Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
149 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

SYNCS: Synthetic Data and Contrastive Self-Supervised Training for Central Sulcus Segmentation (2403.15121v1)

Published 22 Mar 2024 in cs.CV

Abstract: Bipolar disorder (BD) and schizophrenia (SZ) are severe mental disorders with profound societal impact. Identifying risk markers early is crucial for understanding disease progression and enabling preventive measures. The Danish High Risk and Resilience Study (VIA) focuses on understanding early disease processes, particularly in children with familial high risk (FHR). Understanding structural brain changes associated with these diseases during early stages is essential for effective interventions. The central sulcus (CS) is a prominent brain landmark related to brain regions involved in motor and sensory processing. Analyzing CS morphology can provide valuable insights into neurodevelopmental abnormalities in the FHR group. However, segmenting the central sulcus (CS) presents challenges due to its variability, especially in adolescents. This study introduces two novel approaches to improve CS segmentation: synthetic data generation to model CS variability and self-supervised pre-training with multi-task learning to adapt models to new cohorts. These methods aim to enhance segmentation performance across diverse populations, eliminating the need for extensive preprocessing.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (61)
  1. Big self-supervised models advance medical image classification, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 3478–3488.
  2. Automatic classification of sulcal regions of the human brain cortex using pattern recognition, in: Medical imaging 2003: Image processing, SPIE. pp. 1499–1510.
  3. A learning strategy for contrast-agnostic mri segmentation. arXiv preprint arXiv:2003.01995 .
  4. Synthseg: Segmentation of brain MRI scans of any contrast and resolution without retraining. Medical Image Analysis 86, 102789. doi:10.1016/j.media.2023.102789.
  5. Robust machine learning segmentation for large-scale analysis of heterogeneous clinical brain mri datasets. Proceedings of the National Academy of Sciences 120, e2216399120. doi:10.1073/pnas.2216399120.
  6. Automatic labeling of cortical sulci using patch- or CNN-based segmentation techniques combined with bottom-up geometric constraints. Medical Image Analysis 62, 101651. doi:10.1016/j.media.2020.101651.
  7. Brainvisa, 2019. Sulci database. Online.
  8. Impaired motor development in children with familial high risk of schizophrenia or bipolar disorder and the association with psychotic experiences: a 4-year danish observational follow-up study. The Lancet Psychiatry 10, 108–118. doi:10.1016/s2215-0366(22)00402-3.
  9. Impairments of motor function among children with a familial risk of schizophrenia or bipolar disorder at 7 years old in denmark: an observational cohort study. The Lancet Psychiatry 4, 400–408. doi:10.1016/s2215-0366(17)30103-7.
  10. Monai: An open-source framework for deep learning in healthcare.
  11. New morphologic variants of the hand motor cortex as seen with MR imaging in a large study population. American Journal of Neuroradiology 28, 1480–1485. doi:10.3174/ajnr.a0597.
  12. A simple framework for contrastive learning of visual representations, in: International conference on machine learning, PMLR. pp. 1597–1607.
  13. A review of medical image data augmentation techniques for deep learning applications. Journal of Medical Imaging and Radiation Oncology 65, 545–563. doi:https://doi.org/10.1111/1754-9485.13261.
  14. 3d u-net: learning dense volumetric segmentation from sparse annotation, in: Medical Image Computing and Computer-Assisted Intervention–MICCAI 2016: 19th International Conference, Athens, Greece, October 17-21, 2016, Proceedings, Part II 19, Springer. pp. 424–432.
  15. Identification of the central sulcus using the scanner and mri. Journal of Neuroradiology= Journal de Neuroradiologie 24, 187–204.
  16. Automatic 3d intersubject registration of mr volumetric data in standardized talairach space. Journal of computer assisted tomography 18, 192–205.
  17. An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. NeuroImage 31, 968–980. doi:10.1016/j.neuroimage.2006.01.021.
  18. Improving data-efficiency and robustness of medical imaging segmentation using inpainting-based self-supervised learning. Bioengineering 10. doi:10.3390/bioengineering10020207.
  19. The prevalence and burden of bipolar disorder: findings from the global burden of disease study 2013. Bipolar Disorders 18, 440–450. doi:10.1111/bdi.12423.
  20. Cortical surface-based analysis: Ii: Inflation, flattening, and a surface-based coordinate system. NeuroImage 9, 195 – 207.
  21. A feature transfer enabled multi-task deep learning model on medical imaging. Expert Systems with Applications 143, 112957. doi:https://doi.org/10.1016/j.eswa.2019.112957.
  22. Deep learning techniques for medical image segmentation: Achievements and challenges. Journal of Digital Imaging 32, 582–596. doi:10.1007/s10278-019-00227-x.
  23. Self-supervised learning for medical image classification: a systematic review and implementation guidelines. npj Digital Medicine 6. doi:10.1038/s41746-023-00811-0.
  24. Morphological patterns of the collateral sulcus in the human brain. European Journal of Neuroscience 35, 1295–1311. doi:https://doi.org/10.1111/j.1460-9568.2012.08031.x.
  25. Joint super-resolution and synthesis of 1 mm isotropic mp-rage volumes from clinical mri exams with scans of different orientation, resolution and contrast. Neuroimage 237, 118206.
  26. A survey on contrastive self-supervised learning. Technologies 9, 2.
