{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T09:24:56Z","timestamp":1783416296363,"version":"3.54.6"},"reference-count":68,"publisher":"PeerJ","license":[{"start":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T00:00:00Z","timestamp":1783382400000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>\n                    Accurate image segmentation in medical and industrial domains remains challenging due to small object sizes, complex textures, and diverse defect morphologies. To address these limitations, we propose Dense121GAN, a conditional generative adversarial network (cGAN) that integrates a pre-trained DenseNet121 as a frozen encoder. The proposed architecture promotes effective feature reuse and stable information flow through residual skip connections. By leveraging transfer learning, Dense121GAN captures rich hierarchical representations and improves training stability, particularly in data-constrained settings. Extensive experiments were conducted using five-fold cross-validation across 16 heterogeneous datasets. Dense121GAN consistently outperformed UNet and ResNet-based generative adversarial networks (GANs) across multiple evaluation metrics, including Intersection over Union (IoU), Dice coefficient (Dice), Matthews Correlation Coefficient (MCC), Cohen\u2019s Kappa, Structural Similarity Index Measure (SSIM), coefficient of determination (\n                    <jats:italic>R<\/jats:italic>\n                    <jats:sup>2<\/jats:sup>\n                    ), and Symmetric Mean Absolute Percentage Error (SMAPE). The model also demonstrated stable training behavior and mitigated common GAN failure modes such as mode collapse. In addition, Dense121GAN achieved faster convergence compared to the evaluated baseline architectures. These results highlight the effectiveness, robustness, and computational efficiency of Dense121GAN. The proposed framework provides a practical and versatile solution for both clinical diagnostic and industrial inspection applications.\n                  <\/jats:p>","DOI":"10.7717\/peerj-cs.4020","type":"journal-article","created":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T08:49:26Z","timestamp":1783414166000},"page":"e4020","source":"Crossref","is-referenced-by-count":0,"title":["Dense121GAN: transfer learning-enhanced conditional generative adversarial network with DenseNet121 for reliable and efficient segmentation in medical and industrial imaging"],"prefix":"10.7717","volume":"12","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6864-2339","authenticated-orcid":true,"given":"Muhammed","family":"Davud","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"4443","published-online":{"date-parts":[[2026,7,7]]},"reference":[{"key":"10.7717\/peerj-cs.4020\/ref-1","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1511.00111","article-title":"Regional active contours based on variational level sets and machine learning for image segmentation","author":"Abdelsamea","year":"2015"},{"issue":"17","key":"10.7717\/peerj-cs.4020\/ref-2","doi-asserted-by":"publisher","first-page":"17491","DOI":"10.1007\/s11042-024-19703-w","article-title":"Enhancing of polyp image segmentation in colonoscopy images: a comprehensive approach using modified UNet, hybrid color space, and ensemble learning","volume":"84","author":"Aghalari","year":"2024","journal-title":"Multimedia Tools and Applications"},{"key":"10.7717\/peerj-cs.4020\/ref-3","doi-asserted-by":"publisher","first-page":"104863","DOI":"10.1016\/j.dib.2019.104863","article-title":"Dataset of breast ultrasound images","volume":"28","author":"Al-Dhabyani","year":"2020","journal-title":"Data in Brief"},{"key":"10.7717\/peerj-cs.4020\/ref-4","article-title":"CVC-clinicDB polyp dataset","author":"Al-Ghoniy","year":"2021"},{"issue":"1","key":"10.7717\/peerj-cs.4020\/ref-5","doi-asserted-by":"publisher","first-page":"1027","DOI":"10.1038\/s41598-024-84504-y","article-title":"Explainable attention based breast tumor segmentation using a combination of UNet, ResNet, DenseNet, and EfficientNet