1
|
Tu R, Zhang D, Li C, Xiao L, Zhang Y, Cai X, Si W. Multimodal MRI segmentation of key structures for microvascular decompression via knowledge-driven mutual distillation and topological constraints. Int J Comput Assist Radiol Surg 2024:10.1007/s11548-024-03159-2. [PMID: 38739324 DOI: 10.1007/s11548-024-03159-2] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/09/2024] [Accepted: 04/22/2024] [Indexed: 05/14/2024]
Abstract
PURPOSE Microvascular decompression (MVD) is a widely used neurosurgical intervention for the treatment of cranial nerves compression. Segmentation of MVD-related structures, including the brainstem, nerves, arteries, and veins, is critical for preoperative planning and intraoperative decision-making. Automatically segmenting structures related to MVD is still challenging for current methods due to the limited information from a single modality and the complex topology of vessels and nerves. METHODS Considering that it is hard to distinguish MVD-related structures, especially for nerve and vessels with similar topology, we design a multimodal segmentation network with a shared encoder-dual decoder structure and propose a clinical knowledge-driven distillation scheme, allowing reliable knowledge transferred from each decoder to the other. Besides, we introduce a class-wise contrastive module to learn the discriminative representations by maximizing the distance among classes across modalities. Then, a projected topological loss based on persistent homology is proposed to constrain topological continuity. RESULTS We evaluate the performance of our method on in-house dataset consisting of 100 paired HR-T2WI and 3D TOF-MRA volumes. Experiments indicate that our model outperforms the SOTA in DSC by 1.9% for artery, 3.3% for vein and 0.5% for nerve. Visualization results show our method attains improved continuity and less breakage, which is also consistent with intraoperative images. CONCLUSION Our method can comprehensively extract the distinct features from multimodal data to segment the MVD-related key structures and preserve the topological continuity, allowing surgeons precisely perceiving the patient-specific target anatomy and substantially reducing the workload of surgeons in the preoperative planning stage. Our resources will be publicly available at https://github.com/JaronTu/Multimodal_MVD_Seg .
Collapse
Affiliation(s)
- Renzhe Tu
- Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen University Town, 1068 Xueyuan Avenue, Shenzhen, 518055, China
- University of Chinese Academy of Sciences, No.1 Yanqihu East Rd, Beijing, 101408, China
| | - Doudou Zhang
- The Second School of Clinical Medicine, Southern Medical University, No.1023, South Shatai Road, Guangzhou, 510515, China
- Department of Neurosurgery, Guangdong Second Provincial General Hospital, 466 Xingang Middle Road, Guangzhou, 510317, China
- Department of Neurosurgery, the First Affiliated Hospital of Shenzhen University, Health Science Center, Shenzhen Second People's Hospital, Sungang West Road 3002, Shenzhen, 518035, China
| | - Caizi Li
- Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen University Town, 1068 Xueyuan Avenue, Shenzhen, 518055, China.
| | - Linxia Xiao
- Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen University Town, 1068 Xueyuan Avenue, Shenzhen, 518055, China
| | - Yong Zhang
- The Second School of Clinical Medicine, Southern Medical University, No.1023, South Shatai Road, Guangzhou, 510515, China
- Department of Neurosurgery, Guangdong Second Provincial General Hospital, 466 Xingang Middle Road, Guangzhou, 510317, China
| | - Xiaodong Cai
- Department of Neurosurgery, the First Affiliated Hospital of Shenzhen University, Health Science Center, Shenzhen Second People's Hospital, Sungang West Road 3002, Shenzhen, 518035, China
| | - Weixin Si
- Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen University Town, 1068 Xueyuan Avenue, Shenzhen, 518055, China.
