BIBSNet: A Deep Learning Baby Image Brain Segmentation Network for MRI Scans.
العنوان: | BIBSNet: A Deep Learning Baby Image Brain Segmentation Network for MRI Scans. |
---|---|
المؤلفون: | Hendrickson TJ; Minnesota Supercomputing Institute, University of Minnesota.; Masonic Institute for the Developing Brain, University of Minnesota., Reiners P; Masonic Institute for the Developing Brain, University of Minnesota., Moore LA; Masonic Institute for the Developing Brain, University of Minnesota., Perrone AJ; Masonic Institute for the Developing Brain, University of Minnesota., Alexopoulos D; Washington University., Lee EG; Minnesota Supercomputing Institute, University of Minnesota.; Masonic Institute for the Developing Brain, University of Minnesota., Styner M; Department of Psychiatry, University of North Carolina at Chapel Hill., Kardan O; Department of Psychology, University of Chicago.; University of Michigan., Chamberlain TA; Department of Psychology, University of Chicago., Mummaneni A; Department of Psychology, University of Chicago., Caldas HA; Department of Psychology, University of Chicago., Bower B; PrimeNeuro., Stoyell S; Masonic Institute for the Developing Brain, University of Minnesota., Martin T; Masonic Institute for the Developing Brain, University of Minnesota., Sung S; Masonic Institute for the Developing Brain, University of Minnesota., Fair E; Masonic Institute for the Developing Brain, University of Minnesota., Uriarte-Lopez J; Oregon Health & Science University., Rueter AR; Medical School Research Office, University of Minnesota., Yacoub E; Department of Radiology, University of Minnesota.; Center for Magnetic Resonance Research, University of Minnesota., Rosenberg MD; Department of Psychology, University of Chicago., Smyser CD; Washington University., Elison JT; Masonic Institute for the Developing Brain, University of Minnesota.; Institute of Child Development, University of Minnesota.; Department of Pediatrics, University of Minnesota., Graham A; Oregon Health & Science University., Fair DA; Masonic Institute for the Developing Brain, University of Minnesota.; Institute of Child Development, University of Minnesota.; Department of Pediatrics, University of Minnesota., Feczko E; Masonic Institute for the Developing Brain, University of Minnesota.; Department of Pediatrics, University of Minnesota. |
المصدر: | BioRxiv : the preprint server for biology [bioRxiv] 2023 May 03. Date of Electronic Publication: 2023 May 03. |
نوع المنشور: | Preprint |
اللغة: | English |
بيانات الدورية: | Country of Publication: United States NLM ID: 101680187 Publication Model: Electronic Cited Medium: Internet NLM ISO Abbreviation: bioRxiv Subsets: PubMed not MEDLINE |
مستخلص: | Objectives: Brain segmentation of infant magnetic resonance (MR) images is vitally important in studying developmental mental health and disease. The infant brain undergoes many changes throughout the first years of postnatal life, making tissue segmentation difficult for most existing algorithms. Here, we introduce a deep neural network BIBSNet ( B aby and I nfant B rain S egmentation Neural Net work), an open-source, community-driven model that relies on data augmentation and a large sample size of manually annotated images to facilitate the production of robust and generalizable brain segmentations. Experimental Design: Included in model training and testing were MR brain images on 84 participants with an age range of 0-8 months (median postmenstrual ages of 13.57 months). Using manually annotated real and synthetic segmentation images, the model was trained using a 10-fold cross-validation procedure. Testing occurred on MRI data processed with the DCAN labs infant-ABCD-BIDS processing pipeline using segmentations produced from gold standard manual annotation, joint-label fusion (JLF), and BIBSNet to assess model performance. Principal Observations: Using group analyses, results suggest that cortical metrics produced using BIBSNet segmentations outperforms JLF segmentations. Additionally, when analyzing individual differences, BIBSNet segmentations perform even better. Conclusions: BIBSNet segmentation shows marked improvement over JLF segmentations across all age groups analyzed. The BIBSNet model is 600x faster compared to JLF and can be easily included in other processing pipelines. |
معلومات مُعتمدة: | P50 HD103573 United States HD NICHD NIH HHS |
تواريخ الأحداث: | Date Created: 20230330 Latest Revision: 20240329 |
رمز التحديث: | 20240329 |
مُعرف محوري في PubMed: | PMC10055337 |
DOI: | 10.1101/2023.03.22.533696 |
PMID: | 36993540 |
قاعدة البيانات: | MEDLINE |
DOI: | 10.1101/2023.03.22.533696 |
---|