models","volume":"15","author":"Anari","year":"2025","journal-title":"Scientific Reports"},{"key":"10.7717\/peerj-cs.4020\/ref-6","article-title":"BCCD dataset with segmentation masks","author":"Anmol","year":"2018"},{"key":"10.7717\/peerj-cs.4020\/ref-7","article-title":"MVTec defect detection dataset","author":"Anmol","year":"2019"},{"key":"10.7717\/peerj-cs.4020\/ref-8","first-page":"9592","article-title":"MVTec AD\u2013a comprehensive real-world dataset for unsupervised anomaly detection","author":"Bergmann","year":"2019"},{"key":"10.7717\/peerj-cs.4020\/ref-9","doi-asserted-by":"publisher","first-page":"99","DOI":"10.1016\/j.compmedimag.2015.02.007","article-title":"WM-DOVA maps for accurate polyp highlighting in colonoscopy: Validation vs. saliency maps from physicians","volume":"43","author":"Bernal","year":"2015","journal-title":"Computerized Medical Imaging and Graphics"},{"issue":"1","key":"10.7717\/peerj-cs.4020\/ref-10","doi-asserted-by":"publisher","first-page":"123","DOI":"10.1002\/ima.22795","article-title":"Squeeze and multi-context attention for polyp segmentation","volume":"33","author":"Bhattacharya","year":"2023","journal-title":"International Journal of Imaging Systems and Technology"},{"key":"10.7717\/peerj-cs.4020\/ref-11","first-page":"4985","article-title":"End-to-end referring video object segmentation with multimodal transformers","author":"Botach","year":"2022"},{"issue":"1","key":"10.7717\/peerj-cs.4020\/ref-12","doi-asserted-by":"publisher","first-page":"011001","DOI":"10.1088\/2632-2153\/ad1f77","article-title":"Ten years of generative adversarial nets (GANs): a survey of the state-of-the-art","volume":"5","author":"Chakraborty","year":"2024","journal-title":"Machine Learning: Science and Technology"},{"issue":"3","key":"10.7717\/peerj-cs.4020\/ref-13","doi-asserted-by":"publisher","first-page":"803","DOI":"10.1109\/tmi.2017.2764326","article-title":"Multimodal MR synthesis via modality-invariant latent representation","volume":"37","author":"Chartsias","year":"2017","journal-title":"IEEE Transactions on Medical Imaging"},{"issue":"1","key":"10.7717\/peerj-cs.4020\/ref-14","doi-asserted-by":"publisher","first-page":"9951582","DOI":"10.1155\/2022\/9951582","article-title":"Leukocyte segmentation method based on adaptive retinex correction and U-Net","volume":"2022","author":"Chen","year":"2022","journal-title":"Computational and Mathematical Methods in Medicine"},{"key":"10.7717\/peerj-cs.4020\/ref-15","first-page":"3910","article-title":"Test time training for industrial anomaly segmentation","author":"Costanzino","year":"2024"},{"key":"10.7717\/peerj-cs.4020\/ref-16","doi-asserted-by":"crossref","first-page":"70","DOI":"10.1007\/978-3-031-70924-1_6","article-title":"Utilizing transfer learning for enhanced classification of skin lesions through deep learning approaches","volume-title":"Recent Trends and Advances in Artificial Intelligence","author":"Davud","year":"2024"},{"key":"10.7717\/peerj-cs.4020\/ref-17","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.17935252","article-title":"Dense121GAN: cGAN framework for medical and industrial image segmentation","author":"Davud","year":"2025"},{"key":"10.7717\/peerj-cs.4020\/ref-18","doi-asserted-by":"publisher","DOI":"10.34740\/KAGGLE\/DSV\/6107556","article-title":"Blood cell segmentation dataset","author":"Depto","year":"2023"},{"key":"10.7717\/peerj-cs.4020\/ref-19","doi-asserted-by":"publisher","first-page":"101653","DOI":"10.1016\/j.tice.2021.101653","article-title":"Automatic segmentation of blood cells from microscopic slides: a comparative analysis","volume":"73","author":"Depto","year":"2021","journal-title":"Tissue and Cell"},{"key":"10.7717\/peerj-cs.4020\/ref-20","first-page":"1","article-title":"An efficient polyp segmentation network","author":"Erol","year":"2022"},{"key":"10.7717\/peerj-cs.4020\/ref-21","first-page":"302","article-title":"Selective feature aggregation network with area-boundary constraints for polyp segmentation","author":"Fang","year":"2019"},{"key":"10.7717\/peerj-cs.4020\/ref-22","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2302.01027","article-title":"FCB-SwinV2 