{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T06:04:45Z","timestamp":1790057085389,"version":"4.0.1"},"reference-count":43,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T00:00:00Z","timestamp":1790035200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Vellore Institute of Technology, Vellore"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>Accurate differentiation of benign and malignant pulmonary nodules on computed tomography (CT) is important for early lung-cancer assessment. However, classification remains challenging because benign and malignant nodules may exhibit subtle differences in texture, margin irregularity, boundary sharpness, and internal morphology that may not be fully represented by conventional convolutional feature extraction.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>A frequency response Bilinear Lag-Lead (BT-LL) EfficientNetB3 framework was developed for benign-malignant pulmonary nodule classification. The proposed approach integrates Lag-Lead frequency compensation into the pretrained feature-extraction process to refine intermediate representations. The Lag component performs structural smoothing to preserve low-frequency information, whereas the Lead component enhances high-frequency boundary and texture characteristics. These corresponding representations are integrated with the original feature maps through a dedicated fusion operation to obtain frequency-refined features for classification. The compensation filters are fixed, thereby avoiding additional parameter learning within the filtering operation. Patient-level data partitioning was used to ensure that nodules from the same patient were not distributed across different experimental subsets, reducing the risk of information leakage.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The proposed model achieved an accuracy, sensitivity, and specificity of 0.90, an F1-score of 0.81, and an AUROC of 0.95 on the held-out patient-level test set. Performance was additionally assessed using patient-level cross-validation, cross imbalance analysis, an IQ-OTH\/NCCD image-level evaluation and an independent BUS-BRA patient-level evaluation. The experimental analysis further included statistical significance testing using McNemar\u2019s test for pairwise model comparisons, followed by the Holm correction to control for multiple comparisons, while Cochran\u2019s Q test was employed to assess performance differences across multiple model configurations. The combined configuration achieved the best sensitivity and balanced accuracy relative to the branch-wise variants. The Bilinear Lag-Lead (BT-LL) framework thus provides an explicit, frequency response mechanism for integrating contributions of low- and high-frequency information in pulmonary nodule representations.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/frai.2026.1934912","type":"journal-article","created":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T05:39:34Z","timestamp":1790055574000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Frequency response compensation to optimize pretrained deep learning models in lung nodule malignancy classification"],"prefix":"10.3389","volume":"9","author":[{"given":"V.","family":"Sangeetha","sequence":"first","affiliation":[{"name":"School of Computer Science Engineering and Information Systems, Vellore Institute of Technology","place":["Vellore, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"S.","family":"Kalaivani","sequence":"additional","affiliation":[{"name":"School of Computer Science Engineering and Information Systems, Vellore Institute of Technology","place":["Vellore, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,22]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"13099","DOI":"10.1038\/s41598-026-42290-9","article-title":"Enhancing lung cancer classification through a double attention hybrid CNN-HiFuse approach","volume":"16","author":"Aaseegha","year":"2026","journal-title":"Sci. Rep."},{"key":"ref2","doi-asserted-by":"publisher","first-page":"1625183","DOI":"10.3389\/fmed.2025.1625183","article-title":"Automated classification and explainable AI analysis of lung cancer stages using EfficientNet and gradient-weighted class activation mapping","volume":"12","author":"Alqhatni","year":"2025","journal-title":"Front. Med."},{"key":"ref3","volume-title":"The IQ-OTH\/NCCD Lung Cancer Dataset","author":"Al-Yasriy","year":"2023"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"102629","DOI":"10.1016\/j.artmed.2023.102629","article-title":"Learnable DoG convolutional filters for microcalcification detection","volume":"143","author":"Cantone","year":"2023","journal-title":"Artif. Intell. Med."},{"key":"ref5","doi-asserted-by":"publisher","first-page":"112659","DOI":"10.1016\/j.knosys.2024.112659","article-title":"SSTL-FM: self-supervised transfer learning-based fusion model for the classification of benign-malignant lung nodules","volume":"305","author":"Cao","year":"2024","journal-title":"Knowl. Based Syst."