Unsupervised Learning Techniques for Breast Lesion Segmentation on MRI Images: Are We Ready for Automation?
In the era of precision medicine, increasing importance is given to machine learning (ML) applications. In breast cancer, advanced analyses, such as the radiomic process, characterise tumours and predict therapy responses. Breast magnetic resonance imaging (MRI) plays a key role in screening, stagin...
| Main Authors: | , , , , |
|---|---|
| Formato: | Artigo |
| Idioma: | English |
| Publicado: |
MDPI AG
2025-02-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Acceso en liña: | https://www.mdpi.com/2076-3417/15/5/2401 |
| _version_ | 1826531932795568128 |
|---|---|
| author | Marina Fedon Vocaturo Luisa Altabella Giuseppe Cardano Stefania Montemezzi Carlo Cavedon |
| author_facet | Marina Fedon Vocaturo Luisa Altabella Giuseppe Cardano Stefania Montemezzi Carlo Cavedon |
| author_sort | Marina Fedon Vocaturo |
| collection | DOAJ |
| description | In the era of precision medicine, increasing importance is given to machine learning (ML) applications. In breast cancer, advanced analyses, such as the radiomic process, characterise tumours and predict therapy responses. Breast magnetic resonance imaging (MRI) plays a key role in screening, staging, and treatment monitoring. Lesion segmentation on MRI is essential both to assess tumour growth and as a baseline for radiomic feature extraction. Manual segmentation is time-consuming and prone to inter-operator variability, limiting access to large labelled datasets and robust analyses. The use of ML for breast lesion segmentation on MRI has been investigated through a systematic review of PubMed, exploring studies published over the last 10 years. Results are compared in terms of performance, primarily using the Dice score. Early unsupervised methods achieved a mean Dice score of ∼0.75, surpassing traditional supervised methods (∼0.70). In contrast, deep learning (DL) approaches based on U-Net achieved higher average scores of 0.79. Further customised supervised DL approaches reached a mean Dice score of ∼0.83. However, there is still a gap in research on unsupervised DL techniques, which could help reduce bias and human variability. Future work may also explore multiparametric and multitechnique data, integrating more representative samples, including non-mass lesions. |
| first_indexed | 2025-03-14T01:43:12Z |
| format | Article |
| id | doaj.art-c3655ae7019744fc8eff660cb60f3390 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2025-03-14T01:43:12Z |
| publishDate | 2025-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-c3655ae7019744fc8eff660cb60f33902025-03-12T13:52:54ZengMDPI AGApplied Sciences2076-34172025-02-01155240110.3390/app15052401Unsupervised Learning Techniques for Breast Lesion Segmentation on MRI Images: Are We Ready for Automation?Marina Fedon Vocaturo0Luisa Altabella1Giuseppe Cardano2Stefania Montemezzi3Carlo Cavedon4Medical Physics Unit, Department of Pathology and Diagnostics, Azienda Ospedaliera Universitaria Integrata, P.le Stefani 1, 37126 Verona, ItalyMedical Physics Unit, Department of Pathology and Diagnostics, Azienda Ospedaliera Universitaria Integrata, P.le Stefani 1, 37126 Verona, ItalyDepartment of Pathology and Diagnostics, Radiology Unit, Azienda Ospedaliera Universitaria Integrata, P.le Stefani 1, 37126 Verona, ItalyDepartment of Pathology and Diagnostics, Radiology Unit, Azienda Ospedaliera Universitaria Integrata, P.le Stefani 1, 37126 Verona, ItalyMedical Physics Unit, Department of Pathology and Diagnostics, Azienda Ospedaliera Universitaria Integrata, P.le Stefani 1, 37126 Verona, ItalyIn the era of precision medicine, increasing importance is given to machine learning (ML) applications. In breast cancer, advanced analyses, such as the radiomic process, characterise tumours and predict therapy responses. Breast magnetic resonance imaging (MRI) plays a key role in screening, staging, and treatment monitoring. Lesion segmentation on MRI is essential both to assess tumour growth and as a baseline for radiomic feature extraction. Manual segmentation is time-consuming and prone to inter-operator variability, limiting access to large labelled datasets and robust analyses. The use of ML for breast lesion segmentation on MRI has been investigated through a systematic review of PubMed, exploring studies published over the last 10 years. Results are compared in terms of performance, primarily using the Dice score. Early unsupervised methods achieved a mean Dice score of ∼0.75, surpassing traditional supervised methods (∼0.70). In contrast, deep learning (DL) approaches based on U-Net achieved higher average scores of 0.79. Further customised supervised DL approaches reached a mean Dice score of ∼0.83. However, there is still a gap in research on unsupervised DL techniques, which could help reduce bias and human variability. Future work may also explore multiparametric and multitechnique data, integrating more representative samples, including non-mass lesions.https://www.mdpi.com/2076-3417/15/5/2401machine learningdeep learningsegmentationbreast lesion segmentationbreast magnetic resonance imagingbreast cancer |
| spellingShingle | Marina Fedon Vocaturo Luisa Altabella Giuseppe Cardano Stefania Montemezzi Carlo Cavedon Unsupervised Learning Techniques for Breast Lesion Segmentation on MRI Images: Are We Ready for Automation? Applied Sciences machine learning deep learning segmentation breast lesion segmentation breast magnetic resonance imaging breast cancer |
| title | Unsupervised Learning Techniques for Breast Lesion Segmentation on MRI Images: Are We Ready for Automation? |
| title_full | Unsupervised Learning Techniques for Breast Lesion Segmentation on MRI Images: Are We Ready for Automation? |
| title_fullStr | Unsupervised Learning Techniques for Breast Lesion Segmentation on MRI Images: Are We Ready for Automation? |
| title_full_unstemmed | Unsupervised Learning Techniques for Breast Lesion Segmentation on MRI Images: Are We Ready for Automation? |
| title_short | Unsupervised Learning Techniques for Breast Lesion Segmentation on MRI Images: Are We Ready for Automation? |
| title_sort | unsupervised learning techniques for breast lesion segmentation on mri images are we ready for automation |
| topic | machine learning deep learning segmentation breast lesion segmentation breast magnetic resonance imaging breast cancer |
| url | https://www.mdpi.com/2076-3417/15/5/2401 |
| work_keys_str_mv | AT marinafedonvocaturo unsupervisedlearningtechniquesforbreastlesionsegmentationonmriimagesarewereadyforautomation AT luisaaltabella unsupervisedlearningtechniquesforbreastlesionsegmentationonmriimagesarewereadyforautomation AT giuseppecardano unsupervisedlearningtechniquesforbreastlesionsegmentationonmriimagesarewereadyforautomation AT stefaniamontemezzi unsupervisedlearningtechniquesforbreastlesionsegmentationonmriimagesarewereadyforautomation AT carlocavedon unsupervisedlearningtechniquesforbreastlesionsegmentationonmriimagesarewereadyforautomation |