{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T05:04:22Z","timestamp":1789362262567,"version":"build-2803163510"},"reference-count":38,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T00:00:00Z","timestamp":1789344000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Hum. Neurosci."],"abstract":"<jats:p>\n                    Timely and accurate diagnosis of brain tumors using magnetic resonance imaging (MRI) is a major concern in neuro-oncology because of the potential effects of delayed and inconsistent radiological diagnoses on the management of such diseases. Manual analysis of MRIs is time-consuming and highly susceptible to observer variations; thus, there is a need for computer-based and more consistent diagnostic processes. In this study, an automatic diagnostic process based on deep learning was proposed using multiscale contrast enhancement along with lightweight convolutional features and classifiers for diagnosing four classes of brain tumors. The MRI images were analyzed using a Multi-scale Intensity Contrast Amplification (MICA) algorithm that enhances the relevant structural information of the tumors across fine, intermediate, and coarse spatial scales. The refined images were then fed into an in-house convolutional neural network (CNN) backbone architecture that uses depth-wise separable convolution, residual skip connection, and squeeze-and-excitation (SE) attention mechanism to extract compact and discriminative feature representations. To select the optimal decision layer for this representation, three supervised learning methods, namely K-Nearest Neighbors (KNN), Support Vector Machine (SVM), and Random Forest (RF), were trained on this set of features and compared with the softmax head of the CNN. The experiments were performed on a publicly available, three-source combined MRI dataset of 7,023 images across glioma, meningioma, pituitary tumor, and normal categories using both hold-out and 5-fold cross-validation setups. In all the evaluated configurations, the hybrid model of CNN and KNN provided the best hold-out accuracy (96.58%), whereas the hybrid model of CNN and SVM offered the best macro-average precision (98.81%) and area under the ROC curve (99.56%). Both hybrids were statistically superior (\n                    <jats:italic>p<\/jats:italic>\n                    \u202f&amp;lt;\u202f0.001) to the softmax head of the CNN in the paired statistical tests. Class Activation visualization based on gradient analysis revealed that the network concentrated on clinically sensible regions of the tumors rather than other irrelevant background regions. With a compact parameter footprint (0.20\u202fMB in half precision) and an inference throughput of approximately 23.7 images per second on the CPU and 220.1 images per second on an NVIDIA A100 GPU at a batch size of 16, the proposed framework offers an accurate, interpretable, and computationally frugal solution that is suitable for computer-aided brain tumor screening in resource-constrained clinical settings.\n                  <\/jats:p>","DOI":"10.3389\/fnhum.2026.1883543","type":"journal-article","created":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T04:33:55Z","timestamp":1789360435000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Toward reliable computer-aided brain tumor diagnosis: a contrast-enhanced deep learning approach with hybrid KNN classification"],"prefix":"10.3389","volume":"20","author":[{"given":"Krithika","family":"Pandurangan","sequence":"first","affiliation":[{"name":"Department of MBA, SRM Institute of Science and Technology","place":["Chennai, Tamil Nadu, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Doddi Chandana","family":"Gowri","sequence":"additional","affiliation":[{"name":"Department of CSE, SRM University","place":["Amaravathi, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jonnla Naga","family":"Surekha","sequence":"additional","affiliation":[{"name":"Department of Electronics and Communication Engineering, Dr RVR NRI Institute of Technology (Deemed to be University)","place":["Eluru, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pratiksha V.","family":"Chafle","sequence":"additional","affiliation":[{"name":"Department of CSE, G H Raisoni