AI-Assisted Detection and Localization of Spinal Metastatic Lesions
Objectives: The integration of machine learning and radiomics in medical imaging has significantly advanced diagnostic and prognostic capabilities in healthcare. This study focuses on developing and validating an artificial intelligence (AI) model using U-Net architectures for the accurate detection...
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-11-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/14/21/2458 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850062382333689856 |
|---|---|
| author | Edgars Edelmers Artūrs Ņikuļins Klinta Luīze Sprūdža Patrīcija Stapulone Niks Saimons Pūce Elizabete Skrebele Everita Elīna Siņicina Viktorija Cīrule Ance Kazuša Katrina Boločko |
| author_facet | Edgars Edelmers Artūrs Ņikuļins Klinta Luīze Sprūdža Patrīcija Stapulone Niks Saimons Pūce Elizabete Skrebele Everita Elīna Siņicina Viktorija Cīrule Ance Kazuša Katrina Boločko |
| author_sort | Edgars Edelmers |
| collection | DOAJ |
| description | Objectives: The integration of machine learning and radiomics in medical imaging has significantly advanced diagnostic and prognostic capabilities in healthcare. This study focuses on developing and validating an artificial intelligence (AI) model using U-Net architectures for the accurate detection and segmentation of spinal metastases from computed tomography (CT) images, addressing both osteolytic and osteoblastic lesions. Methods: Our methodology employs multiple variations of the U-Net architecture and utilizes two distinct datasets: one consisting of 115 polytrauma patients for vertebra segmentation and another comprising 38 patients with documented spinal metastases for lesion detection. Results: The model demonstrated strong performance in vertebra segmentation, achieving Dice Similarity Coefficient (DSC) values between 0.87 and 0.96. For metastasis segmentation, the model achieved a DSC of 0.71 and an F-beta score of 0.68 for lytic lesions but struggled with sclerotic lesions, obtaining a DSC of 0.61 and an F-beta score of 0.57, reflecting challenges in detecting dense, subtle bone alterations. Despite these limitations, the model successfully identified isolated metastatic lesions beyond the spine, such as in the sternum, indicating potential for broader skeletal metastasis detection. Conclusions: The study concludes that AI-based models can augment radiologists’ capabilities by providing reliable second-opinion tools, though further refinements and diverse training data are needed for optimal performance, particularly for sclerotic lesion segmentation. The annotated CT dataset produced and shared in this research serves as a valuable resource for future advancements. |
| format | Article |
| id | doaj-art-77c4901d81e64ddfa208a3b974f60705 |
| institution | DOAJ |
| issn | 2075-4418 |
| language | English |
| publishDate | 2024-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj-art-77c4901d81e64ddfa208a3b974f607052025-08-20T02:49:56ZengMDPI AGDiagnostics2075-44182024-11-011421245810.3390/diagnostics14212458AI-Assisted Detection and Localization of Spinal Metastatic LesionsEdgars Edelmers0Artūrs Ņikuļins1Klinta Luīze Sprūdža2Patrīcija Stapulone3Niks Saimons Pūce4Elizabete Skrebele5Everita Elīna Siņicina6Viktorija Cīrule7Ance Kazuša8Katrina Boločko9Faculty of Medicine, Rīga Stradiņš University, LV-1010 Riga, LatviaFaculty of Computer Science, Information Technology and Energy, Riga Technical University, LV-1048 Riga, LatviaFaculty of Medicine, Rīga Stradiņš University, LV-1010 Riga, LatviaFaculty of Medicine, Rīga Stradiņš University, LV-1010 Riga, LatviaFaculty of Computer Science, Information Technology and Energy, Riga Technical University, LV-1048 Riga, LatviaFaculty of Civil and Mechanical Engineering, Riga Technical University, LV-1048 Riga, LatviaFaculty of Biology, University of Latvia, LV-1004 Riga, LatviaDepartment of Radiology, Faculty of Medicine, Rīga Stradiņš