  27. Influence of Maturation, Pathology and Functional Lateralization on 3D Sulcal Morphology using MRI. Ph.D. thesis. Technical University of Denmark, DTU Compute.
  28. A geometric method for automatic extraction of sulcal fundi. IEEE transactions on medical imaging 26, 530–540.
  29. Adolescent brain maturation and cortical folding: Evidence for reductions in gyrification. PLoS ONE 9, e84914. doi:10.1371/journal.pone.0084914.
  30. Age-related morphology trends of cortical sulci. Human Brain Mapping 26, 210–220. URL: https://doi.org/10.1002/hbm.20198, doi:10.1002/hbm.20198.
  31. A library of cortical morphology analysis tools to study development, aging and genetics of cerebral cortex. Neuroinformatics 10, 81–96. doi:10.1007/s12021-011-9127-9.
  32. New human-specific brain landmark: The depth asymmetry of superior temporal sulcus. Proceedings of the National Academy of Sciences 112, 1208–1213. doi:10.1073/pnas.1412389112.
  33. A closer look at loss weighting in multi-task learning. arXiv preprint arXiv:2111.10603 .
  34. Differential patterns of gyral and sulcal morphological changes during normal aging process. Frontiers in Aging Neuroscience 13. doi:10.3389/fnagi.2021.625931.
  35. A review of deep-learning-based medical image segmentation methods. Sustainability 13, 1224. doi:10.3390/su13031224.
  36. Labeling lateral prefrontal sulci using spherical data augmentation and context-aware training. NeuroImage 229, 117758. doi:https://doi.org/10.1016/j.neuroimage.2021.117758.
  37. Visualizing data using t-sne. Journal of Machine Learning Research 9, 2579–2605. URL: http://jmlr.org/papers/v9/vandermaaten08a.html.
  38. From 3d magnetic resonance images to structural representations of the cortex topography using topology preserving deformations. Journal of Mathematical Imaging and Vision 5, 297–318. doi:10.1007/bf01250286.
  39. Constructing the cerebral cortex: neurogenesis and fate determination. Neuron 15, 761–768.
  40. Humanistic burden in schizophrenia: A literature review. Journal of Psychiatric Research 54, 85–93. doi:10.1016/j.jpsychires.2014.03.021.
  41. Sulcal pattern and morphology of the superior temporal sulcus. NeuroImage 22, 706–719. doi:https://doi.org/10.1016/j.neuroimage.2004.01.023.
  42. Representation learning with contrastive predictive coding. arXiv preprint arXiv:1807.03748 .
  43. Cortical sulci recognition and spatial normalization. Medical Image Analysis 15, 529–550. doi:https://doi.org/10.1016/j.media.2011.02.008. special section on IPMI 2009.
  44. Fast and sequence-adaptive whole-brain segmentation using parametric bayesian modeling. NeuroImage 143, 235–249. doi:https://doi.org/10.1016/j.neuroimage.2016.09.011.
  45. Environmental risk factors for schizophrenia and bipolar disorder and their relationship to genetic risk: Current knowledge and future directions. Frontiers in Genetics 12. doi:10.3389/fgene.2021.686666.
  46. Sulcation of the intraparietal sulcus is related to symbolic but not non-symbolic number skills. Developmental Cognitive Neuroscience 51, 100998. doi:https://doi.org/10.1016/j.dcn.2021.100998.
  47. Evaluation of atlas selection strategies for atlas-based image segmentation with application to confocal microscopy images of bee brains. NeuroImage 21, 1428–1442.
  48. Tversky loss function for image segmentation using 3d fully convolutional deep networks, in: Wang, Q., Shi, Y., Suk, H.I., Suzuki, K. (Eds.), Machine Learning in Medical Imaging, Springer International Publishing, Cham. pp. 379–387.
  49. Atlanta world: An expectation maximization framework for simultaneous low-level edge grouping and camera calibration in complex man-made environments, in: Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004., IEEE. pp. I–I.
  50. Joint sulci detection using graphical models and boosted priors, in: IPMI, pp. 98–109.
  51. The danish high risk and resilience study—VIA 11: Study protocol for the first follow-up of the VIA 7 cohort -522 children born to parents with schizophrenia spectrum disorders or bipolar disorder and controls being re-examined for the first time at age 11. Frontiers in Psychiatry 9. doi:10.3389/fpsyt.2018.00661.
  52. The danish high risk and resilience study – VIA 7 - a cohort study of 520 7-year-old children born of parents diagnosed with either schizophrenia, bipolar disorder or neither of these two mental disorders. BMC Psychiatry 15. doi:10.1186/s12888-015-0616-5.
  53. Hierachical isosurface segmentation based on discrete curvature. UC Davis: Institute for Data Analysis and Visualization .
  54. The development of gyrification in childhood and adolescence. Brain and Cognition 72, 36–45. doi:10.1016/j.bandc.2009.10.009.
  55. Uncovering a tripartite landmark in posterior cingulate cortex. Science Advances 8. doi:10.1126/sciadv.abn9516.
  56. Automatic segmentation of human brain sulci. Medical Image Analysis 12, 442–451. doi:https://doi.org/10.1016/j.media.2008.01.003.
  57. Volumetric segmentation and characterisation of the paracingulate sulcus on MRI scans. bioRxiv doi:10.1101/859496.
  58. SimpleITK image-analysis notebooks: a collaborative environment for education and reproducible research. Journal of Digital Imaging 31, 290–303. doi:10.1007/s10278-017-0037-8.
  59. Contrastive learning with temporal correlated medical images: A case study using lung segmentation in chest x-rays (invited paper), in: 2021 IEEE/ACM International Conference On Computer Aided Design (ICCAD), pp. 1–7.
  60. Morphological changes in the central sulcus of children with isolated growth hormone deficiency versus idiopathic short stature. Developmental Neurobiology 81, 36–46. doi:10.1002/dneu.22797.
  61. Multi-task learning for segmentation and classification of tumors in 3d automated breast ultrasound images. Medical Image Analysis 70, 101918. doi:https://doi.org/10.1016/j.media.2020.101918.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets