Article
Oncology
Botao Zhao, Yan Ren, Ziqi Yu, Jinhua Yu, Tingying Peng, Xiao-Yong Zhang
Summary: An automatically unsupervised segmentation toolbox named AUCseg was proposed in this work for high-grade gliomas, achieving good segmentation performance and fast computing speed on the BraTS2018 dataset.
FRONTIERS IN ONCOLOGY
(2021)
Article
Neurosciences
Shao-Lun Lu, Heng-Chun Liao, Feng-Ming Hsu, Chun-Chih Liao, Feipei Lai, Furen Xiao
Summary: The ICTS dataset consists of contrast-enhanced T1-weighted images of 1500 patients, with tumors labeled by qualified neurosurgeons and radiation oncologists. This dataset is publicly available for ongoing benchmarking through an online evaluation system.
Article
Computer Science, Artificial Intelligence
Stine Hansen, Samuel Kuttner, Michael Kampffmeyer, Tom-Vegard Markussen, Rune Sundset, Silje Kjaernes Oen, Live Eikenes, Robert Jenssen
Summary: This study presents an unsupervised supervoxel-based clustering framework for lung tumor segmentation in hybrid PET/MRI imaging. The framework successfully detected 15 out of 19 tumors in an unsupervised manner, with a considerable performance increase when segmenting across patients, resulting in a higher mean dice score. Results indicate that spectral clustering and Manhattan hierarchical clustering have potential for tumor segmentation in PET/MRI with low missed tumors and false-positives, with spectral clustering showing more robustness to noise.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Orhun Aydin, Mark. V. Janikas, Renato Martins Assuncao, Ting-Hwan Lee
Summary: This study investigates the performance of regionalization algorithms on a simulated benchmark dataset, utilizing various metrics to evaluate quality. The analysis reveals strengths and weaknesses of each method, demonstrating computational efficiency and preferred circumstances.
INTERNATIONAL JOURNAL OF GEOGRAPHICAL INFORMATION SCIENCE
(2021)
Article
Neurosciences
Zhenwei Wang, Mengshen He, Yifan Lv, Enjie Ge, Shu Zhang, Ning Qiang, Tianming Liu, Fan Zhang, Xiang Li, Bao Ge
Summary: In this study, geometric features were used for fiber tract segmentation, and a novel descriptor (FiberGeoMap) was developed to effectively depict the shapes and positions of fiber streamlines. Experimental results showed that the proposed method outperformed existing methods in both the number of categories and segmentation accuracy for differentiating 103 various fiber tracts. Additionally, the method identified statistically different fiber tracts on fractional anisotropy (FA), mean diffusion (MD), and fiber number ratio in autism.
Article
Mathematics
Branislav Panic, Marko Nagode, Jernej Klemenc, Simon Oman
Summary: Unsupervised image segmentation is an important task in computer vision systems, and a mixture model can be used to obtain segmented images. However, problems arise when the optimal mixture model contains a large number of components. This paper investigates methods for merging the components of the mixture model and their usefulness for unsupervised image segmentation.
Article
Computer Science, Artificial Intelligence
Fumin Guo, Dante Pi Capaldi, David G. McCormack, Aaron Fenster, Grace Parraga
Summary: The study developed a lung segmentation method using CKKM algorithm combined with U-net and atlas-based approaches, which improved the accuracy and precision of imaging pulmonary structural abnormalities.
MEDICAL IMAGE ANALYSIS
(2021)
Article
Computer Science, Information Systems
Abdul Haseeb Nizamani, Zhigang Chen, Ahsan Ahmed Nizamani, Uzair Aslam Bhatti
Summary: The precision of medical image segmentation is of immense significance in modern healthcare. Deep learning techniques have revolutionized this field by automating manual segmentation processes, but challenges like intricate structures and indistinct features persist. This study introduces three novel feature-enhanced hybrid UNet models for brain tumor MRI image segmentation and demonstrates their excellence through rigorous experimentation, surpassing current state-of-the-art models.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2023)
Article
Computer Science, Interdisciplinary Applications
Baoshi Chen, Lingling Zhang, Hongyan Chen, Kewei Liang, Xuzhu Chen
Summary: The proposed machine learning-based method in this paper demonstrates high accuracy in automatically detecting, segmenting, and classifying brain tumors, with a 96.05% accuracy for automatically classifying brain tumors. Further studies should focus on obtaining more negative examples and exploring the performance of deep learning algorithms for automatic diagnosis and segmentation of brain tumors.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2021)
Article
Neurosciences
Fang-Cheng Yeh, Andrei Irimia, Dhiego Chaves de Almeida Bastos, Alexandra J. Golby
Summary: This review discusses the challenges of anatomical accuracy in fiber tracking, explores the impact of different white matter pathways on tracking methods, and summarizes the pros and cons of commonly-used techniques. Additionally, it introduces the progress in clinical applications of tractography in patients with brain tumors and traumatic brain injury, highlighting current limitations and future directions for development.
Article
Chemistry, Multidisciplinary
Adham Aleid, Khalid Alhussaini, Reem Alanazi, Meaad Altwaimi, Omar Altwijri, Ali S. Saad
Summary: Artificial intelligence (AI) is used in medical imaging to enhance automatic diagnosis and early detection of brain tumors. This study proposes a classical automatic segmentation method based on MRI images, using a multilevel thresholding technique and harmony search algorithm. The results show competitive accuracy and improved execution time compared to CNN and DLA methods.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Information Systems
Ibrahim Mahmoud El-Henawy, Mostafa Elbaz, Zainab H. Ali, Noha Sakr
Summary: This paper proposes a novel framework that uses 3D MRI images from Kaggle and applies different diverse models to remove Rician and speckle noise for tumor segmentation. Experimental results show the efficiency of the proposed framework against classical filters such as Bilateral, Frost, Kuan, and Lee.
CMC-COMPUTERS MATERIALS & CONTINUA
(2023)
Article
Multidisciplinary Sciences
Fernando Yepes-Calderon, J. Gordon McComb
Summary: The size of brain's ventricles is crucial in diagnosing and treating neurological disorders. This study proposes a method using artificial intelligence and 3D printed models to accurately measure the ventricular volume. The results provide certified volumes for different age groups and hydrocephalus patients.
SCIENTIFIC REPORTS
(2022)
Article
Computer Science, Information Systems
Himanshu Mittal, Avinash Chandra Pandey, Mukesh Saraswat, Sumit Kumar, Raju Pal, Garv Modwel
Summary: This paper reviews various clustering-based image segmentation methods, with a focus on partitional clustering methods including K-means, histogram-based, and meta-heuristic methods. It also discusses various performance parameters for quantitative evaluation of segmentation results, along with publicly available benchmark datasets for image segmentation.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Puneet Kumar, R. K. Agrawal, Dhirendra Kumar
Summary: In this study, an improved fuzzy bounded k-plane clustering method (FBkPC_S1) is proposed, which efficiently clusters non-spherically distributed data and handles noise. The method utilizes FCM objective function to constrain cluster planes and incorporates local spatial information in the objective function of FkPC to handle noise. Extensive experiments on synthetic image and human brain MRI datasets demonstrate the fast and robust performance of the proposed method in providing accurate segmentation in the presence of noise artifacts. The proposed FBkPC_S1 method achieves superior average segmentation accuracy and Dice score compared to related methods.
APPLIED SOFT COMPUTING
(2023)