Article
Engineering, Biomedical
Yan Dong, Ting Wang, Chiyuan Ma, Zhenxing Li, Ryad Chellali
Summary: In brain tumor segmentation, both high-precision local information and global contextual information are crucial. This paper proposes a brain tumor segmentation model called DE-Uformer, which utilizes both CNN encoder and Transformer encoder to extract local features and global representations. A nested encoder-aware feature fusion (NEaFF) module is introduced to effectively fuse the information from both encoders. Experimental results demonstrate that the proposed model outperforms state-of-the-art methods in brain tumor segmentation tasks.
PHYSICS IN MEDICINE AND BIOLOGY
(2023)
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
Biology
Ramin Ranjbarzadeh, Annalina Caputo, Erfan Babaee Tirkolaee, Saeid Jafarzadeh Ghoushchi, Malika Bendechache
Summary: This study reviews recent Artificial Intelligence (AI) methods for diagnosing brain tumors using MRI images. MRI has become a widely used noninvasive imaging technique in the diagnosis and segmentation of brain tumors. However, the rapid growth of technology has created a gap between the availability of these technologies and the number of medical staff who can utilize them. Therefore, developing robust automated brain tumor detection techniques has become a major focus of research.
COMPUTERS IN BIOLOGY AND MEDICINE
(2023)
Article
Engineering, Biomedical
Zahra Sobhaninia, Nader Karimi, Pejman Khadivi, Shadrokh Samavi
Summary: This paper presents an approach that simultaneously segments and classifies brain tumors in MRI images using a framework that contains MRI image enhancement and tumor region detection. A network called Multiscale Cascaded Multitask Network is proposed, which is based on a multitask learning approach containing segmentation and classification tasks. The proposed method achieves high accuracy in both segmentation (96.27 and 95.88 for DCS and mean IoU, respectively) and classification (97.988 accuracy).
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Chemistry, Multidisciplinary
Weitao Yang, Cuijun Deng, Xiudong Shi, Yan Xu, Chenyu Dai, Hui Wang, Kexin Bian, Tianming Cui, Bingbo Zhang
Summary: This study presents a structural and molecular fusion MRI nanoprobe for differential diagnosis of benign and malignant tumors. The nanoprobe enables both structural and molecular imaging of tumors, and can catalyze chemodynamic therapy within the tumor. This innovative nanoprobe strategy allows for the diagnosis of tumors from both spatial and molecular perspectives in a single-step MRI imaging, with potential applications in precision intervention.
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
Computer Science, Information Systems
Syed Nauyan Rashid, Muhammad Hanif, Usman Habib, Akhtar Khalil, Omair Inam, Hafeez Ur Rehman
Summary: In this study, a novel deep learning-based method is proposed to improve the diagnosis and classification accuracy of gliomas, with a focus on the enhancing region. By combining data preprocessing, patch extraction, patch preprocessing, and a deep learning model, better results were achieved for all types of gliomas, including the enhancing region.
CMC-COMPUTERS MATERIALS & CONTINUA
(2022)
Article
Computer Science, Information Systems
Abhishek Bal, Minakshi Banerjee, Amlan Chakrabarti, Punit Sharma
Summary: Automated brain tumor segmentation of MR image is a challenging task. The proposed method using rough-fuzzy C-means and shape based topological properties achieved better performance in terms of statistical volume metrics compared to previous algorithms.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Engineering, Biomedical
Ahmet Ilhan, Boran Sekeroglu, Rahib Abiyev
Summary: This study proposes an efficient system for the segmentation of complete brain tumors from MRI images using a deep learning architecture called U-net. The system utilizes tumor localization and enhancement methods to improve the segmentation ability. Testing on benchmark datasets shows that the proposed methods achieve high accuracy and low cost segmentation of brain tumors in MRI images.
INTERNATIONAL JOURNAL OF COMPUTER ASSISTED RADIOLOGY AND SURGERY
(2022)
Article
Medicine, General & Internal
Duygu Sinanc Terzi, Nuh Azginoglu
Summary: Transfer learning is important when labeled data is scarce, but the contribution of natural image datasets as pre-training sources to success in different fields, such as medical imaging, is still controversial. This study quantitatively compared the effect of transfer learning for medical object detection using natural and medical image datasets. The results showed that transfer learning from the medical image dataset was more successful and showed better convergence performance than the MS COCO pre-trained model, despite having fewer data.
