A5090663798- Brain Tumor Detection and Classification
- Traumatic Brain Injury and Neurovascular Disturbances
- Medical Imaging Techniques and Applications
- Radiomics and Machine Learning in Medical Imaging
- Advanced MRI Techniques and Applications
- Glioma Diagnosis and Treatment
- Neurosurgical Procedures and Complications
- Management of metastatic bone disease
- Brain Metastases and Treatment
- Meningioma and schwannoma management
- Intracerebral and Subarachnoid Hemorrhage Research
- Spine and Intervertebral Disc Pathology
- Medical Imaging and Analysis
- Advanced Neural Network Applications
- Acute Ischemic Stroke Management
- Medical Image Segmentation Techniques
- Spinal Fractures and Fixation Techniques
- Vascular Malformations Diagnosis and Treatment
- Advanced Radiotherapy Techniques
- Cerebrospinal fluid and hydrocephalus
- Generative Adversarial Networks and Image Synthesis
- Augmented Reality Applications
- AI in cancer detection
- Intracranial Aneurysms: Treatment and Complications
- Cancer Diagnosis and Treatment
National Taiwan University Hospital
2015-2025
Central South University
2025
Central South University of Forestry and Technology
2025
National Taiwan University
2011-2024
National Taiwan University of Arts
2020-2024
Taipei Hospital
2006-2024
Kunming University of Science and Technology
2024
E-Da Hospital
2024
National Taiwan University of Science and Technology
2011-2023
Shanghai Jiao Tong University
2014
Artificial intelligence (AI) in healthcare, especially medical imaging, faces challenges due to data scarcity and privacy concerns. Addressing these, we introduce Med-DDPM, a diffusion model designed for 3D semantic brain MRI synthesis. This effectively tackles issues by integrating conditioning. involves the channel-wise concatenation of conditioning image input, enabling control generation. Med-DDPM demonstrates superior stability performance compared existing imaging synthesis methods. It...
Graph neural networks have excellent performance and powerful representation capabilities, been widely used to handle Few-shot image classification problems. The feature extraction module of graph has always designed as a fixed convolutional network (CNN), but due the intrinsic properties convolution operations, its receiving domain is limited. This method limitations in capturing global information easily ignores key image. In order extract comprehensive critical information, new CA-MFE...
Non-enhanced head computed tomography is widely used for patients presenting with trauma or stroke, given acute intracranial hemorrhage significantly influences clinical decision-making. This study aimed to develop a deep learning algorithm, referred as DeepCT, detect on non-enhanced images and evaluate its applicability. We retrospectively collected 1,815 image sets from single center model training. Additional 3 centers were construct an independent validation dataset (VAL) 2 test datasets...
In recent years, magnetic resonance imaging (MRI) has become important in brain tumor diagnosis. Using this modality, physicians can locate specific pathologies by analyzing differences tissue character presented different types of MR images.This paper uses an algorithm integrating fuzzy-c-mean (FCM) and region growing techniques for automated image segmentation from patients with menigioma. Only non-contrasted T1 T2 -weighted images are included the analysis. The study's aims to correctly...
Abstract Background Stereotactic radiosurgery (SRS), a validated treatment for brain tumors, requires accurate tumor contouring. This manual segmentation process is time-consuming and prone to substantial inter-practitioner variability. Artificial intelligence (AI) with deep neural networks have increasingly been proposed use in lesion detection but seldom clinical setting. Methods We conducted randomized, cross-modal, multi-reader, multispecialty, multi-case study evaluate the impact of AI...
<p>The demand for artificial intelligence (AI) in healthcare is rapidly increasing. However, significant challenges arise from data scarcity and privacy concerns, particularly medical imaging. While existing generative models have achieved success image synthesis image-to-image translation tasks, there remains a gap the generation of 3D semantic images. To address this gap, we introduce Med-DDPM, diffusion model specifically designed synthesis, effectively tackling issues. The novelty...