| |
Collapse
|
2
|
Zhang W, Yin M, Jiang M, Dai Q. Partitioned estimation methodology of biological neuronal networks with topology-based module detection. Comput Biol Med 2023; 154:106552. [PMID: 36738704 DOI: 10.1016/j.compbiomed.2023.106552] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/24/2022] [Revised: 12/27/2022] [Accepted: 01/11/2023] [Indexed: 02/02/2023]
Abstract
Parameter estimation of neuronal networks is closely related with information processing mechanisms in neural systems. Estimation of synaptic parameters for neuronal networks was an time consuming task. Due to complex interactions between neurons, computational efficiency and accuracy of estimation methods is relatively low. Meanwhile, inherent topological properties such as core-periphery and modular structures are not fully considered in estimation. In order to improve the efficiency and accuracy of estimation, this study proposes a two-stage PartitionMLE method which introduces detected neuronal modules as topological constraints in estimation. The proposed PartitionMLE method firstly decomposes the system into multiple non-overlapping neuronal modules, by performing topology-based module detection. Dynamic parameters including intra-modular and inter-modular parameters are estimated in two stages, using detected hubs to connect non-overlapping neuronal modules. The contributions of PartitionMLE method are two-folds: reducing estimation errors and improving the model interpretability. Experiments about neuronal networks consisting of Hodgkin-Huxley (HH) and leaky integrate-and-firing (LIF) neurons validated the effectiveness of the PartitionMLE method, with comparison to the single-stage MLE method.
Collapse
Affiliation(s)
- Wei Zhang
- Zhejiang Sci-Tech University, Second Street 928, Hangzhou, 310018, China.
| | - Muqi Yin
- Institute of Cyber-Systems and Control, Zhejiang University, Zheda Road 38, Hangzhou, 310027, China
| | - Mingfeng Jiang
- Zhejiang Sci-Tech University, Second Street 928, Hangzhou, 310018, China
| | - Qi Dai
- Zhejiang Sci-Tech University, Second Street 928, Hangzhou, 310018, China.
| |
Collapse
|
3
|
Rosa A. The Physical Behavior of Interphase Chromosomes: Polymer Theory and Coarse-Grain Computer Simulations. Methods Mol Biol 2022; 2301:235-58. [PMID: 34415539 DOI: 10.1007/978-1-0716-1390-0_12] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/12/2023]
Abstract
Fluorescence in situ hybridization and chromosome conformation capture methods point to the same conclusion: that chromosomes appear to the external observer as compact structures with a highly nonrandom three-dimensional organization. In this work, we recapitulate the efforts made by us and other groups to rationalize this behavior in terms of the mathematical language and tools of polymer physics. After a brief introduction dedicated to some crucial experiments dissecting the structure of interphase chromosomes, we discuss at a nonspecialistic level some fundamental aspects of theoretical and numerical polymer physics. Then, we inglobe biological and polymer aspects into a polymer model for interphase chromosomes which moves from the observation that mutual topological constraints, such as those typically present between polymer chains in ordinary melts, induce slow chain dynamics and "constraint" chromosomes to resemble double-folded randomly branched polymer conformations. By explicitly turning these ideas into a multi-scale numerical algorithm which is described here in full details, we can design accurate model polymer conformations for interphase chromosomes and offer them for systematic comparison to experiments. The review is concluded by discussing the limitations of our approach and pointing to promising perspectives for future work.
