{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,9]],"date-time":"2026-07-09T15:39:25Z","timestamp":1783611565623,"version":"3.55.0"},"reference-count":59,"publisher":"MDPI AG","issue":"8","license":[{"start":{"date-parts":[[2023,4,20]],"date-time":"2023-04-20T00:00:00Z","timestamp":1681948800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Digital Transformation Office of the Presidency of the Republic of T\u00fcrkiye"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Diagnostics"],"abstract":"<jats:p>This paper proposes ensemble strategies for the deep learning object detection models carried out by combining the variants of a model and different models to enhance the anatomical and pathological object detection performance in brain MRI. In this study, with the help of the novel Gazi Brains 2020 dataset, five different anatomical parts and one pathological part that can be observed in brain MRI were identified, such as the region of interest, eye, optic nerves, lateral ventricles, third ventricle, and a whole tumor. Firstly, comprehensive benchmarking of the nine state-of-the-art object detection models was carried out to determine the capabilities of the models in detecting the anatomical and pathological parts. Then, four different ensemble strategies for nine object detectors were applied to boost the detection performance using the bounding box fusion technique. The ensemble of individual model variants increased the anatomical and pathological object detection performance by up to 10% in terms of the mean average precision (mAP). In addition, considering the class-based average precision (AP) value of the anatomical parts, an up to 18% AP improvement was achieved. Similarly, the ensemble strategy of the best different models outperformed the best individual model by 3.3% mAP. Additionally, while an up to 7% better FAUC, which is the area under the TPR vs. FPPI curve, was achieved on the Gazi Brains 2020 dataset, a 2% better FAUC score was obtained on the BraTS 2020 dataset. The proposed ensemble strategies were found to be much more efficient in finding the anatomical and pathological parts with a small number of anatomic objects, such as the optic nerve and third ventricle, and producing higher TPR values, especially at low FPPI values, compared to the best individual methods.<\/jats:p>","DOI":"10.3390\/diagnostics13081494","type":"journal-article","created":{"date-parts":[[2023,4,21]],"date-time":"2023-04-21T01:33:43Z","timestamp":1682040823000},"page":"1494","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":15,"title":["An Ensemble of Deep Learning Object Detection Models for Anatomical and Pathological Regions in Brain MRI"],"prefix":"10.3390","volume":"13","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2345-8666","authenticated-orcid":false,"given":"Ramazan","family":"Terzi","sequence":"first","affiliation":[{"name":"Department of Big Data and Artificial Intelligence, Digital Transformation Office of the Presidency of Republic of T\u00fcrkiye, Ankara 06100, Turkey"},{"name":"Department of Computer Engineering, Amasya University, Amasya 05100, Turkey"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2023,4,20]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"202","DOI":"10.1053\/j.semnuclmed.2010.12.001","article-title":"Evolution of brain imaging instrumentation","volume":"41","author":"Abraham","year":"2011","journal-title":"Semin. Nucl. Med."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"718","DOI":"10.1016\/j.neubiorev.2006.06.001","article-title":"Brain development in children and adolescents: Insights from anatomical magnetic resonance imaging","volume":"30","author":"Lenroot","year":"2006","journal-title":"Neurosci. Biobehav. Rev."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"538","DOI":"10.1016\/j.neuroimage.2006.04.214","article-title":"A stereotaxic MRI template set for the rat brain with tissue class distribution maps and co-registered anatomical atlas: Application to pharmacological MRI","volume":"32","author":"Schwarz","year":"2006","journal-title":"Neuroimage"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"64","DOI":"10.1016\/j.artmed.2018.08.008","article-title":"Deep convolutional neural networks for brain image analysis on magnetic resonance imaging: A review","volume":"95","author":"Bernal","year":"2019","journal-title":"Artif. Intell. Med."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"486","DOI":"10.1162\/neco.2007.09-06-340","article-title":"Brain reading using full brain support vector machines for object recognition: There is no \u201cface\u201d identification area","volume":"20","author":"Hanson","year":"2008","journal-title":"Neural Comput."