transformer for polyp segmentation","author":"Fitzgerald","year":"2023"},{"key":"10.7717\/peerj-cs.4020\/ref-23","first-page":"289","article-title":"Synthetic data augmentation using GAN for improved liver lesion classification","author":"Frid-Adar","year":"2018"},{"key":"10.7717\/peerj-cs.4020\/ref-24","first-page":"27","article-title":"Generative adversarial nets","author":"Goodfellow","year":"2014"},{"issue":"1","key":"10.7717\/peerj-cs.4020\/ref-25","doi-asserted-by":"publisher","first-page":"012029","DOI":"10.1088\/1742-6596\/2273\/1\/012029","article-title":"Deep learning (CNN) and transfer learning: a review","volume":"2273","author":"Gupta","year":"2022","journal-title":"Journal of Physics: Conference Series"},{"issue":"1","key":"10.7717\/peerj-cs.4020\/ref-26","doi-asserted-by":"publisher","first-page":"3614","DOI":"10.1038\/s41598-025-86588-6","article-title":"Adapting a style based generative adversarial network to create images depicting cleft lip deformity","volume":"15","author":"Hayajneh","year":"2025","journal-title":"Scientific Reports"},{"key":"10.7717\/peerj-cs.4020\/ref-27","first-page":"770","article-title":"Deep residual learning for image recognition","author":"He","year":"2016"},{"key":"10.7717\/peerj-cs.4020\/ref-28","doi-asserted-by":"publisher","first-page":"e2016","DOI":"10.7717\/peerj-cs.2016","article-title":"Predictive maintenance in industry 4.0: a survey of planning models and machine learning techniques","volume":"10","author":"Hector","year":"2024","journal-title":"PeerJ Computer Science"},{"key":"10.7717\/peerj-cs.4020\/ref-29","first-page":"1989","article-title":"CyCADA: cycle-consistent adversarial domain adaptation","author":"Hoffman","year":"2018"},{"key":"10.7717\/peerj-cs.4020\/ref-30","first-page":"4700","article-title":"Densely connected convolutional networks","author":"Huang","year":"2017"},{"key":"10.7717\/peerj-cs.4020\/ref-31","doi-asserted-by":"publisher","first-page":"104667","DOI":"10.1016\/j.bspc.2023.104667","article-title":"Tuberculosis chest X-ray detection using CNN-based hybrid segmentation and classification approach","volume":"84","author":"Iqbal","year":"2023","journal-title":"Biomedical Signal Processing and Control"},{"key":"10.7717\/peerj-cs.4020\/ref-32","first-page":"1","article-title":"Comparison of deep neural networks for lung image segmentation on chest X-ray","author":"Jena","year":"2024"},{"key":"10.7717\/peerj-cs.4020\/ref-33","first-page":"451","article-title":"Kvasir-SEG: a segmented polyp dataset","author":"Jha","year":"2020"},{"key":"10.7717\/peerj-cs.4020\/ref-34","article-title":"Kvasir-seg dataset","author":"Kaggle","year":"2021a"},{"key":"10.7717\/peerj-cs.4020\/ref-35","article-title":"Kvasir-Sessile dataset","author":"Kaggle","year":"2021b"},{"key":"10.7717\/peerj-cs.4020\/ref-36","doi-asserted-by":"publisher","first-page":"49","DOI":"10.13164\/mendel.2022.2.049","article-title":"Segmentation of chest X-ray images using U-net model","volume":"28","author":"Kamil","year":"2022","journal-title":"Mendel"},{"key":"10.7717\/peerj-cs.4020\/ref-37","article-title":"Chest X-ray lung segmentation dataset","author":"Karmakar","year":"2020"},{"key":"10.7717\/peerj-cs.4020\/ref-38","article-title":"Chest X-ray lungs segmentation (montgomery and shenzhen)","author":"Karmakar","year":"2022"},{"key":"10.7717\/peerj-cs.4020\/ref-39","doi-asserted-by":"publisher","first-page":"187536\u2013187571","DOI":"10.1109\/ACCESS.2024.3470122","article-title":"Image data augmentation approaches: a comprehensive survey and future directions","volume":"12","author":"Kumar","year":"2024","journal-title":"IEEE Access"},{"key":"10.7717\/peerj-cs.4020\/ref-40","first-page":"1","article-title":"Industrial surface defect detection via multi-scale mask cross-layer fusion network","author":"Liu","year":"2024"},{"issue":"11","key":"10.7717\/peerj-cs.4020\/ref-41","doi-asserted-by":"publisher","first-page":"5799","DOI":"10.1007\/s00521-023-09374-3","article-title":"CellSegUNet: an improved deep segmentation model for the cell segmentation based on UNet++and residual