Article
Oncology
Tahir Mohammad Ali, Ali Nawaz, Attique Ur Rehman, Rana Zeeshan Ahmad, Abdul Rehman Javed, Thippa Reddy Gadekallu, Chin-Ling Chen, Chih-Ming Wu
Summary: Magnetic resonance imaging is commonly used for brain tumor identification, but it is time-consuming and complex. This paper proposes an attention-based convolutional neural network for brain tumor segmentation, using a pre-trained VGG19 network and attention gate for noise induction and denoising. The algorithm achieved good segmentation results on the BRATS'20 dataset.
FRONTIERS IN ONCOLOGY
(2022)
Article
Anatomy & Morphology
Andrea Chiappiniello, Roberto Tarducci, Cristina Muscio, Maria Grazia Bruzzone, Marco Bozzali, Pietro Tiraboschi, Anna Nigri, Claudia Ambrosi, Elena Chipi, Stefania Ferraro, Cristina Festari, Roberto Gasparotti, Ruben Gianeri, Giovanni Giulietti, Lorella Mascaro, Chiara Montanucci, Valentina Nicolosi, Cristina Rosazza, Laura Serra, Giovanni B. Frisoni, Daniela Perani, Fabrizio Tagliavini, Jorge Jovicich
Summary: This study evaluated the test-retest reproducibility of hippocampal subfield segmentation using FreeSurfer 6.0 and found that longitudinal pipelines with 3D-T1 and 3D-FLAIR data resulted in more accurate and spatially reproducible segmentation compared to cross-sectional pipelines. Additionally, including high-resolution 2D-T2 data in the longitudinal pipeline may further improve reproducibility.
BRAIN STRUCTURE & FUNCTION
(2021)
Article
Computer Science, Artificial Intelligence
Orcan Alpar
Summary: In this paper, a mathematical fuzzy inference-based fusion framework is proposed to enhance the segmentation of FLAIR sequences and overcome the limitations of FLAIR images. The framework achieved a high dice score coefficient (DSC) and showed promising results for whole tumor segmentation.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Nacer Farajzadeh, Nima Sadeghzadeh, Mahdi Hashemzadeh
Summary: Detecting brain tumors is crucial for patients' survival, and Magnetic Resonance Imaging (MRI) has been proven to be the most accurate method. However, the accuracy of evaluation by human specialists can be compromised due to fatigue, lack of expertise, and insufficiency of information in the images. This study proposes a segmentation approach to assist specialists in accurately detecting brain tumors, achieving the highest accuracy compared to previous studies.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Neurosciences
Liangjun Chen, Zhengwang Wu, Fenqiang Zhao, Ya Wang, Weili Lin, Li Wang, Gang Li
Summary: In this study, a context-guided, attention-based, coarse-to-fine deep framework is proposed to accurately segment subcortical structures from infant brain magnetic resonance (MR) images. The framework utilizes a SDM-Unet to predict signed distance maps (SDMs) at the coarse stage, which are then integrated with multi-modal intensity images to refine the segmentation using a multi-source and multi-path attention Unet (M2A-Unet) at the fine stage. The proposed framework achieves higher segmentation accuracy and exhibits good generalizability in both qualitative and quantitative evaluations.
Article
Radiology, Nuclear Medicine & Medical Imaging
Lisa C. Adams, Praveen Jayapal, Shakthi K. Ramasamy, Wipawee Morakote, Kristen Yeom, Lucia Baratto, Heike E. Daldrup-Link
Summary: Ferumoxytol is an approved ultrasmall iron oxide nanoparticle that has been increasingly used as an MRI contrast agent, particularly in pediatric patients. Unlike gadolinium-based contrast agents, it is biodegradable and has no potential risk of nephrogenic systemic fibrosis. It has unique MRI properties, including long-lasting vascular retention, making it suitable for various applications, such as vascular, cardiac, and cancer imaging. It is also being researched for its potential use in cellular and molecular imaging and as a potential cancer therapeutic agent.