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"449","DOI":"10.1007\/s10278-017-9983-4","article-title":"Deep learning for brain MRI segmentation: State of the art and future directions","volume":"30","author":"Akkus","year":"2017","journal-title":"J. Digit. Imaging"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"70","DOI":"10.1016\/j.mri.2019.06.018","article-title":"Anatomical context improves deep learning on the brain age estimation task","volume":"62","author":"Bermudez","year":"2019","journal-title":"Magn. Reson. Imaging"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"212","DOI":"10.1109\/42.387703","article-title":"Object recognition in brain CT-scans: Knowledge-based fusion of data from multiple feature extractors","volume":"14","author":"Li","year":"1995","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1097\/00019052-199702000-00003","article-title":"Functional anatomy of object recognition in humans: Evidence from positron emission tomography and functional magnetic resonance imaging","volume":"10","author":"Bly","year":"1997","journal-title":"Curr. Opin. Neurol."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"115","DOI":"10.1016\/j.neuroimage.2006.05.061","article-title":"Automatic anatomical brain MRI segmentation combining label propagation and decision fusion","volume":"33","author":"Heckemann","year":"2006","journal-title":"NeuroImage"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"495","DOI":"10.1109\/TMI.2007.908121","article-title":"Brain anatomical structure segmentation by hybrid discriminative\/generative models","volume":"27","author":"Tu","year":"2008","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"777","DOI":"10.1109\/TMI.2011.2180920","article-title":"Hierarchical scale-based multiobject recognition of 3-D anatomical structures","volume":"31","author":"Bagci","year":"2011","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"713","DOI":"10.1109\/TMI.2011.2175940","article-title":"Hierarchical statistical shape models of multiobject anatomical structures: Application to brain MRI","volume":"31","author":"Cerrolaza","year":"2011","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"929","DOI":"10.1016\/j.media.2013.05.004","article-title":"Joint segmentation of anatomical and functional images: Applications in quantification of lesions from PET, PET-CT, MRI-PET, and MRI-PET-CT images","volume":"17","author":"Bagci","year":"2013","journal-title":"Med. Image Anal."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"de Brebisson, A., and Montana, G. (2015, January 7\u201312). Deep neural networks for anatomical brain segmentation. Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Boston, MA, USA.","DOI":"10.1109\/CVPRW.2015.7301312"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"156","DOI":"10.1016\/j.media.2019.02.007","article-title":"Evaluating reinforcement learning agents for anatomical landmark detection","volume":"53","author":"Alansary","year":"2019","journal-title":"Med. Image Anal."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"362","DOI":"10.1016\/j.procs.2016.05.244","article-title":"A new method of brain tissues segmentation from MRI with accuracy estimation","volume":"85","author":"Roy","year":"2016","journal-title":"Procedia Comput. Sci."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"769","DOI":"10.1080\/03772063.2017.1331757","article-title":"An iterative implementation of level set for precise segmentation of brain tissues and abnormality detection from MR images","volume":"63","author":"Roy","year":"2017","journal-title":"IETE J. Res."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"101692","DOI":"10.1016\/j.media.2020.101692","article-title":"Fully automatic brain tumor segmentation with deep learning-based selective attention using overlapping patches and multi-class weighted cross-entropy","volume":"63","author":"Akil","year":"2020","journal-title":"Med. Image Anal."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"1476","DOI":"10.1109\/JBHI.2018.2791863","article-title":"Anatomical landmark based deep feature representation for MR images in brain disease diagnosis","volume":"22","author":"Liu","year":"2018","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"73436","DOI":"10.1109\/ACCESS.2019.2920005","article-title":"Hippocampus localization using a two-stage ensemble Hough convolutional neural network","volume":"7","author":"Basher","year":"2019","journal-title":"IEEE Access"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Chegraoui, H., Philippe, C., Dangouloff-Ros, V., Grigis, A., Calmon, R., Boddaert, N., Frouin, F., Grill, J., and Frouin, V. (2021). Object Detection Improves Tumour Segmentation in MR Images of Rare Brain Tumours. Cancers, 13.","DOI":"10.3390\/cancers13236113"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"425","DOI":"10.1117\/1.1289771","article-title":"Automatic detection of foreign objects in computed radiography","volume":"5","author":"Xuan","year":"2000","journal-title":"J. Biomed. Opt."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"105539","DOI":"10.1016\/j.compbiomed.2022.105539","article-title":"A classification of MRI brain tumor based on two stage feature level ensemble of deep CNN models","volume":"146","author":"Aurna","year":"2022","journal-title":"Comput. Biol. Med."