Universal conditional networks (UniCoN) for multi-age embryonic cartilage segmentation with Sparsely annotated data
Abstract Osteochondrodysplasia, affecting 2–3% of newborns globally, is a group of bone and cartilage disorders that often result in head malformations, contributing to childhood morbidity and reduced quality of life. Current research on this disease using mouse models faces challenges since it invo...
Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-87797-9 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571826571575296 |
|---|---|
| author | Nishchal Sapkota Yejia Zhang Zihao Zhao Maria Jose Gomez Yuhan Hsi Jordan A. Wilson Kazuhiko Kawasaki Greg Holmes Meng Wu Ethylin Wang Jabs Joan T. Richtsmeier Susan M. Motch Perrine Danny Z. Chen |
| author_facet | Nishchal Sapkota Yejia Zhang Zihao Zhao Maria Jose Gomez Yuhan Hsi Jordan A. Wilson Kazuhiko Kawasaki Greg Holmes Meng Wu Ethylin Wang Jabs Joan T. Richtsmeier Susan M. Motch Perrine Danny Z. Chen |
| author_sort | Nishchal Sapkota |
| collection | DOAJ |
| description | Abstract Osteochondrodysplasia, affecting 2–3% of newborns globally, is a group of bone and cartilage disorders that often result in head malformations, contributing to childhood morbidity and reduced quality of life. Current research on this disease using mouse models faces challenges since it involves accurately segmenting (precisely delineating) the developing cartilage in 3D micro-CT images of embryonic mice. Tackling this segmentation task with deep learning (DL) methods is laborious due to the big burden of manual image annotation, expensive due to the high acquisition costs of 3D micro-CT images, and difficult due to embryonic cartilage’s complex and rapidly changing shapes. While DL approaches have been proposed to automate cartilage segmentation, most such models have limited accuracy and generalizability, especially across data from different embryonic age groups. To address these limitations, we propose novel DL methods that can be adopted by any DL architectures—including Convolutional Neural Networks (CNNs), Transformers, or hybrid models—which effectively leverage age and spatial information to enhance model performance. Specifically, we propose two new mechanisms, one conditioned on discrete age categories and the other on continuous image crop locations, to enable an accurate representation of cartilage shape changes across ages and local shape details throughout the cranial region. Extensive experiments on multi-age cartilage segmentation datasets show significant and consistent performance improvements when integrating our conditional modules into popular DL segmentation architectures. On average, we achieve a 1.7% Dice score increase with minimal computational overhead and a 7.5% improvement on unseen data. These results highlight the potential of our approach for developing robust, universal models capable of handling diverse datasets with limited annotated data, a key challenge in DL-based medical image analysis. |
| format | Article |
| id | doaj-art-d41a6c0104184885b08a1371d35425b4 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-d41a6c0104184885b08a1371d35425b42025-02-02T12:17:29ZengNature PortfolioScientific Reports2045-23222025-01-0115111210.1038/s41598-025-87797-9Universal conditional networks (UniCoN) for multi-age embryonic cartilage segmentation with Sparsely annotated dataNishchal Sapkota0Yejia Zhang1Zihao Zhao2Maria Jose Gomez3Yuhan Hsi4Jordan A. Wilson5Kazuhiko Kawasaki6Greg Holmes7Meng Wu8Ethylin Wang Jabs9Joan T. Richtsmeier10Susan M. Motch Perrine11Danny Z. Chen12Department of Computer Science and Engineering, University of Notre DameDepartment of Computer Science and Engineering, University of Notre DameDepartment of Computer Science and Engineering, University of Notre DameDepartment of Computer Science and Engineering, University of Notre DameDepartment of Anthropology, The Pennsylvania State UniversityDepartment of Anthropology, The Pennsylvania State UniversityDepartment of Anthropology, The Pennsylvania State UniversityDepartment of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai, Icahn Medical InstituteDepartment of Clinical Genomics, Mayo ClinicDepartment of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai, Icahn Medical InstituteDepartment of Anthropology, The Pennsylvania State UniversityDepartment of Anthropology, The Pennsylvania State UniversityDepartment of Computer Science and Engineering, University of Notre DameAbstract Osteochondrodysplasia, affecting 2–3% of newborns globally, is a group of bone and cartilage disorders that often result in head malformations, contributing to childhood morbidity and reduced quality of life. Current research on this disease using mouse models faces challenges