University, LV-1010 Riga, LatviaFaculty of Medicine, Rīga Stradiņš University, LV-1010 Riga, LatviaDepartment of Computer Graphics and Computer Vision, Riga Technical University, LV-1048 Riga, LatviaObjectives: The integration of machine learning and radiomics in medical imaging has significantly advanced diagnostic and prognostic capabilities in healthcare. This study focuses on developing and validating an artificial intelligence (AI) model using U-Net architectures for the accurate detection and segmentation of spinal metastases from computed tomography (CT) images, addressing both osteolytic and osteoblastic lesions. Methods: Our methodology employs multiple variations of the U-Net architecture and utilizes two distinct datasets: one consisting of 115 polytrauma patients for vertebra segmentation and another comprising 38 patients with documented spinal metastases for lesion detection. Results: The model demonstrated strong performance in vertebra segmentation, achieving Dice Similarity Coefficient (DSC) values between 0.87 and 0.96. For metastasis segmentation, the model achieved a DSC of 0.71 and an F-beta score of 0.68 for lytic lesions but struggled with sclerotic lesions, obtaining a DSC of 0.61 and an F-beta score of 0.57, reflecting challenges in detecting dense, subtle bone alterations. Despite these limitations, the model successfully identified isolated metastatic lesions beyond the spine, such as in the sternum, indicating potential for broader skeletal metastasis detection. Conclusions: The study concludes that AI-based models can augment radiologists’ capabilities by providing reliable second-opinion tools, though further refinements and diverse training data are needed for optimal performance, particularly for sclerotic lesion segmentation. The annotated CT dataset produced and shared in this research serves as a valuable resource for future advancements.https://www.mdpi.com/2075-4418/14/21/2458artificial intelligencespinal metastasesvertebrae segmentationcomputer tomographymedical imaginginstance segmentation |
| spellingShingle | Edgars Edelmers Artūrs Ņikuļins Klinta Luīze Sprūdža Patrīcija Stapulone Niks Saimons Pūce Elizabete Skrebele Everita Elīna Siņicina Viktorija Cīrule Ance Kazuša Katrina Boločko AI-Assisted Detection and Localization of Spinal Metastatic Lesions Diagnostics artificial intelligence spinal metastases vertebrae segmentation computer tomography medical imaging instance segmentation |
| title | AI-Assisted Detection and Localization of Spinal Metastatic Lesions |
| title_full | AI-Assisted Detection and Localization of Spinal Metastatic Lesions |
| title_fullStr | AI-Assisted Detection and Localization of Spinal Metastatic Lesions |
| title_full_unstemmed | AI-Assisted Detection and Localization of Spinal Metastatic Lesions |
| title_short | AI-Assisted Detection and Localization of Spinal Metastatic Lesions |
| title_sort | ai assisted detection and localization of spinal metastatic lesions |
| topic | artificial intelligence spinal metastases vertebrae segmentation computer tomography medical imaging instance segmentation |
| url | https://www.mdpi.com/2075-4418/14/21/2458 |
| work_keys_str_mv | AT edgarsedelmers aiassisteddetectionandlocalizationofspinalmetastaticlesions AT artursnikulins aiassisteddetectionandlocalizationofspinalmetastaticlesions AT klintaluizesprudza aiassisteddetectionandlocalizationofspinalmetastaticlesions AT patricijastapulone aiassisteddetectionandlocalizationofspinalmetastaticlesions AT nikssaimonspuce aiassisteddetectionandlocalizationofspinalmetastaticlesions AT elizabeteskrebele aiassisteddetectionandlocalizationofspinalmetastaticlesions AT everitaelinasinicina aiassisteddetectionandlocalizationofspinalmetastaticlesions AT viktorijacirule aiassisteddetectionandlocalizationofspinalmetastaticlesions AT ancekazusa aiassisteddetectionandlocalizationofspinalmetastaticlesions AT katrinabolocko aiassisteddetectionandlocalizationofspinalmetastaticlesions |