},{"key":"ref6","first-page":"3435","author":"Chen","year":"2019"},{"key":"ref7","first-page":"4479","article-title":"Fast Fourier convolution","volume":"33","author":"Chi","year":"2020","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref8","doi-asserted-by":"publisher","first-page":"256","DOI":"10.1093\/biomet\/37.3-4.256","article-title":"The comparison of percentages in matched samples","volume":"37","author":"Cochran","year":"1950","journal-title":"Biometrika"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"11","DOI":"10.1016\/j.diii.2022.11.007","article-title":"Artificial intelligence: a critical review of applications for lung nodule and lung cancer","volume":"104","author":"de Margerie-Mellon","year":"2023","journal-title":"Diagn. Interv. Imaging"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"837","DOI":"10.2307\/2531595","article-title":"Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach","volume":"44","author":"DeLong","year":"1988","journal-title":"Biometrics"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"1648548","DOI":"10.3389\/fonc.2025.1648548","article-title":"Advances in modelling the risk of benign and malignant lung nodules","volume":"15","author":"Du","year":"2025","journal-title":"Front. Oncol."},{"key":"ref12","doi-asserted-by":"publisher","first-page":"3110","DOI":"10.1002\/mp.16812","article-title":"BUS-BRA: a breast ultrasound dataset for assessing computer-aided diagnosis systems","volume":"51","author":"G\u00f3mez-Flores","year":"2024","journal-title":"Med. Phys."},{"key":"ref13","doi-asserted-by":"publisher","first-page":"1550939","DOI":"10.3389\/fnbot.2025.1550939","article-title":"Multi-scale image edge detection based on spatial-frequency domain interactive attention","volume":"19","author":"Guo","year":"2025","journal-title":"Front. Neurorobot."},{"key":"ref14","doi-asserted-by":"publisher","first-page":"106118","DOI":"10.1016\/j.ijmedinf.2025.106118","article-title":"An attention aided wavelet convolutional neural network for lung nodule characterization","volume":"205","author":"Halder","year":"2025","journal-title":"Int. J. Med. Inform."},{"key":"ref15","doi-asserted-by":"publisher","first-page":"12707","DOI":"10.1038\/s41598-025-97645-5","article-title":"Explainable AI for lung cancer detection via a custom CNN on CT images","volume":"15","author":"Hammad","year":"2025","journal-title":"Sci. Rep."},{"key":"ref16","doi-asserted-by":"publisher","first-page":"24427","DOI":"10.1038\/s41598-024-75886-0","article-title":"Simulating clinical features on chest radiographs for medical image exploration and CNN explainability using a style-based generative adversarial autoencoder","volume":"14","author":"Hasenstab","year":"2024","journal-title":"Sci. Rep."},{"key":"ref17","doi-asserted-by":"publisher","first-page":"1423693","DOI":"10.3389\/fcomp.2024.1423693","article-title":"Lung tumor segmentation: a review of the state of the art","volume":"6","author":"Hiraman","year":"2024","journal-title":"Front. Comput. Sci."},{"key":"ref18","doi-asserted-by":"publisher","first-page":"65","DOI":"10.2307\/4615733","article-title":"A simple sequentially rejective multiple test procedure","volume":"6","author":"Holm","year":"1979","journal-title":"Scand. J. Stat."},{"key":"ref19","doi-asserted-by":"publisher","first-page":"107378","DOI":"10.1016\/j.compbiomed.2023.107378","article-title":"BAWGNet: boundary aware wavelet guided network for the nuclei segmentation in histopathology images","volume":"165","author":"Imtiaz","year":"2023","journal-title":"Comput. Biol. Med."},{"key":"ref20","doi-asserted-by":"publisher","first-page":"109233","DOI":"10.48550\/arXiv.2409.09951","article-title":"Optimal ablation for interpretability","volume":"37","author":"Li","year":"2024","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref21","doi-asserted-by":"publisher","first-page":"1773338","DOI":"10.3389\/fmed.2026.1773338","article-title":"LNMSNet: a multi-task deep learning network for pulmonary nodules segmentation and malignancy classification","volume":"13","author":"Liu","year":"2026","journal-title":"Front. Med."},{"key":"ref22","doi-asserted-by":"publisher","first-page":"281","DOI":"10.3311\/ppee.21533","article-title":"Practical automatic tuning of lag-lead compensators for feedback control systems","volume":"67","author":"Lu","year":"2023","journal-title":"Period. Polytech. Electr. Eng. Comput. Sci."},{"key":"ref23","doi-asserted-by":"publisher","first-page":"4357","DOI":"10.1109\/TIP.2018.2835143","article-title":"Gabor convolutional networks","volume":"27","author":"Luan","year":"2018","journal-title":"IEEE Trans. Image Process."},{"key":"ref24","doi-asserted-by":"publisher","first-page":"309","DOI":"10.1007\/s00418-020-01961-y","article-title":"Wavelet decomposition facilitates training on small datasets for medical image classification by deep learning","volume":"155","author":"Masquelin","year":"2021","journal-title":"Histochem. Cell Biol."},{"key":"ref25","doi-asserted-by":"publisher","first-page":"153","DOI":"10.1007\/BF02295996","article-title":"Note on the sampling error of the difference between correlated proportions or percentages","volume":"12","author":"McNemar","year":"1947","journal-title":"Psychometrika"},{"key":"ref26","doi-asserted-by":"publisher","first-page":"1901.08644","DOI":"10.48550\/arXiv.1901.08644","article-title":"Ablation studies in artificial neural networks","author":"Meyes","year":"2019","journal-title":"arXiv preprint"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"68441","DOI":"10.1007\/s11042-023-17996-x","article-title":"Weighted ensemble CNN for lung nodule classification: an evolutionary approach","volume":"83","author":"Naik","year":"2024","journal-title":"Multimed. Tools Appl."