University Amravati","place":["Maharashtra, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kolagotla Venkateswara","family":"Reddy","sequence":"additional","affiliation":[{"name":"Department of Mechanical Engineering, Lakireddy Balireddy College of Engineering","place":["Vijayawada, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shonak","family":"Bansal","sequence":"additional","affiliation":[{"name":"Department of Electronics and Communication Engineering, Chandigarh University","place":["Gharuan, Punjab, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Krishna","family":"Prakash","sequence":"additional","affiliation":[{"name":"Department of Information and Communication Technology, Marwadi University","place":["Rajkot, Gujarat, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Md. Shabiul","family":"Islam","sequence":"additional","affiliation":[{"name":"Centre for Advanced Devices and Systems, Centre of Excellence for Robotics and Sensing Technologies, Multimedia University","place":["Cyberjaya, Selangor, Malaysia"]},{"name":"Faculty of Artificial Intelligence and Engineering, Multimedia University","place":["Cyberjaya, Selangor, Malaysia"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mohammad","family":"Tariqul Islam","sequence":"additional","affiliation":[{"name":"Department of Electrical, Electronic and Systems Engineering, Faculty of Engineering and Built Environment, UKM Bangi","place":["Bangi, Selangor, Malaysia"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,14]]},"reference":[{"key":"ref1","author":"Nickparvar","year":"2026"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"27","DOI":"10.1080\/0954898X.2023.2275720","article-title":"Brain tumour classification using MRI images based on lenet with golden teacher learning optimization","volume":"35","author":"Aluri","year":"2024","journal-title":"Netw. Comput. Neural Syst."},{"key":"ref3","first-page":"1128","volume-title":"Deep Learning-Based Approach for Improved Brain Tumor Identification in CT Images using Sophisticated Neural Network Models 2025 5th International Conference on Advancement in Electronics and Communication Engineering (AECE)","author":"Bhuvaneshwari","year":"2025"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1023\/a:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Mach. Learn."},{"key":"ref5","first-page":"1","volume-title":"Brain Tumor Detection using Machine Learning 2024 Second International Conference on Data Science and Information System (ICDSIS)","author":"Castelino","year":"2024"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"273","DOI":"10.1023\/a:1022627411411","article-title":"Support-vector networks","volume":"20","author":"Cort\u00e9s","year":"1995","journal-title":"Mach. Learn."},{"key":"ref7","doi-asserted-by":"publisher","first-page":"347","DOI":"10.1016\/j.ecl.2020.04.002","article-title":"The epidemiology of pituitary adenomas","volume":"49","author":"Daly","year":"2020","journal-title":"Endocrinol. Metab. Clin. N. Am."},{"key":"ref8","doi-asserted-by":"publisher","first-page":"653","DOI":"10.3390\/info15100653","article-title":"Enhancing brain tumor detection through custom convolutional neural networks and interpretability-driven analysis","volume":"15","author":"Dewage","year":"2024","journal-title":"Information"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"105419","DOI":"10.1016\/j.bspc.2023.105419","article-title":"Hybrid archimedes sine cosine optimization enabled deep learning for multilevel brain tumor classification using mri images","volume":"87","author":"Geetha","year":"2024","journal-title":"Biomed. Signal Process. Control"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"108971","DOI":"10.1016\/j.compbiomed.2024.108971","article-title":"Brain-gcn-net: graph-convolutional neural network for brain tumor identification","volume":"180","author":"G\u00fcrsoy","year":"2024","journal-title":"Comput. Biol. Med."