[[2026,8,3]],"date-time":"2026-08-03T06:17:40Z","timestamp":1785737860000},"page":"2436","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["SpecAtt-Net: Time-Frequency-Based Image Representations of Slice-Level Radiomic Features for Explainable Lung Cancer Classification via Slice-Wise Activation Mapping"],"prefix":"10.3390","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0733-3652","authenticated-orcid":false,"given":"Merve","family":"Ceyhan","sequence":"first","affiliation":[{"name":"Department of Computer Engineering, Faculty of Engineering and Architecture, Eski\u015fehir Osmangazi University, Odunpazar\u0131, Eski\u015fehir 26040, Turkey"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"U\u011fur","family":"G\u00fcrel","sequence":"additional","affiliation":[{"name":"Department of Computer Engineering, Faculty of Engineering and Architecture, Eski\u015fehir Osmangazi University, Odunpazar\u0131, Eski\u015fehir 26040, Turkey"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,8,1]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"53","DOI":"10.5306\/wjco.v11.i2.53","article-title":"Objective response rate assessment in oncology: Current situation and future expectations","volume":"11","author":"Aykan","year":"2020","journal-title":"World J. Clin. Oncol."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Pi\u00f1ero, F., Dirchwolf, M., and Pess\u00f4a, M.G. (2020). Biomarkers in Hepatocellular Carcinoma: Diagnosis, Prognosis and Treatment Response Assessment. Cells, 9.","DOI":"10.3390\/cells9061370"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"100426","DOI":"10.1016\/j.ejro.2022.100426","article-title":"Imaging response assessment for oncology: An algorithmic approach","volume":"9","author":"Ruchalski","year":"2022","journal-title":"Eur. J. Radiol."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"20","DOI":"10.1186\/s13058-022-01516-0","article-title":"Radiomics predicts the prognosis of patients with locally advanced breast cancer by reflecting the heterogeneity of tumor cells and the tumor microenvironment","volume":"24","author":"Wang","year":"2022","journal-title":"Breast Cancer Res."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"72","DOI":"10.1038\/s41523-024-00678-8","article-title":"Integrative radiomics clustering analysis to decipher breast cancer heterogeneity and prognostic indicators through multiparametric MRI","volume":"10","author":"He","year":"2024","journal-title":"npj Breast Cancer"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"1717","DOI":"10.1148\/rg.2021210037","article-title":"Radiomics in oncology: A practical guide","volume":"41","author":"Shur","year":"2021","journal-title":"Radiographics"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"2568","DOI":"10.1109\/TSP.2009.2028095","article-title":"Short-time fractional fourier transform and its applications","volume":"58","author":"Tao","year":"2010","journal-title":"IEEE Trans. Signal Process."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Yao, S., Piao, A., Jiang, W., Zhao, Y., Shao, H., Liu, S., Liu, D., Li, J., Wang, T., and Hu, S. (2019, January 13\u201317). STFNets: Learning sensing signals from the time-frequency perspective with short-time fourier neural networks. Proceedings of the Web Conference\u2014Proceedings of the World Wide Web Conference, San Francisco, CA, USA.","DOI":"10.1145\/3308558.3313426"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"338821","DOI":"10.1016\/j.aca.2021.338821","article-title":"Accurate diagnosis of lung tissues for 2D Raman spectrogram by deep learning based on short-time Fourier transform","volume":"1179","author":"Qi","year":"2021","journal-title":"Anal. Chim. Acta"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Preuss, K., Thach, N., Liang, X., Baine, M., Chen, J., Zhang, C., Du, H., Yu, H., Lin, C., and Hollingsworth, M.A. (2022). Using Quantitative Imaging for Personalized Medicine in Pancreatic Cancer: A Review of Radiomics and Deep Learning Applications. Cancers, 14.","DOI":"10.3390\/cancers14071654"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Babu, T.R.G., Saravanakumar, U., and Pattanaik, B. (2024). Computational Imaging and Analytics in Biomedical Engineering: Algorithms and Applications, Apple Academic Press.","DOI":"10.1201\/9781032669687"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Khoria, S., and Singh, R. (2025). Predicting lung cancer using short-time fourier transform and bidirectional long short-term memory networks. Proc. Indian Natl. Sci. Acad., 1\u201316.","DOI":"10.1007\/s43538-025-00455-6"},{"key":"ref_13","unstructured":"(2026, June 01). LUNG-PET-CT-DX\u2014The Cancer Imaging Archive (TCIA). Available online: https:\/\/www.cancerimagingarchive.net\/collection\/lung-pet-ct-dx\/."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"328","DOI":"10.1148\/radiol.2020191145","article-title":"The Image Biomarker Standardization Initiative: Standardized Quantitative Radiomics for High-Throughput Image-based Phenotyping","volume":"295","author":"Zwanenburg","year":"2020","journal-title":"Radiology"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1016\/0165-1684(88)90040-0","article-title":"A short-time Fourier transform","volume":"14","author":"Owens","year":"1988","journal-title":"Signal Process."},{"key":"ref_16","unstructured":"(2026, June 01). SciPy. Available online: https:\/\/scipy.org\/."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"92871","DOI":"10.1109\/ACCESS.2019.2928017","article-title":"ECG Arrhythmia Classification Using STFT-Based Spectrogram and Convolutional Neural Network","volume":"7","author":"Huang","year":"2019","journal-title":"IEEE Access"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"1234","DOI":"10.1016\/j.mri.2012.06.010","article-title":"Radiomics: The process and the challenges","volume":"30","author":"Kumar","year":"2012","journal-title":"Magn. Reson. Imaging"},{"key":"ref_19","unstructured":"(2026, June 01). Deep Learning. Available online: https:\/\/www.deeplearningbook.org\/."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"886","DOI":"10.1016\/j.tranon.2017.08.007","article-title":"2D and 3D CT Radiomics Features Prognostic Performance Comparison in Non-Small Cell Lung Cancer","volume":"10","author":"Shen","year":"2017","journal-title":"Transl. Oncol."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Wan, Q., Zhou, J., Xia, X., Hu, J., Wang, P., Peng, Y., Zhang, T., Sun, J., Song, Y., and Yang, G. (2021). Diagnostic Performance of 2D and 3D T2WI-Based Radiomics Features With Machine Learning Algorithms to Distinguish Solid Solitary Pulmonary Lesion. Front. Oncol., 11.","DOI":"10.3389\/fonc.2021.683587"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Sueur, J. (2018). Spectrographic Visualization. Sound Analysis and Synthesis with R, Springer.","DOI":"10.1007\/978-3-319-77647-7_11"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"2137","DOI":"10.1109\/LSP.2015.2463093","article-title":"Time-Frequency Filtering Based on Spectrogram Zeros","volume":"22","author":"Flandrin","year":"2015","journal-title":"IEEE Signal Process"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"273","DOI":"10.1023\/A:1022627411411","article-title":"Support-vector networks","volume":"20","author":"Cortes","year":"1995","journal-title":"Mach. Learn."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Mach. Learn."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Chen, T., and Guestrin, C. (2016, January 13\u201317). XGBoost: A scalable tree boosting system. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA.","DOI":"10.1145\/2939672.2939785"},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Daydulo, Y.D., Thamineni, B.L., and Dawud, A.A. (2023). Cardiac arrhythmia detection using deep learning approach and time frequency representation of ECG signals. BMC Med. Inform. Decis. Mak., 23.","DOI":"10.1186\/s12911-023-02326-w"},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Gao, D., Tang, X., Wan, M., Huang, G., and Zhang, Y. (2023). EEG driving fatigue detection based on log-Mel spectrogram and convolutional recurrent neural networks. Front. Neurosci., 17.","DOI":"10.3389\/fnins.2023.1136609"},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"M S, K., Rajaguru, H., and Nair, A.R. (2023). Evaluation and Exploration of Machine Learning and Convolutional Neural Network Classifiers in Detection of Lung Cancer from Microarray Gene\u2014A Paradigm Shift. Bioengineering, 10.","DOI":"10.20944\/preprints202307.0519.v1"},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Eleyan, A., Bayram, F., and Eleyan, G. (2024). Spectrogram-Based Arrhythmia Classification Using Three-Channel Deep Learning Model with Feature Fusion. Appl. Sci., 14.","DOI":"10.3390\/app14219936"},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"e31827","DOI":"10.1016\/j.heliyon.2024.e31827","article-title":"A real-time epilepsy seizure detection approach based on EEG using short-time Fourier transform and Google-Net convolutional neural network","volume":"10","author":"Shen","year":"2024","journal-title":"Heliyon"},{"key":"ref_32","first-page":"140","article-title":"Classification of EEG Spectrogram Images with Deep Learning Models for Alcoholism Detection","volume":"2","author":"Yildirim","year":"2025","journal-title":"Curr. Trends Comput."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"11667","DOI":"10.1007\/s00500-024-09866-x","article-title":"Deep learning and feature fusion-based lung sound recognition model to diagnoses the respiratory diseases","volume":"28","author":"Shehab","year":"2024","journal-title":"Soft Comput."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Zheng, B., Luo, W., Zhang, M., and Jin, H. (2025). Arrhythmia classification based on multi-input convolutional neural network with attention mechanism. PLoS ONE, 20.","DOI":"10.1371\/journal.pone.0326079"},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"e70155","DOI":"10.1002\/ett.70155","article-title":"Deep Learning Based Classification and Combined Transform Based Feature Extraction Approach for Mental Stress Prediction of Human Beings Using EEG","volume":"36","author":"Agarwal","year":"2025","journal-title":"Trans. Emerg. Telecommun. Technol."},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Karim, A., Ryu, S., and Jeong, I.C. (2025). Deep Learning-Enhanced Spectrogram Analysis for Anatomical Region Classification in Biomedical Signals. Appl. Sci., 15.","DOI":"10.3390\/app15105313"},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"e15734056388107","DOI":"10.2174\/0115734056388107250710120917","article-title":"Evaluation of Deep Learning Methods for Pulmonary Disease Classification","volume":"21","author":"Singh","year":"2025","journal-title":"Curr. Med. Imaging"},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"45233","DOI":"10.1038\/s41598-025-28832-7","article-title":"Deep Learning-Driven Early Diagnosis of Respiratory Diseases using CNN-RNN Fusion on Lung Sound Data","volume":"15","author":"Bikku","year":"2025","journal-title":"Sci. Rep."}],"container-title":["Diagnostics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/15\/2436\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,4]],"date-time":"2026-08-04T12:45:40Z","timestamp":1785847540000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/15\/2436"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,1]]},"references-count":38,"journal-issue":{"issue":"15","published-online":{"date-parts":[[2026,8]]}},"alternative-id":["diagnostics16152436"],"URL":"https:\/\/doi.org\/10.3390\/diagnostics16152436","relation":{},"ISSN":["2075-4418"],"issn-type":[{"value":"2075-4418","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,1]]}}}