since it involves accurately segmenting (precisely delineating) the developing cartilage in 3D micro-CT images of embryonic mice. Tackling this segmentation task with deep learning (DL) methods is laborious due to the big burden of manual image annotation, expensive due to the high acquisition costs of 3D micro-CT images, and difficult due to embryonic cartilage’s complex and rapidly changing shapes. While DL approaches have been proposed to automate cartilage segmentation, most such models have limited accuracy and generalizability, especially across data from different embryonic age groups. To address these limitations, we propose novel DL methods that can be adopted by any DL architectures—including Convolutional Neural Networks (CNNs), Transformers, or hybrid models—which effectively leverage age and spatial information to enhance model performance. Specifically, we propose two new mechanisms, one conditioned on discrete age categories and the other on continuous image crop locations, to enable an accurate representation of cartilage shape changes across ages and local shape details throughout the cranial region. Extensive experiments on multi-age cartilage segmentation datasets show significant and consistent performance improvements when integrating our conditional modules into popular DL segmentation architectures. On average, we achieve a 1.7% Dice score increase with minimal computational overhead and a 7.5% improvement on unseen data. These results highlight the potential of our approach for developing robust, universal models capable of handling diverse datasets with limited annotated data, a key challenge in DL-based medical image analysis.https://doi.org/10.1038/s41598-025-87797-9Embryonic cartilage segmentationMulti-age image dataMicro-CTConditional trainingSelf-attention |
| spellingShingle | Nishchal Sapkota Yejia Zhang Zihao Zhao Maria Jose Gomez Yuhan Hsi Jordan A. Wilson Kazuhiko Kawasaki Greg Holmes Meng Wu Ethylin Wang Jabs Joan T. Richtsmeier Susan M. Motch Perrine Danny Z. Chen Universal conditional networks (UniCoN) for multi-age embryonic cartilage segmentation with Sparsely annotated data Scientific Reports Embryonic cartilage segmentation Multi-age image data Micro-CT Conditional training Self-attention |
| title | Universal conditional networks (UniCoN) for multi-age embryonic cartilage segmentation with Sparsely annotated data |
| title_full | Universal conditional networks (UniCoN) for multi-age embryonic cartilage segmentation with Sparsely annotated data |
| title_fullStr | Universal conditional networks (UniCoN) for multi-age embryonic cartilage segmentation with Sparsely annotated data |
| title_full_unstemmed | Universal conditional networks (UniCoN) for multi-age embryonic cartilage segmentation with Sparsely annotated data |
| title_short | Universal conditional networks (UniCoN) for multi-age embryonic cartilage segmentation with Sparsely annotated data |
| title_sort | universal conditional networks unicon for multi age embryonic cartilage segmentation with sparsely annotated data |
| topic | Embryonic cartilage segmentation Multi-age image data Micro-CT Conditional training Self-attention |
| url | https://doi.org/10.1038/s41598-025-87797-9 |
| work_keys_str_mv | AT nishchalsapkota universalconditionalnetworksuniconformultiageembryoniccartilagesegmentationwithsparselyannotateddata AT yejiazhang universalconditionalnetworksuniconformultiageembryoniccartilagesegmentationwithsparselyannotateddata AT zihaozhao universalconditionalnetworksuniconformultiageembryoniccartilagesegmentationwithsparselyannotateddata AT mariajosegomez universalconditionalnetworksuniconformultiageembryoniccartilagesegmentationwithsparselyannotateddata AT yuhanhsi universalconditionalnetworksuniconformultiageembryoniccartilagesegmentationwithsparselyannotateddata AT jordanawilson universalconditionalnetworksuniconformultiageembryoniccartilagesegmentationwithsparselyannotateddata AT kazuhikokawasaki universalconditionalnetworksuniconformultiageembryoniccartilagesegmentationwithsparselyannotateddata AT gregholmes universalconditionalnetworksuniconformultiageembryoniccartilagesegmentationwithsparselyannotateddata AT mengwu universalconditionalnetworksuniconformultiageembryoniccartilagesegmentationwithsparselyannotateddata AT ethylinwangjabs universalconditionalnetworksuniconformultiageembryoniccartilagesegmentationwithsparselyannotateddata AT joantrichtsmeier universalconditionalnetworksuniconformultiageembryoniccartilagesegmentationwithsparselyannotateddata AT susanmmotchperrine universalconditionalnetworksuniconformultiageembryoniccartilagesegmentationwithsparselyannotateddata AT dannyzchen universalconditionalnetworksuniconformultiageembryoniccartilagesegmentationwithsparselyannotateddata |