},{"key":"ref11","doi-asserted-by":"publisher","first-page":"1","DOI":"10.3171\/2023.8.SPINE23190","article-title":"Treatment of intramedullary spinal cord tumors: a modified Delphi technique of the North American Spine Society section of spine oncology: presented at the 2023 AANS\/CNS joint section on disorders of the spine and peripheral nerves","volume":"40","author":"Hersh","year":"2023","journal-title":"J. Neurosurg. Spine"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"6319","DOI":"10.1038\/s41598-022-10309-6","article-title":"A deep learning model to classify and detect brain abnormalities in portable microwave based imaging system","volume":"12","author":"Hossain","year":"2022","journal-title":"Sci. Rep."},{"key":"ref13","doi-asserted-by":"publisher","first-page":"302","DOI":"10.3390\/bios13030302","article-title":"Brain tumor segmentation and classification from sensor-based portable microwave brain imaging system using lightweight deep learning models","volume":"13","author":"Hossain","year":"2023","journal-title":"Biosensors"},{"key":"ref14","first-page":"1","volume-title":"Brain Tumor Detection using Extended Optimization Based Multi-Layer Perceptron 2025 International Conference on Computing Technologies & Data Communication (ICCTDC)","author":"Jemina","year":"2025"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"40177","DOI":"10.1038\/s41598-025-23813-2","article-title":"An intelligent brain tumor detection model using lightweight hybrid twin attentive pyramid convolutional network","volume":"15","author":"Jemina","year":"2025","journal-title":"Sci. Rep."},{"key":"ref16","doi-asserted-by":"publisher","first-page":"19789","DOI":"10.1007\/s13369-025-10290-y","article-title":"DBTNet: a dual-stream neural network for effective brain tumor detection in MRI images","volume":"50","author":"Kamireddy","year":"2025","journal-title":"Arab. J. Sci. Eng."},{"key":"ref17","doi-asserted-by":"publisher","first-page":"126633","DOI":"10.1016\/j.eswa.2025.126633","article-title":"Brain tumor classification from MRI images using exponential-walruses hunting optimization driven SqueezeNet","volume":"271","author":"Kanchanamala","year":"2025","journal-title":"Expert Syst. Appl."},{"key":"ref18","doi-asserted-by":"publisher","first-page":"248","DOI":"10.1007\/s10586-024-04940-3","article-title":"Detection of MRI brain tumor using residual skip block based modified MobileNet model","volume":"28","author":"Khan","year":"2025","journal-title":"Clust. Comput."},{"key":"ref19","doi-asserted-by":"publisher","first-page":"29765","DOI":"10.1038\/s41598-025-13281-z","article-title":"A novel brain MRI classification framework integrating tuned single scale retinex and empirical wavelet entropy features","volume":"15","author":"Kumar","year":"2025","journal-title":"Sci. Rep."},{"key":"ref20","doi-asserted-by":"publisher","first-page":"1535478","DOI":"10.3389\/fonc.2025.1535478","article-title":"Explainable AI in medical imaging: an interpretable and collaborative federated learning model for brain tumor classification","volume":"15","author":"Latif","year":"2025","journal-title":"Front. Oncol."},{"key":"ref21","doi-asserted-by":"publisher","first-page":"3596","DOI":"10.1038\/s41598-025-86752-y","article-title":"Dilated SE-DenseNet for brain tumor MRI classification","volume":"15","author":"Mao","year":"2025","journal-title":"Sci. Rep."},{"key":"ref22","doi-asserted-by":"publisher","first-page":"540","DOI":"10.1177\/08953996251317416","article-title":"Multi-axis transformer based U-net with class balanced ensemble model for lung disease classification using X-ray images","volume":"33","author":"Maruthai","year":"2025","journal-title":"J. X Ray Sci. Technol."},{"key":"ref23","doi-asserted-by":"publisher","first-page":"e01224","DOI":"10.48550\/arXiv.1908.01224","article-title":"Smooth grad-cam++: an enhanced inference level visualization technique for deep convolutional neural network models","author":"Omeiza","year":"2019","journal-title":"arXiv"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"1883","DOI":"10.4249\/scholarpedia.1883","article-title":"K-nearest neighbor","volume":"4","author":"Peterson","year":"2009","journal-title":"Scholarpedia"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"465","DOI":"10.1007\/s00521-026-12171-3","article-title":"Automated MRI brain abnormality detection using U-net and ReX-net","volume":"38","author":"Prakash","year":"2026","journal-title":"Neural Comput. & Applic."