Please use this identifier to cite or link to this item:
10.3390/tomography9050141
Title: | Modular Neural Networks for Osteoporosis Detection in Mandibular Cone-Beam Computed Tomography Scans |
Authors: | Namatevs, Ivars Nikulins, Arturs Edelmers, Edgars Neimane, Laura Slaidiņa, Anda Radziņš, Oskars Sudars, Kaspars Department of Morphology Institute of Anatomy and Anthropology Department of Conservative Dentistry and Oral Health Department of Prosthetic Dentistry Department of Orthodontics |
Keywords: | artificial intelligence;CBCT;convolutional neural network;dentistry;osteoporosis;deep learning;3.2 Clinical medicine;2.6 Medical engineering;1.1. Scientific article indexed in Web of Science and/or Scopus database |
Issue Date: | Oct-2023 |
Citation: | Namatevs , I , Nikulins , A , Edelmers , E , Neimane , L , Slaidiņa , A , Radziņš , O & Sudars , K 2023 , ' Modular Neural Networks for Osteoporosis Detection in Mandibular Cone-Beam Computed Tomography Scans ' , Tomography (Ann Arbor, Mich.) , vol. 9 , no. 5 , pp. 1772-1786 . https://doi.org/10.3390/tomography9050141 |
Abstract: | In this technical note, we examine the capabilities of deep convolutional neural networks (DCNNs) for diagnosing osteoporosis through cone-beam computed tomography (CBCT) scans of the mandible. The evaluation was conducted using 188 patients’ mandibular CBCT images utilizing DCNN models built on the ResNet-101 framework. We adopted a segmented three-phase method to assess osteoporosis. Stage 1 focused on mandibular bone slice identification, Stage 2 pinpointed the coordinates for mandibular bone cross-sectional views, and Stage 3 computed the mandibular bone’s thickness, highlighting osteoporotic variances. The procedure, built using ResNet-101 networks, showcased efficacy in osteoporosis detection using CBCT scans: Stage 1 achieved a remarkable 98.85% training accuracy, Stage 2 minimized L1 loss to a mere 1.02 pixels, and the last stage’s bone thickness computation algorithm reported a mean squared error of 0.8377. These findings underline the significant potential of AI in osteoporosis identification and its promise for enhanced medical care. The compartmentalized method endorses a sturdier DCNN training and heightened model transparency. Moreover, the outcomes illustrate the efficacy of a modular transfer learning method for osteoporosis detection, even when relying on limited mandibular CBCT datasets. The methodology given is accompanied by the source code available on GitLab. |
Description: | Publisher Copyright: © 2023 by the authors. |
DOI: | 10.3390/tomography9050141 |
ISSN: | 2379-1381 |
Appears in Collections: | Research outputs from Pure / Zinātniskās darbības rezultāti no ZDIS Pure |
Files in This Item:
File | Size | Format | |
---|---|---|---|
Modular_Neural_Networks_for_Osteoporosis_Detection.pdf | 5.51 MB | Adobe PDF | View/Open![]() |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.