Microwave Breast Sensing via Deep Learning for Tumor Spatial Localization by Probability Maps
Background: microwave imaging (MWI) has emerged as a promising modality for breast cancer screening, offering cost-effective, rapid, safe and comfortable exams. However, the practical application of MWI for tumor detection and localization is hampered by its inherent low resolution and low detection...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-10-01
|
| Series: | Bioengineering |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2306-5354/10/10/1153 |
| _version_ | 1797574693380685824 |
|---|---|
| author | Marijn Borghouts Michele Ambrosanio Stefano Franceschini Maria Maddalena Autorino Vito Pascazio Fabio Baselice |
| author_facet | Marijn Borghouts Michele Ambrosanio Stefano Franceschini Maria Maddalena Autorino Vito Pascazio Fabio Baselice |
| author_sort | Marijn Borghouts |
| collection | DOAJ |
| description | Background: microwave imaging (MWI) has emerged as a promising modality for breast cancer screening, offering cost-effective, rapid, safe and comfortable exams. However, the practical application of MWI for tumor detection and localization is hampered by its inherent low resolution and low detection capability. Methods: this study aims to generate an accurate tumor probability map directly from the scattering matrix. This direct conversion makes the probability map independent of specific image formation techniques and thus potentially complementary to any image formation technique. An approach based on a convolutional neural network (CNN) is used to convert the scattering matrix into a tumor probability map. The proposed deep learning model is trained using a large realistic numerical dataset of two-dimensional (2D) breast slices. The performance of the model is assessed through visual inspection and quantitative measures to assess the predictive quality at various levels of detail. Results: the results demonstrate a remarkably high accuracy (0.9995) in classifying profiles as healthy or diseased, and exhibit the model’s ability to accurately locate the core of a single tumor (within 0.9 cm for most cases). Conclusion: overall, this research demonstrates that an approach based on neural networks (NN) for direct conversion from scattering matrices to tumor probability maps holds promise in advancing state-of-the-art tumor detection algorithms in the MWI domain. |
| first_indexed | 2024-03-10T21:26:52Z |
| format | Article |
| id | doaj.art-ddc5c738230948299374ca9cc4720641 |
| institution | Directory Open Access Journal |
| issn | 2306-5354 |
| language | English |
| last_indexed | 2024-03-10T21:26:52Z |
| publishDate | 2023-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Bioengineering |
| spelling | doaj.art-ddc5c738230948299374ca9cc47206412023-11-19T15:41:43ZengMDPI AGBioengineering2306-53542023-10-011010115310.3390/bioengineering10101153Microwave Breast Sensing via Deep Learning for Tumor Spatial Localization by Probability MapsMarijn Borghouts0Michele Ambrosanio1Stefano Franceschini2Maria Maddalena Autorino3Vito Pascazio4Fabio Baselice5Department of Biomedical Engineering, Technical University of Eindhoven, 5600 MB Eindhoven, The NetherlandsDepartment of Economics, Law, Cybersecurity and Sports Sciences, University of Naples “Parthenope”, 80035 Nola, ItalyDepartment of Engineering, University of Napoli “Parthenope”, 80143 Naples, ItalyDepartment of Engineering, University of Napoli “Parthenope”, 80143 Naples, ItalyDepartment of Engineering, University of Napoli “Parthenope”, 80143 Naples, ItalyDepartment of Engineering, University of Napoli “Parthenope”, 80143 Naples, ItalyBackground: microwave imaging (MWI) has emerged as a promising modality for breast cancer screening, offering cost-effective, rapid, safe and comfortable exams. However, the practical application of MWI for tumor detection and localization is hampered by its inherent low resolution and low detection capability. Methods: this study aims to generate an accurate tumor probability map directly from the scattering matrix. This direct conversion makes the probability map independent of specific image formation techniques and thus potentially complementary to any image formation technique. An approach based on a convolutional neural network (CNN) is used to convert the scattering matrix into a tumor probability map. The proposed deep learning model is trained using a large realistic numerical dataset of two-dimensional (2D) breast slices. The performance of the model is assessed through visual inspection and quantitative measures to assess the predictive quality at various levels of detail. Results: the results demonstrate a remarkably high accuracy (0.9995) in classifying profiles as healthy or diseased, and exhibit the model’s ability to accurately locate the core of a single tumor (within 0.9 cm for most cases). Conclusion: overall, this research demonstrates that an approach based on neural networks (NN) for direct conversion from scattering matrices to tumor probability maps holds promise in advancing state-of-the-art tumor detection algorithms in the MWI domain.https://www.mdpi.com/2306-5354/10/10/1153microwave imagingneural networkstumor localizationbreast cancerearly detectionbiomedical engineering |
| spellingShingle | Marijn Borghouts Michele Ambrosanio Stefano Franceschini Maria Maddalena Autorino Vito Pascazio Fabio Baselice Microwave Breast Sensing via Deep Learning for Tumor Spatial Localization by Probability Maps Bioengineering microwave imaging neural networks tumor localization breast cancer early detection biomedical engineering |
| title | Microwave Breast Sensing via Deep Learning for Tumor Spatial Localization by Probability Maps |
| title_full | Microwave Breast Sensing via Deep Learning for Tumor Spatial Localization by Probability Maps |
| title_fullStr | Microwave Breast Sensing via Deep Learning for Tumor Spatial Localization by Probability Maps |
| title_full_unstemmed | Microwave Breast Sensing via Deep Learning for Tumor Spatial Localization by Probability Maps |
| title_short | Microwave Breast Sensing via Deep Learning for Tumor Spatial Localization by Probability Maps |
| title_sort | microwave breast sensing via deep learning for tumor spatial localization by probability maps |
| topic | microwave imaging neural networks tumor localization breast cancer early detection biomedical engineering |
| url | https://www.mdpi.com/2306-5354/10/10/1153 |
| work_keys_str_mv | AT marijnborghouts microwavebreastsensingviadeeplearningfortumorspatiallocalizationbyprobabilitymaps AT micheleambrosanio microwavebreastsensingviadeeplearningfortumorspatiallocalizationbyprobabilitymaps AT stefanofranceschini microwavebreastsensingviadeeplearningfortumorspatiallocalizationbyprobabilitymaps AT mariamaddalenaautorino microwavebreastsensingviadeeplearningfortumorspatiallocalizationbyprobabilitymaps AT vitopascazio microwavebreastsensingviadeeplearningfortumorspatiallocalizationbyprobabilitymaps AT fabiobaselice microwavebreastsensingviadeeplearningfortumorspatiallocalizationbyprobabilitymaps |