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"108105","DOI":"10.1016\/j.compeleceng.2022.108105","article-title":"A deep learning approach for brain tumor classification using MRI images","volume":"101","author":"Aamir","year":"2022","journal-title":"Comput. Electr. Eng."},{"key":"ref_26","first-page":"62","article-title":"Brain tumor segmentation in multimodal MRI via pixel-level and feature-level image fusion","volume":"16648714","author":"Liu","year":"2023","journal-title":"Multimodal Brain Image Fusion Methods Eval. Appl."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"626677","DOI":"10.3389\/fpsyt.2021.626677","article-title":"Improving individual brain age prediction using an ensemble deep learning framework","volume":"12","author":"Kuo","year":"2021","journal-title":"Front. Psychiatry"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"101660","DOI":"10.1016\/j.compmedimag.2019.101660","article-title":"Deep CNN ensembles and suggestive annotations for infant brain MRI segmentation","volume":"79","author":"Dolz","year":"2020","journal-title":"Comput. Med. Imaging Graph."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.inffus.2016.10.003","article-title":"MRI segmentation fusion for brain tumor detection","volume":"36","author":"Cabria","year":"2017","journal-title":"Inf. Fusion"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"25","DOI":"10.3389\/fncom.2020.00025","article-title":"Brain tumor segmentation using an ensemble of 3d u-nets and overall survival prediction using radiomic features","volume":"14","author":"Feng","year":"2020","journal-title":"Front. Comput. Neurosci."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"211","DOI":"10.1515\/comp-2022-0242","article-title":"Deep learning-based ensemble model for brain tumor segmentation using multi-parametric MR scans","volume":"12","author":"Das","year":"2022","journal-title":"Open Comput. Sci."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"104018","DOI":"10.1016\/j.bspc.2022.104018","article-title":"Performance enhancement of MRI-based brain tumor classification using suitable segmentation method and deep learning-based ensemble algorithm","volume":"78","author":"Tandel","year":"2022","journal-title":"Biomed. Signal Process. Control"},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Islam, M., Reza, M.T., Kaosar, M., and Parvez, M.Z. (2022). Effectiveness of Federated Learning and CNN Ensemble Architectures for Identifying Brain Tumors Using MRI Images. Neural Process. Lett., 1\u201331.","DOI":"10.1007\/s11063-022-11014-1"},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Ghafourian, E., Samadifam, F., Fadavian, H., Jerfi Canatalay, P., Tajally, A., and Channumsin, S. (2023). An Ensemble Model for the Diagnosis of Brain Tumors through MRIs. Diagnostics, 13.","DOI":"10.3390\/diagnostics13030561"},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Kang, J., Ullah, Z., and Gwak, J. (2021). MRI-based brain tumor classification using ensemble of deep features and machine learning classifiers. Sensors, 21.","DOI":"10.3390\/s21062222"},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Kurmi, A., Biswas, S., Sen, S., Sinitca, A., Kaplun, D., and Sarkar, R. (2022). An Ensemble of CNN Models for Parkinson\u2019s Disease Detection Using DaTscan Images. Diagnostics, 12.","DOI":"10.3390\/diagnostics12051173"},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Chatterjee, S., and Byun, Y.C. (2022). Voting Ensemble Approach for Enhancing Alzheimer\u2019s Disease Classification. Sensors, 22.","DOI":"10.3390\/s22197661"},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Zahoor, M.M., Qureshi, S.A., Bibi, S., Khan, S.H., Khan, A., Ghafoor, U., and Bhutta, M.R. (2022). A new deep hybrid boosted and ensemble learning-based brain tumor analysis using MRI. Sensors, 22.","DOI":"10.3390\/s22072726"},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"1993","DOI":"10.1109\/TMI.2014.2377694","article-title":"The multimodal brain tumor image segmentation benchmark (BRATS)","volume":"34","author":"Menze","year":"2014","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"55","DOI":"10.1016\/j.jalz.2005.06.003","article-title":"Ways toward an early diagnosis in Alzheimer\u2019s disease: The Alzheimer\u2019s Disease Neuroimaging Initiative (ADNI)","volume":"1","author":"Mueller","year":"2005","journal-title":"Alzheimer\u2019s Dement."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"89","DOI":"10.1016\/j.jneumeth.2014.11.011","article-title":"Extra tree forests for sub-acute ischemic stroke lesion segmentation in MR sequences","volume":"240","author":"Maier","year":"2015","journal-title":"J. Neurosci. Methods"},{"key":"ref_42","unstructured":"(2023, January 23). GaziBrais2020. Synapse. Available online: https:\/\/www.synapse.org\/#!Synapse:syn22159468."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"170117","DOI":"10.1038\/sdata.2017.117","article-title":"Advancing the cancer genome atlas glioma MRI collections with expert segmentation labels and radiomic features","volume":"4","author":"Bakas","year":"2017","journal-title":"Sci. Data"},{"key":"ref_44","unstructured":"Bakas, S., Reyes, M., Jakab, A., Bauer, S., Rempfler, M., Crimi, A., Shinohara, R.T., Berger, C., Ha, S.M., and Rozycki, M. (2018). Identifying the best machine learning algorithms for brain tumor segmentation, progression assessment, and overall survival prediction in the BRATS challenge. arXiv."