AMERICAN JOURNAL OF ROENTGENOLOGY
(2023)
Article
Medicine, Research & Experimental
Mariam Tolba, Z. Jason Qian, Hung-Fu Lin, Kristen W. W. Yeom, Mai Thy Truong
Summary: This study aimed to determine the feasibility of using machine learning for objective assessment of aesthetic outcomes in auricular reconstructive surgery. By utilizing convolutional neural networks, images were analyzed and assigned percent scores based on confidence of classification, showing potential for objective evaluation of surgical outcomes.
Article
Computer Science, Artificial Intelligence
Adi Szeskin, Shalom Rochman, Snir Weiss, Richard Lederman, Jacob Sosna, Leo Joskowicz
Summary: This paper presents a fully automatic end-to-end pipeline for liver lesion changes analysis in consecutive CT scans. The pipeline, using the SimU-Net model, outperforms the traditional 3D R2-UNet model in lesion detection, segmentation, and matching, providing accurate and comprehensive results that can assist radiologists in evaluating radiological oncology.
MEDICAL IMAGE ANALYSIS
(2023)
Review
Engineering, Biomedical
Xingqi Fan, Qiyang Zhu, Puxun Tu, Leo Joskowicz, Xiaojun Chen
Summary: This review summarizes the application of artificial intelligence, deep learning, augmented reality, and robotics in image-guided orthopedic surgery. It covers key technologies used in the pre-operative and intra-operative stages, as well as the combination of surgical navigation system with AR and robotic technology. The current issues and prospects of the IGOS system are also discussed.
PHYSICS IN MEDICINE AND BIOLOGY
(2023)
Article
Imaging Science & Photographic Technology
Roy Eagleson, Leo Joskowicz
Summary: This paper discusses the fundamental principles of analyzing AR/VR systems for medical imaging and computer-assisted interventions. It introduces the three key concepts of analysis (verification, evaluation, and validation) and defines them using examples of AR/VR systems. The paper also defines and relates the concepts of system specifications, measurement accuracy, uncertainty, and observer variability to the analysis principles, illustrated with examples of working AR/VR systems.
JOURNAL OF IMAGING
(2023)
Review
Radiology, Nuclear Medicine & Medical Imaging
Moss Y. Y. Zhao, Elizabeth Tong, Rui Duarte Armindo, Amanda Woodward, Kristen W. Yeom, Michael E. Moseley, Greg Zaharchuk
Summary: Cerebral blood flow (CBF) is an important parameter to assess brain health. Quantitative measurement of CBF can be obtained using medical imaging techniques. However, there is a lack of CBF data in healthy children due to difficulties in pediatric neuroimaging. Understanding the factors affecting pediatric CBF and its normal range is crucial for optimal CBF measurement in pediatric neuroradiology.
JOURNAL OF MAGNETIC RESONANCE IMAGING
(2023)
Article
Computer Science, Interdisciplinary Applications
Bolun Zeng, Huixiang Wang, Jiangchang Xu, Puxun Tu, Leo Joskowicz, Xiaojun Chen
Summary: Pelvic fractures are severe traumatic injuries with high rates of morbidity and mortality. Accurate diagnosis and surgical planning require effective identification and localization of the fracture zones, which is challenging due to the complexity of pelvic fractures. This study proposes a novel two-stage method that combines the symmetry properties of pelvic anatomy and the symmetric feature differences caused by fractures to overcome the limitations of existing methods. The method utilizes a Siamese deep neural network with supervised contrastive learning and a structural attention mechanism to minimize confusion and enhance recognition of fracture zones.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2023)
Article
Engineering, Biomedical
Shalom Rochman, Adi Szeskin, Richard Lederman, Jacob Sosna, Leo Joskowicz
Summary: This paper proposes a graph-based method for the automatic detection and classification of lesion changes in CT scans. The method achieves high matching rate and detection accuracy on lung and liver datasets. The analysis of lesion changes improves quantitative follow-up, evaluation of disease status, and assessment of treatment efficacy.
