ChronoMID-Cross-modal neural networks for 3-D temporal medical imaging data.
ChronoMID-neural networks for temporally-varying, hence Chrono, Medical Imaging Data-makes the novel application of cross-modal convolutional neural networks (X-CNNs) to the medical domain. In this paper, we present multiple approaches for incorporating temporal information into X-CNNs and compare t...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2020-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://doi.org/10.1371/journal.pone.0228962 |
| _version_ | 1819199433542205440 |
|---|---|
| author | Alexander G Rakowski Petar Veličković Enrico Dall'Ara Pietro Liò |
| author_facet | Alexander G Rakowski Petar Veličković Enrico Dall'Ara Pietro Liò |
| author_sort | Alexander G Rakowski |
| collection | DOAJ |
| description | ChronoMID-neural networks for temporally-varying, hence Chrono, Medical Imaging Data-makes the novel application of cross-modal convolutional neural networks (X-CNNs) to the medical domain. In this paper, we present multiple approaches for incorporating temporal information into X-CNNs and compare their performance in a case study on the classification of abnormal bone remodelling in mice. Previous work developing medical models has predominantly focused on either spatial or temporal aspects, but rarely both. Our models seek to unify these complementary sources of information and derive insights in a bottom-up, data-driven approach. As with many medical datasets, the case study herein exhibits deep rather than wide data; we apply various techniques, including extensive regularisation, to account for this. After training on a balanced set of approximately 70000 images, two of the models-those using difference maps from known reference points-outperformed a state-of-the-art convolutional neural network baseline by over 30pp (> 99% vs. 68.26%) on an unseen, balanced validation set comprising around 20000 images. These models are expected to perform well with sparse data sets based on both previous findings with X-CNNs and the representations of time used, which permit arbitrarily large and irregular gaps between data points. Our results highlight the importance of identifying a suitable description of time for a problem domain, as unsuitable descriptors may not only fail to improve a model, they may in fact confound it. |
| first_indexed | 2024-12-23T03:16:16Z |
| format | Article |
| id | doaj.art-8328a8c9a59a4a87823f0a2b47c0334a |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2024-12-23T03:16:16Z |
| publishDate | 2020-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-8328a8c9a59a4a87823f0a2b47c0334a2022-12-21T18:02:05ZengPublic Library of Science (PLoS)PLoS ONE1932-62032020-01-01152e022896210.1371/journal.pone.0228962ChronoMID-Cross-modal neural networks for 3-D temporal medical imaging data.Alexander G RakowskiPetar VeličkovićEnrico Dall'AraPietro LiòChronoMID-neural networks for temporally-varying, hence Chrono, Medical Imaging Data-makes the novel application of cross-modal convolutional neural networks (X-CNNs) to the medical domain. In this paper, we present multiple approaches for incorporating temporal information into X-CNNs and compare their performance in a case study on the classification of abnormal bone remodelling in mice. Previous work developing medical models has predominantly focused on either spatial or temporal aspects, but rarely both. Our models seek to unify these complementary sources of information and derive insights in a bottom-up, data-driven approach. As with many medical datasets, the case study herein exhibits deep rather than wide data; we apply various techniques, including extensive regularisation, to account for this. After training on a balanced set of approximately 70000 images, two of the models-those using difference maps from known reference points-outperformed a state-of-the-art convolutional neural network baseline by over 30pp (> 99% vs. 68.26%) on an unseen, balanced validation set comprising around 20000 images. These models are expected to perform well with sparse data sets based on both previous findings with X-CNNs and the representations of time used, which permit arbitrarily large and irregular gaps between data points. Our results highlight the importance of identifying a suitable description of time for a problem domain, as unsuitable descriptors may not only fail to improve a model, they may in fact confound it.https://doi.org/10.1371/journal.pone.0228962 |
| spellingShingle | Alexander G Rakowski Petar Veličković Enrico Dall'Ara Pietro Liò ChronoMID-Cross-modal neural networks for 3-D temporal medical imaging data. PLoS ONE |
| title | ChronoMID-Cross-modal neural networks for 3-D temporal medical imaging data. |
| title_full | ChronoMID-Cross-modal neural networks for 3-D temporal medical imaging data. |
| title_fullStr | ChronoMID-Cross-modal neural networks for 3-D temporal medical imaging data. |
| title_full_unstemmed | ChronoMID-Cross-modal neural networks for 3-D temporal medical imaging data. |
| title_short | ChronoMID-Cross-modal neural networks for 3-D temporal medical imaging data. |
| title_sort | chronomid cross modal neural networks for 3 d temporal medical imaging data |
| url | https://doi.org/10.1371/journal.pone.0228962 |
| work_keys_str_mv | AT alexandergrakowski chronomidcrossmodalneuralnetworksfor3dtemporalmedicalimagingdata AT petarvelickovic chronomidcrossmodalneuralnetworksfor3dtemporalmedicalimagingdata AT enricodallara chronomidcrossmodalneuralnetworksfor3dtemporalmedicalimagingdata AT pietrolio chronomidcrossmodalneuralnetworksfor3dtemporalmedicalimagingdata |