Bibliographic Details
| Title: |
A Hybrid Transformer-Convolutional Neural Network for Segmentation of Intracerebral Hemorrhage and Perihematomal Edema on Non-Contrast Head Computed Tomography (CT) with Uncertainty Quantification to Improve Confidence |
| Authors: |
Anh T. Tran, Dmitriy Desser, Tal Zeevi, Gaby Abou Karam, Fiona Dierksen, Andrea Dell’Orco, Helge Kniep, Uta Hanning, Jens Fiehler, Julia Zietz, Pina C. Sanelli, Ajay Malhotra, James S. Duncan, Sanjay Aneja, Guido J. Falcone, Adnan I. Qureshi, Kevin N. Sheth, Jawed Nawabi, Seyedmehdi Payabvash |
| Source: |
Bioengineering, Vol 11, Iss 12, p 1274 (2024) |
| Publisher Information: |
MDPI AG, 2024. |
| Publication Year: |
2024 |
| Collection: |
LCC:Technology
LCC:Biology (General) |
| Subject Terms: |
intracerebral hemorrhage, perihematomal edema, segmentation, Technology, Biology (General), QH301-705.5 |
| More Details: |
Intracerebral hemorrhage (ICH) and perihematomal edema (PHE) are key imaging markers of primary and secondary brain injury in hemorrhagic stroke. Accurate segmentation and quantification of ICH and PHE can help with prognostication and guide treatment planning. In this study, we combined Swin-Unet Transformers with nnU-NETv2 convolutional network for segmentation of ICH and PHE on non-contrast head CTs. We also applied test-time data augmentations to assess individual-level prediction uncertainty, ensuring high confidence in prediction. The model was trained on 1782 CT scans from a multicentric trial and tested in two independent datasets from Yale (n = 396) and University of Berlin Charité Hospital and University Medical Center Hamburg-Eppendorf (n = 943). Model performance was evaluated with the Dice coefficient and Volume Similarity (VS). Our dual Swin-nnUNET model achieved a median (95% confidence interval) Dice = 0.93 (0.90–0.95) and VS = 0.97 (0.95–0.98) for ICH, and Dice = 0.70 (0.64–0.75) and VS = 0.87 (0.80–0.93) for PHE segmentation in the Yale cohort. Dice = 0.86 (0.80–0.90) and VS = 0.91 (0.85–0.95) for ICH and Dice = 0.65 (0.56–0.70) and VS = 0.86 (0.77–0.93) for PHE segmentation in the Berlin/Hamburg-Eppendorf cohort. Prediction uncertainty was associated with lower segmentation accuracy, smaller ICH/PHE volumes, and infratentorial location. Our results highlight the benefits of a dual transformer-convolutional neural network architecture for ICH/PHE segmentation and test-time augmentation for uncertainty quantification. |
| Document Type: |
article |
| File Description: |
electronic resource |
| Language: |
English |
| ISSN: |
2306-5354 |
| Relation: |
https://www.mdpi.com/2306-5354/11/12/1274; https://doaj.org/toc/2306-5354 |
| DOI: |
10.3390/bioengineering11121274 |
| Access URL: |
https://doaj.org/article/215928c76a20488dbb57d56884979aa4 |
| Accession Number: |
edsdoj.215928c76a20488dbb57d56884979aa4 |
| Database: |
Directory of Open Access Journals |