INTERNATIONAL JOURNAL OF COMPUTER ASSISTED RADIOLOGY AND SURGERY
(2023)
Article
Engineering, Biomedical
Avichai Haimi, Shaul Beyth, Moshe Gross, Ori Safran, Leo Joskowicz
Summary: This paper presents a novel automatic method for the computation of glenoid bone loss in CT scans. The method consists of four steps: computation of an oblique plane, selection of the glenoid oblique CT slice, computation of the best-fit circle, and quantification of the bone loss. The evaluation results show that the method has a mean absolute error of 2.3±2.9 mm (4.67±3.32%) and can assist orthopedists in surgical planning.
INTERNATIONAL JOURNAL OF COMPUTER ASSISTED RADIOLOGY AND SURGERY
(2023)
Article
Engineering, Biomedical
Avigail Suna, Amit Davidson, Yoram Weil, Leo Joskowicz
Summary: This paper presents a novel automatic pipeline for calculating six anatomical radiographic parameters associated with distal radius fractures, providing objective support for clinical treatment decision-making.
INTERNATIONAL JOURNAL OF COMPUTER ASSISTED RADIOLOGY AND SURGERY
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Brendan Kelly, Mesha Martinez, Huy Do, Joel Hayden, Yuhao Huang, Vivek Yedavalli, Chang Ho, Pearse A. Keane, Ronan Killeen, Aonghus Lawlor, Michael E. Moseley, Kristen W. Yeom, Edward H. H. Lee
Summary: A model has been developed to classify DSA videos based on the presence of large vessel occlusion, location of occlusion, and efficacy of reperfusion. The model achieved high accuracy in identifying occlusion and classifying the location, and also showed promising results in evaluating thrombectomy efficacy.
EUROPEAN RADIOLOGY
(2023)
Article
Clinical Neurology
Bossmat Yehuda, Aviad Rabinowich, Daphna Link-Sourani, Netanell Avisdris, Ori Ben-Zvi, Bella Specktor-Fadida, Leo Joskowicz, Liat Ben-Sira, Elka Miller, Dafna Ben Bashat
Summary: This study presents an automatic pipeline for quantification of fetal brain gyrification based on routine 2D MR imaging data. It provides normal developmental curves and successfully differentiate fetuses with lissencephaly and polymicrogyria from controls. The method can be helpful in radiologic assessment and early identification of fetuses with cortical malformations.
AMERICAN JOURNAL OF NEURORADIOLOGY
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Aviad Rabinowich, Netanell Avisdris, Bossmat Yehuda, Ayala Zilberman, Tamir Graziani, Bar Neeman, Bella Specktor-Fadida, Dafna Link-Sourani, Yair Wexler, Jacky Herzlich, Karina Krajden Haratz, Leo Joskowicz, Liat Ben Sira, Liran Hiersch, Dafna Ben Bashat
Summary: Smaller, leaner fetuses are malnourished and will experience unfavorable outcomes.
JOURNAL OF MAGNETIC RESONANCE IMAGING
(2023)
Article
Engineering, Biomedical
Nir Mazor, Gili Dar, Richard Lederman, Naama Lev-Cohain, Jacob Sosna, Leo Joskowicz
Summary: This paper presents a novel method MC3DU-Net for the detection and segmentation of pancreatic cysts in MRI studies. The method can accurately and reliably perform automatic detection and segmentation of pancreatic cysts, providing a precise method for disease evaluation.
INTERNATIONAL JOURNAL OF COMPUTER ASSISTED RADIOLOGY AND SURGERY
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Netanell Avisdris, Daphna Link Sourani, Liat Ben-Sira, Leo Joskowicz, Gustavo Malinger, Simcha Yagel, Elka Miller, Dafna Ben Bashat
Summary: This study successfully differentiated hypo-/hypertelorism in fetuses using automatic biometric measurements and machine learning classification based on MRI. The results showed that the newly defined ratios and the ML multi-parametric classifier improved the accuracy of distinguishing abnormal from normal fetuses with the condition. The developed fully automatic method demonstrated high performance on varied clinical imaging data.
EUROPEAN RADIOLOGY
(2023)
