Spotting L3 slice in CT scans using deep convolutional network and transfer learning
Abstract
In this article, we present a complete automated system for spotting a particular slice in a complete 3D Computed Tomography exam (CT scan). Our approach does not require any assumptions on which part of the patient's body is covered by the scan. It relies on an original machine learning regression approach. Our models are learned using the transfer learning trick by exploiting deep architectures that have been pre-trained on imageNet database, and therefore it requires very little annotation for its training. The whole pipeline consists of three steps: i) conversion of the CT scans into Maximum Intensity Projection (MIP) images, ii) prediction from a Convolutional Neural Network (CNN) applied in a sliding window fashion over the MIP image, and iii) robust analysis of the prediction sequence to predict the height of the desired slice within the whole CT scan. Our approach is applied to the detection of the third lumbar vertebra (L3) slice that has been found to be representative to the whole body composition. Our system is evaluated on a database collected in our clinical center, containing 642 CT scans from different patients. We obtained an average localization error of 1.91±2.69 slices (less than 5 mm) in an average time of less than 2.5 s/CT scan, allowing integration of the proposed system into daily clinical routines.
Related
- Learning without relearning physical symmetry: an SO(3)-equivariant convolutional network (CNN) on the sphere
- Estimating "Realized" Skewness using Convolutional Neural Network
- Computations of the slice genus and the unknotting number of links via machine learning
- Stochastic dynamics generation of directed percolation using a two-channel convolutional neural network
- Aligning entity of knowledge graph based on graph convolutional network with maximum translated similarity
BibTeX
@article{belharbi2026spotting,
title = {{Spotting L3 slice in CT scans using deep convolutional network and transfer learning}},
author = {Soufiane Belharbi and Clément Chatelain and Romain Hérault and Sébastien Adam and Sébastien Thureau and Mathieu Chastan and Romain Modzelewski},
journal = {Computers in Biology and Medicine},
year = {2026},
month = oct,
doi = {10.1016/j.compbiomed.2017.05.018},
url = {https://hal.science/hal-01643960v1},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Spotting L3 slice in CT scans using deep convolutional network and transfer learning
AU - Soufiane Belharbi
AU - Clément Chatelain
AU - Romain Hérault
AU - Sébastien Adam
AU - Sébastien Thureau
AU - Mathieu Chastan
AU - Romain Modzelewski
PY - 2026
DA - 2026/10/05
JO - Computers in Biology and Medicine
DO - 10.1016/j.compbiomed.2017.05.018
UR - https://hal.science/hal-01643960v1
AB - In this article, we present a complete automated system for spotting a particular slice in a complete 3D Computed Tomography exam (CT scan). Our approach does not require any assumptions on which part of the patient's body is covered by the scan. It relies on an original machine learning regression approach. Our models are learned using the transfer learning trick by exploiting deep architectures that have been pre-trained on imageNet database, and therefore it requires very little annotation for its training. The whole pipeline consists of three steps: i) conversion of the CT scans into Maximum Intensity Projection (MIP) images, ii) prediction from a Convolutional Neural Network (CNN) applied in a sliding window fashion over the MIP image, and iii) robust analysis of the prediction sequence to predict the height of the desired slice within the whole CT scan. Our approach is applied to the detection of the third lumbar vertebra (L3) slice that has been found to be representative to the whole body composition. Our system is evaluated on a database collected in our clinical center, containing 642 CT scans from different patients. We obtained an average localization error of 1.91±2.69 slices (less than 5 mm) in an average time of less than 2.5 s/CT scan, allowing integration of the proposed system into daily clinical routines.
SN - 0010-4825
ER -