},{"key":"ref28","first-page":"763","author":"Qin","year":"2021"},{"key":"ref29","doi-asserted-by":"publisher","first-page":"3522","DOI":"10.1038\/s41598-024-53528-9","article-title":"Navigating the nuances: comparative analysis and hyperparameter optimisation of neural architectures on contrast-enhanced MRI for liver and liver tumour segmentation","volume":"14","author":"Quinton","year":"2024","journal-title":"Sci. Rep."},{"key":"ref30","doi-asserted-by":"publisher","first-page":"106902","DOI":"10.1016\/j.engappai.2023.106902","article-title":"Lung-EffNet: lung cancer classification using EfficientNet from CT-scan images","volume":"126","author":"Raza","year":"2023","journal-title":"Eng. Appl. Artif. Intell."},{"key":"ref31","doi-asserted-by":"publisher","first-page":"e00032","DOI":"10.14218\/ERHM.2025.00032","article-title":"Enhanced pulmonary nodule detection and classification using artificial intelligence on LIDC-IDRI data","volume":"11","author":"Salhi","year":"2026","journal-title":"Explor. Res. Hypothesis Med."},{"key":"ref32","doi-asserted-by":"publisher","first-page":"1353873","DOI":"10.3389\/frai.2024.1353873","article-title":"Image restoration in frequency space using complex-valued CNNs","volume":"7","author":"Shah","year":"2024","journal-title":"Front. Artif. Intell."},{"key":"ref33","doi-asserted-by":"publisher","first-page":"111443","DOI":"10.1016\/j.asoc.2024.111443","article-title":"A frequency-domain approach with learnable filters for image classification","volume":"155","author":"Stuchi","year":"2024","journal-title":"Appl. Soft Comput."},{"key":"ref34","doi-asserted-by":"publisher","first-page":"108505","DOI":"10.1016\/j.compbiomed.2024.108505","article-title":"Nodule-CLIP: lung nodule classification based on multi-modal contrastive learning","volume":"175","author":"Sun","year":"2024","journal-title":"Comput. Biol. Med."},{"key":"ref35","first-page":"6105","author":"Tan","year":"2019"},{"key":"ref36","doi-asserted-by":"publisher","first-page":"13885","DOI":"10.1007\/s11042-024-19475-3","article-title":"VSNet: classification of pulmonary nodules in 3D using vision transformer and sequence spatial attention mechanism","volume":"84","author":"Tang","year":"2024","journal-title":"Multimed. Tools Appl."},{"key":"ref37","doi-asserted-by":"publisher","first-page":"48","DOI":"10.1038\/s41746-022-00592-y","article-title":"Machine learning for medical imaging: methodological failures and recommendations for the future","volume":"5","author":"Varoquaux","year":"2022","journal-title":"NPJ Digit. Med."},{"key":"ref38","doi-asserted-by":"publisher","first-page":"39268","DOI":"10.1038\/s41598-025-23106-8","article-title":"Sensitivity analysis of data augmentation methods on performance of deep learning model for lung sounds classification","volume":"15","author":"Wang","year":"2025","journal-title":"Sci. Rep."},{"key":"ref39","doi-asserted-by":"publisher","first-page":"1636008","DOI":"10.3389\/fmed.2025.1636008","article-title":"DCAI: a dual cross-attention integration framework for benign-malignant classification of pulmonary nodules","volume":"12","author":"Wang","year":"2025","journal-title":"Front. Med."},{"key":"ref40","doi-asserted-by":"publisher","first-page":"24852","DOI":"10.1109\/access.2025.3529127","article-title":"Squeeze-and-excitation vision transformer for lung nodule classification","volume":"13","author":"Xue","year":"2025","journal-title":"IEEE Access"},{"key":"ref41","doi-asserted-by":"publisher","first-page":"107479","DOI":"10.1016\/j.bspc.2024.107479","article-title":"ETMO-NAS: an efficient two-step multimodal one-shot NAS for lung nodules classification","volume":"104","author":"Yu","year":"2025","journal-title":"Biomed. Signal Process. Control"},{"key":"ref42","doi-asserted-by":"publisher","first-page":"6802","DOI":"10.1109\/TNNLS.2024.3399164","article-title":"Pyramid pixel context adaption network for medical image classification with supervised contrastive learning","volume":"36","author":"Zhang","year":"2024","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"ref43","doi-asserted-by":"publisher","first-page":"e23508","DOI":"10.1016\/j.heliyon.2023.e23508","article-title":"Attention-guided deep neural network with a multichannel architecture for lung nodule classification","volume":"10","author":"Zheng","year":"2024","journal-title":"Heliyon"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1934912\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T05:39:35Z","timestamp":1790055575000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1934912\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,22]]},"references-count":43,"alternative-id":["10.3389\/frai.2026.1934912"],"URL":"https:\/\/doi.org\/10.3389\/frai.2026.1934912","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,22]]},"article-number":"1934912"}}