UNet models","volume":"36","author":"Metlek","year":"2024","journal-title":"Neural Computing and Applications"},{"key":"10.7717\/peerj-cs.4020\/ref-42","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1411.1784","article-title":"Conditional generative adversarial nets","author":"Mirza","year":"2014"},{"key":"10.7717\/peerj-cs.4020\/ref-43","doi-asserted-by":"publisher","first-page":"e3248","DOI":"10.7717\/peerj-cs.3248","article-title":"Deep feature fusion for melanoma detection using ResNet-VGG networks with attention mechanism","volume":"11","author":"Naeem","year":"2025","journal-title":"PeerJ Computer Science"},{"key":"10.7717\/peerj-cs.4020\/ref-44","doi-asserted-by":"publisher","first-page":"115275","DOI":"10.1016\/j.eswa.2021.115275","article-title":"Semi-supervised anomaly detection for visual quality inspection","volume":"183","author":"Napoletano","year":"2021","journal-title":"Expert Systems with Applications"},{"issue":"10","key":"10.7717\/peerj-cs.4020\/ref-45","doi-asserted-by":"publisher","first-page":"e3","DOI":"10.23915\/distill.00003","article-title":"Deconvolution and checkerboard artifacts","volume":"1","author":"Odena","year":"2016","journal-title":"Distill"},{"key":"10.7717\/peerj-cs.4020\/ref-46","first-page":"71","article-title":"Fully convolutional neural network for lungs segmentation from chest X-rays","author":"Rashid","year":"2018"},{"issue":"3","key":"10.7717\/peerj-cs.4020\/ref-47","doi-asserted-by":"publisher","first-page":"2557024","DOI":"10.1142\/s0218001425570241","article-title":"T-GAN: transformer generative adversarial network for brain tumor segmentation","volume":"40","author":"Reddy","year":"2026","journal-title":"International Journal of Pattern Recognition and Artificial Intelligence"},{"key":"10.7717\/peerj-cs.4020\/ref-48","first-page":"234","article-title":"U-Net: convolutional networks for biomedical image segmentation","author":"Ronneberger","year":"2015"},{"key":"10.7717\/peerj-cs.4020\/ref-49","article-title":"Breast ultrasound images dataset (BUSI)","author":"Shah","year":"2020"},{"key":"10.7717\/peerj-cs.4020\/ref-50","doi-asserted-by":"publisher","first-page":"e1813","DOI":"10.7717\/peerj-cs.1813","article-title":"Blood cell image segmentation and classification: a systematic review","volume":"10","author":"Shahzad","year":"2024","journal-title":"PeerJ Computer Science"},{"issue":"9","key":"10.7717\/peerj-cs.4020\/ref-51","doi-asserted-by":"crossref","first-page":"2132","DOI":"10.3390\/diagnostics12092132","article-title":"Segmentation-based classification deep learning model embedded with explainable AI for COVID-19 detection in chest X-ray scans","volume":"12","author":"Sharma","year":"2022","journal-title":"Diagnostics"},{"key":"10.7717\/peerj-cs.4020\/ref-52","first-page":"588","article-title":"Multi-scale convolutional neural networks for lung nodule classification","author":"Shen","year":"2015"},{"issue":"1","key":"10.7717\/peerj-cs.4020\/ref-53","doi-asserted-by":"publisher","first-page":"2995","DOI":"10.1007\/s11042-023-15307-y","article-title":"Deep learning-based automated steel surface defect segmentation: a comparative experimental study","volume":"83","author":"Sime","year":"2024","journal-title":"Multimedia Tools and Applications"},{"key":"10.7717\/peerj-cs.4020\/ref-54","doi-asserted-by":"publisher","first-page":"285","DOI":"10.1016\/j.cmpb.2019.06.005","article-title":"An automatic method for lung segmentation and reconstruction in chest X-ray using deep neural networks","volume":"177","author":"Souza","year":"2019","journal-title":"Computer Methods and Programs in Biomedicine"},{"issue":"7","key":"10.7717\/peerj-cs.4020\/ref-55","doi-asserted-by":"publisher","first-page":"1455","DOI":"10.1109\/tbme.2015.2496264","article-title":"A dataset for breast cancer histopathological image classification","volume":"63","author":"Spanhol","year":"2015","journal-title":"IEEE Transactions on Biomedical Engineering"},{"issue":"1","key":"10.7717\/peerj-cs.4020\/ref-56","doi-asserted-by":"publisher","first-page":"135","DOI":"10.46387\/bjesr.1261689","article-title":"Mikroskobik