Treatment of giant cerebral arteriovenous malformations (AVMs) remains a challenge.To propose hypofractionated stereotactic radiotherapy (HSRT) as part staged treatment, and evaluate its effect by analyzing AVM volume changes.From 2001 to 2007, 20 AVMs larger than 5 cm were treated HSRT followed up using magnetic resonance imaging. Patients' median age was 34 years (8-61 years). Eleven patients presented with hemorrhage 9 seizure. Ten had previous embolization radiosurgery failed in 4....
Stereotaxic radiosurgery (SRS) non-invasively and precisely ablates brain tumors or glioma located in the location where is surgically inaccessible with aid of three-dimensional coordination system. This technique can also treat functional psychiatric disorders, yet its dosimetry curative mechanism remain to be elucidated. In this study, a miniature pig model was utilized verify effect stereotaxic on white matter tract various doses delivered by CyberKnife. As porcine bears high resemblance...
Stereotactic radiosurgery represents a noninvasive alternative treatment for intracranial metastases.To investigate the outcome of linear accelerator-based stereotactic (linac-SRS) brainstem metastases.We retrospectively reviewed our database patients who were diagnosed with metastases and underwent linac-SRS between 1997 2008 at University California, Los Angeles.A total 45 48 treated. The median target volume was 0.40 mL (range, 0.02-5.70 mL), prescription dose 14 Gy 10-17 Gy) 90% isodose...
We propose and evaluate a systematic approach to detect classify Patient/Problem, Intervention, Comparison Outcome (PICO) from the medical literature. The training test corpora were generated systematically automatically structured PubMed abstracts. 23,472 sentences by exact pattern match of head words P-I-O categories. Afterward, terms with top frequencies used as features Naïve Bayesian classifier. This achieves F-measure values 0.91 for 0.75 Intervention 0.88 Outcome, comparable previous...
Introduction The accuracy of radiation delivery is increasingly important as radiotherapy technology continues to develop. goal this study was evaluate intrafractional motion during intracranial radiosurgery and the relationship between change treatment time. Methods Materials A total 50 records with 5988 images, all acquired treatments CyberKnife Radiosurgery System, were retrospectively analyzed in study. We measured translation rotation including superior-inferior (SI), right-left (RL),...
Semantic segmentation of medical images with deep learning models is rapidly being developed. In this study, we benchmarked state-of-the-art algorithms on our clinical stereotactic radiosurgery dataset. The dataset consists 1688 patients various brain lesions (pituitary tumors, meningioma, schwannoma, metastases, arteriovenous malformation, and trigeminal neuralgia), divided the into a training set (1557 patients) test (131 patients). This study demonstrates strengths weaknesses...
This study introduces Polyp-DDPM, a diffusion-based method for generating realistic images of polyps conditioned on masks, aimed at enhancing the segmentation gastrointestinal (GI) tract polyps. Our approach addresses challenges data limitations, high annotation costs, and privacy concerns associated with medical images. By conditioning diffusion model masks-binary masks that represent abnormal areas-Polyp-DDPM outperforms state-of-the-art methods in terms image quality (achieving Frechet...
Cerebral arteriovenous malformations (AVMs) are abnormal connections between the arteries and veins, with possible serious consequences of intracranial hemorrhage. The curative treatment for AVMs includes microsurgery radiosurgery, sometimes embolization as an adjunct. However, controversies exist options available large to giant AVMs. Hypofractionated stereotactic radiotherapy (HSRT) is one option such difficult lesions. We aim review recent literature, looking at outcome HSRT in terms AVM...
Vestibular assessment in patients with acoustic tumor (so-called vestibular schwannoma, VS) via ocular vestibular-evoked myogenic potential (oVEMP) and cervical VEMP (cVEMP) tests are not often discussed the neurosurgical literature.This study conducted physiological morphological assessments for VS before after CyberKnife radiosurgery.Twenty unilateral underwent a battery of comprising facial nerve function test, audiometry, caloric, oVEMP cVEMP 2 years treatment at mean dosage 18 Gy 3...