},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"e6821","DOI":"10.1002\/cpe.6821","article-title":"False positive repression: Data centric pipeline for object detection in brain MRI","volume":"34","author":"Terzi","year":"2021","journal-title":"Concurr. Comput. Pract. Exp."},{"key":"ref_46","unstructured":"Sultana, F., Sufian, A., and Dutta, P. (2020). Intelligent Computing: Image Processing Based Applications, Springer."},{"key":"ref_47","unstructured":"Chen, K., Wang, J., Pang, J., Cao, Y., Xiong, Y., Li, X., Sun, S., Feng, W., Liu, Z., and Xu, J. (2019). MMDetection: Open mmlab detection toolbox and benchmark. arXiv."},{"key":"ref_48","doi-asserted-by":"crossref","unstructured":"Lin, T.Y., Goyal, P., Girshick, R., He, K., and Doll\u00e1r, P. (2017, January 22\u201329). Focal loss for dense object detection. Proceedings of the 2017 IEEE International Conference on Computer Vision (ICCV), Venice, Italy.","DOI":"10.1109\/ICCV.2017.324"},{"key":"ref_49","unstructured":"Redmon, J., and Farhadi, A. (2018). Yolov3: An incremental improvement. arXiv."},{"key":"ref_50","doi-asserted-by":"crossref","unstructured":"Tian, Z., Shen, C., Chen, H., and He, T. (November, January 27). Fcos: Fully convolutional one-stage object detection. Proceedings of the 2019 IEEE\/CVF International Conference on Computer Vision (ICCV), Seoul, Republic of Korea.","DOI":"10.1109\/ICCV.2019.00972"},{"key":"ref_51","doi-asserted-by":"crossref","unstructured":"Ghiasi, G., Lin, T.Y., and Le, Q.V. (2019, January 15\u201320). Nas-fpn: Learning scalable feature pyramid architecture for object detection. Proceedings of the 2019 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA.","DOI":"10.1109\/CVPR.2019.00720"},{"key":"ref_52","doi-asserted-by":"crossref","unstructured":"Zhang, S., Chi, C., Yao, Y., Lei, Z., and Li, S.Z. (2020, January 13\u201319). Bridging the gap between anchor-based and anchor-free detection via adaptive training sample selection. Proceedings of the 2020 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA.","DOI":"10.1109\/CVPR42600.2020.00978"},{"key":"ref_53","doi-asserted-by":"crossref","unstructured":"Zhang, H., Wang, Y., Dayoub, F., and Sunderhauf, N. (2021, January 20\u201325). Varifocalnet: An iou-aware dense object detector. Proceedings of the 2021 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA.","DOI":"10.1109\/CVPR46437.2021.00841"},{"key":"ref_54","unstructured":"Ren, S., He, K., Girshick, R., and Sun, J. (2015). Faster r-cnn: Towards real-time object detection with region proposal networks. Adv. Neural Inf. Process. Syst., 28."},{"key":"ref_55","doi-asserted-by":"crossref","unstructured":"Zhang, H., Chang, H., Ma, B., Wang, N., and Chen, X. (2020, January 23\u201328). Dynamic R-CNN: Towards high quality object detection via dynamic training. Proceedings of the Computer Vision\u2013ECCV 2020: 16th European Conference, Glasgow, UK. Proceedings, Part XV 16.","DOI":"10.1007\/978-3-030-58555-6_16"},{"key":"ref_56","doi-asserted-by":"crossref","first-page":"1483","DOI":"10.1109\/TPAMI.2019.2956516","article-title":"Cascade R-CNN: High quality object detection and instance segmentation","volume":"43","author":"Cai","year":"2019","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_57","doi-asserted-by":"crossref","first-page":"104117","DOI":"10.1016\/j.imavis.2021.104117","article-title":"Weighted boxes fusion: Ensembling boxes from different object detection models","volume":"107","author":"Solovyev","year":"2021","journal-title":"Image Vis. Comput."},{"key":"ref_58","doi-asserted-by":"crossref","unstructured":"Rezatofighi, H., Tsoi, N., Gwak, J., Sadeghian, A., Reid, I., and Savarese, S. (2019, January 15\u201320). Generalized intersection over union: A metric and a loss for bounding box regression. Proceedings of the 2019 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA.","DOI":"10.1109\/CVPR.2019.00075"},{"key":"ref_59","doi-asserted-by":"crossref","first-page":"98","DOI":"10.1007\/s11263-014-0733-5","article-title":"The pascal visual object classes challenge: A retrospective","volume":"111","author":"Everingham","year":"2015","journal-title":"Int. J. Comput. Vis."}],"container-title":["Diagnostics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-4418\/13\/8\/1494\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T19:20:10Z","timestamp":1760124010000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-4418\/13\/8\/1494"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,4,20]]},"references-count":59,"journal-issue":{"issue":"8","published-online":{"date-parts":[[2023,4]]}},"alternative-id":["diagnostics13081494"],"URL":"https:\/\/doi.org\/10.3390\/diagnostics13081494","relation":{},"ISSN":["2075-4418"],"issn-type":[{"value":"2075-4418","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,4,20]]}}}