Collapse
|
4
|
Liu Y, Wang X, Wu Z, López-Linares K, Macía I, Ru X, Zhao H, González Ballester MA, Zhang C. Automated anatomical labeling of a topologically variant abdominal arterial system via probabilistic hypergraph matching. Med Image Anal 2021; 75:102249. [PMID: 34743037 DOI: 10.1016/j.media.2021.102249] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/26/2019] [Revised: 09/14/2021] [Accepted: 09/17/2021] [Indexed: 10/20/2022]
Abstract
Automated anatomical vessel labeling of the abdominal arterial system is a crucial topic in medical image processing. One reason for this is the importance of the abdominal arterial system in the human body, and another is that such labeling is necessary for the related disease diagnoses, treatments and epidemiological population analyses. We define a hypergraph representation of the abdominal arterial system as a family tree model with a probabilistic hypergraph matching framework for automated vessel labeling. Then we treat the labelling problem as the convex optimization problem and solve it with the maximum a posteriori(MAP) combined the likelihood obtained by geometric labelling with the family tree topology-based knowledge. Geometrically, we utilize XGBoost ensemble learning with an intrinsic geometric feature importance analysis for branch-level labeling. In topology, the defined family tree model of the abdominal arterial system is transferred as a Markov chain model using a constrained traversal order method and further the Markov chain model is optimized by a hidden Markov model (HMM). The probability distribution of the target branches for each candidate anatomical name is predicted and effectively embedded in the HMM model. This approach is evaluated with the leave-one-out method on 37 clinical patients' abdominal arteries, and the average accuracy is 91.94%. The obtained results are better than those of the state-of-art method with an F1 score of 93.00% and a recall of 93.00%, as the proposed method simultaneously handles the anatomical structural variability and discriminates between the symmetric branches. It is demonstrated to be suitable for labelling branches of the abdominal arterial system and can also be extended to similar tubular organ networks, such as arterial or airway networks.
Collapse
Affiliation(s)
- Yue Liu
- School of Artificial Intelligence, Beijing Normal University, China
| | - Xingce Wang
- School of Artificial Intelligence, Beijing Normal University, China.
| | - Zhongke Wu
- School of Artificial Intelligence, Beijing Normal University, China.
| | - Karen López-Linares
- Vicomtech Foundation, San Sebastián, Spain; Biodonostia Health Research Institute, San Sebastián, Spain; BCN MedTech, Dept. of Information and Communication Technologies, Universitát Pompeu Fabra, Barcelona, Spain
| | - Iván Macía
- Vicomtech Foundation, San Sebastián, Spain; Biodonostia Health Research Institute, San Sebastián, Spain
| | - Xudong Ru
- School of Artificial Intelligence, Beijing Normal University, China
| | - Haichuan Zhao
- School of Artificial Intelligence, Beijing Normal University, China
| | - Miguel A González Ballester
- BCN MedTech, Dept. of Information and Communication Technologies, Universitát Pompeu Fabra, Barcelona, Spain; ICREA, Barcelona, Spain
| | - Chong Zhang
- BCN MedTech, Dept. of Information and Communication Technologies, Universitát Pompeu Fabra, Barcelona, Spain.
| |
Collapse
|
5
|
Abstract
The Kirchhoff-Plateau problem concerns the equilibrium shapes of a system in which a flexible filament in the form of a closed loop is spanned by a liquid film, with the filament being modeled as a Kirchhoff rod and the action of the spanning surface being solely due to surface tension. We establish the existence of an equilibrium shape that minimizes the total energy of the system under the physical constraint of noninterpenetration of matter, but allowing for points on the surface of the bounding loop to come into contact. In our treatment, the bounding loop retains a finite cross-sectional thickness and a nonvanishing volume, while the liquid film is represented by a set with finite two-dimensional Hausdorff measure. Moreover, the region where the liquid film touches the surface of the bounding loop is not prescribed a priori. Our mathematical results substantiate the physical relevance of the chosen model. Indeed, no matter how strong is the competition between surface tension and the elastic response of the filament, the system is always able to adjust to achieve a configuration that complies with the physical constraints encountered in experiments.
Collapse
Affiliation(s)
- Giulio G. Giusteri
- Mathematical Soft Matter Unit, Okinawa Institute of Science and Technology Graduate University, 1919-1 Tancha, Onna, Okinawa 904-0495 Japan
| | - Luca Lussardi
- Dipartimento di Matematica e Fisica, Università Cattolica del Sacro Cuore, via Musei 41, 25121 Brescia, Italy
| | - Eliot Fried
- Mathematical Soft Matter Unit, Okinawa Institute of Science and Technology Graduate University, 1919-1 Tancha, Onna, Okinawa 904-0495 Japan
| |
Collapse
|