kan h\u00fccre g\u00f6r\u00fcnt\u00fclerinin g\u00fcncel derin \u00f6\u011frenme mimarileri ile b\u00f6l\u00fctlemesi","volume":"5","author":"Topta\u015f","year":"2023","journal-title":"M\u00fchendislik Bilimleri ve Ara\u015ft\u0131rmalar\u0131 Dergisi"},{"key":"10.7717\/peerj-cs.4020\/ref-57","first-page":"3992","article-title":"Multi-scale patch-based representation learning for image anomaly detection and segmentation","author":"Tsai","year":"2022"},{"key":"10.7717\/peerj-cs.4020\/ref-58","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2409.04356","article-title":"Serp-Mamba: advancing high-resolution retinal vessel segmentation with selective state-space model","author":"Wang","year":"2024a"},{"issue":"6","key":"10.7717\/peerj-cs.4020\/ref-59","doi-asserted-by":"publisher","first-page":"1607","DOI":"10.1007\/s11760-021-02115-w","article-title":"Multiscale transunet++: dense hybrid U-Net with transformer for medical image segmentation","volume":"16","author":"Wang","year":"2022","journal-title":"Signal, Image and Video Processing"},{"key":"10.7717\/peerj-cs.4020\/ref-60","first-page":"5527","article-title":"Language-driven interactive shadow detection","author":"Wang","year":"2024b"},{"issue":"12","key":"10.7717\/peerj-cs.4020\/ref-61","doi-asserted-by":"publisher","first-page":"4457","DOI":"10.1109\/tmi.2024.3426953","article-title":"Video-instrument synergistic network for referring video instrument segmentation in robotic surgery","volume":"43","author":"Wang","year":"2024c","journal-title":"IEEE Transactions on Medical Imaging"},{"key":"10.7717\/peerj-cs.4020\/ref-62","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2401.13560","article-title":"SegMamba: long-range sequential modeling mamba for 3D medical image segmentation","author":"Xing","year":"2024"},{"issue":"10","key":"10.7717\/peerj-cs.4020\/ref-63","doi-asserted-by":"publisher","first-page":"1034","DOI":"10.3390\/bioengineering11101034","article-title":"Advances in medical image segmentation: a comprehensive review of traditional, deep learning and hybrid approaches","volume":"11","author":"Xu","year":"2024","journal-title":"Bioengineering"},{"key":"10.7717\/peerj-cs.4020\/ref-64","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2509.00833","article-title":"SegDINO: an efficient design for medical and natural image segmentation with DINO-V3","author":"Yang","year":"2025"},{"key":"10.7717\/peerj-cs.4020\/ref-65","first-page":"7354","article-title":"Self-attention generative adversarial networks","author":"Zhang","year":"2019"},{"key":"10.7717\/peerj-cs.4020\/ref-66","first-page":"258","article-title":"An image segmentation method based on improved watershed algorithm","author":"Zhang","year":"2010"},{"issue":"6","key":"10.7717\/peerj-cs.4020\/ref-67","doi-asserted-by":"publisher","first-page":"1856","DOI":"10.1109\/tmi.2019.2959609","article-title":"UNet++: redesigning skip connections to exploit multiscale features in image segmentation","volume":"39","author":"Zhou","year":"2019","journal-title":"IEEE Transactions on Medical Imaging"},{"key":"10.7717\/peerj-cs.4020\/ref-68","first-page":"2223","article-title":"Unpaired image-to-image translation using cycle-consistent adversarial networks","author":"Zhu","year":"2017"}],"container-title":["PeerJ Computer Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/peerj.com\/articles\/cs-4020.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-4020.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-4020.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-4020.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T08:49:31Z","timestamp":1783414171000},"score":1,"resource":{"primary":{"URL":"https:\/\/peerj.com\/articles\/cs-4020"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,7]]},"references-count":68,"alternative-id":["10.7717\/peerj-cs.4020"],"URL":"https:\/\/doi.org\/10.7717\/peerj-cs.4020","archive":["CLOCKSS","LOCKSS","Portico"],"relation":{},"ISSN":["2376-5992"],"issn-type":[{"value":"2376-5992","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,7]]},"article-number":"e4020"}}