},{"key":"ref26","doi-asserted-by":"publisher","first-page":"64758","DOI":"10.1109\/ACCESS.2023.3288017","article-title":"Automated segmentation of brain tumor MRI images using deep learning","volume":"11","author":"Rajendran","year":"2023","journal-title":"IEEE Access"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"1405586","DOI":"10.3389\/fnhum.2024.1405586","article-title":"BrainCDNet: a concatenated deep neural network for the detection of brain tumors from MRI images","volume":"18","author":"Reddy","year":"2024","journal-title":"Front. Hum. Neurosci."},{"key":"ref28","doi-asserted-by":"publisher","first-page":"240","DOI":"10.3174\/ajnr.A8368","article-title":"Meningioma: molecular updates from the 2021 World Health Organization classification of CNS tumors and imaging correlates","volume":"46","author":"Soni","year":"2025","journal-title":"Am. J. Neuroradiol."},{"key":"ref29","doi-asserted-by":"publisher","first-page":"1301","DOI":"10.12785\/ijcds\/1301106","article-title":"A new distinctive methodology for the classification of brain MR images using histogram based local feature descriptors","volume":"13","author":"Sowjanya","year":"2023","journal-title":"International Journal of Computing and Digital Systems"},{"key":"ref30","doi-asserted-by":"publisher","first-page":"209","DOI":"10.3322\/caac.21660","article-title":"Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries","volume":"71","author":"Sung","year":"2021","journal-title":"CA Cancer J. Clin."},{"key":"ref31","doi-asserted-by":"publisher","first-page":"340","DOI":"10.12720\/jait.15.3.340-354","article-title":"Efficient brain tumor classification with a hybrid CNN-SVM approach in MRI","volume":"15","author":"Suryawanshi","year":"2024","journal-title":"Journal of Advances in Information Technology"},{"key":"ref32","first-page":"1","volume-title":"BrainVisionNet: A deep Learning-based Approach to Evaluate the Potential of Microwave Imaging for Classification of brain Tumors 2023 International Conference on Next-Generation Computing, IoT and Machine Learning (NCIM)","author":"Taha","year":"2023"},{"key":"ref33","first-page":"1","volume-title":"Brain tumor Detection from MRI Images based on ResNet18 2023 6th International Conference on Information Systems and Computer Networks (ISCON)","author":"Tang","year":"2023"},{"key":"ref34","doi-asserted-by":"publisher","first-page":"64","DOI":"10.3390\/engproc2025087064","article-title":"Optimizing brain tumor classification: integrating deep learning and machine learning with hyperparameter tuning","volume":"87","author":"Velpula","year":"2025","journal-title":"Eng. Proc."},{"key":"ref35","first-page":"551","article-title":"Multiscale image contrast amplification (MUSICA) medical imaging 1994: image processing","author":"Vuylsteke","year":"1994"},{"key":"ref36","doi-asserted-by":"publisher","first-page":"14611","DOI":"10.1007\/s00521-021-05841-x","article-title":"Deep neural network correlation learning mechanism for CT brain tumor detection","volume":"35","author":"Wo\u017aniak","year":"2023","journal-title":"Neural Comput. & Applic."},{"key":"ref37","doi-asserted-by":"publisher","first-page":"47","DOI":"10.1177\/08953996241299996","article-title":"Enhancing brain tumor classification by integrating radiomics and deep learning features: a comprehensive study utilizing ensemble methods on MRI scans","volume":"33","author":"Yin","year":"2025","journal-title":"J. Xray Sci. Technol."},{"key":"ref38","doi-asserted-by":"publisher","first-page":"10026","DOI":"10.48550\/arXiv.2404.10026","article-title":"Distributed federated learning-based deep learning model for privacy MRI brain tumor detection","author":"Zhou","year":"2024","journal-title":"arXiv"}],"container-title":["Frontiers in Human Neuroscience"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnhum.2026.1883543\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T04:34:03Z","timestamp":1789360443000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnhum.2026.1883543\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,14]]},"references-count":38,"alternative-id":["10.3389\/fnhum.2026.1883543"],"URL":"https:\/\/doi.org\/10.3389\/fnhum.2026.1883543","relation":{},"ISSN":["1662-5161"],"issn-type":[{"value":"1662-5161","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,14]]},"article-number":"1883543"}}