AI Article Synopsis
- A new hierarchical image matting model is introduced for automatically extracting blood vessels from fundus images, improving segmentation accuracy without requiring a manually created trimap.
- The method generates a trimap based on the region features of blood vessels, which simplifies the process significantly for users.
- The proposed model demonstrates high accuracy (up to 96.0%) and efficiency, outperforming other advanced techniques in speed and effectiveness on three well-known fundus image datasets.
Article Abstract
In this paper, a hierarchical image matting model is proposed to extract blood vessels from fundus images. More specifically, a hierarchical strategy is integrated into the image matting model for blood vessel segmentation. Normally the matting models require a user specified trimap, which separates the input image into three regions: the foreground, background and unknown regions. However, creating a user specified trimap is laborious for vessel segmentation tasks. In this paper, we propose a method that first generates trimap automatically by utilizing region features of blood vessels, then applies a hierarchical image matting model to extract the vessel pixels from the unknown regions. The proposed method has low calculation time and outperforms many other state-of-art supervised and unsupervised methods. It achieves a vessel segmentation accuracy of 96.0%, 95.7% and 95.1% in an average time of 10.72s, 15.74s and 50.71s on images from three publicly available fundus image datasets DRIVE, STARE, and CHASE DB1, respectively.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1109/TIP.2018.2885495 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Clinical and radiological characteristics of mediastinal melioidosis: A four-year retrospective cohort from Sabah, Malaysia.
Respir Med
November 2024
Department of Respiratory Medicine, Queen Elizabeth Hospital, Karung Berkunci No. 2029, 88586, Kota Kinabalu, Sabah, Malaysia. Electronic address:
Article Synopsis
- Melioidosis is a serious tropical infection with limited understanding of how it affects the mediastinum, leading to a study that aimed to categorize its radiological features and assess patient demographics.
- A retrospective analysis of CT scans from 70 out of 486 melioidosis patients identified 41 with mediastinal involvement, revealing diverse morphological presentations including necrotic lymph nodes and fluid collections.
- Half of the patients had diabetes, and the study suggests that further research is needed to understand the implications of these findings and the prognosis of mediastinal melioidosis.
Understanding multicentric haemangioendothelioma: diagnostic dilemmas and treatment strategies.
BMJ Case Rep
August 2024
Department of Radiodiagnosis, SIMATS Deemed University Saveetha Medical College and Hospital, Chennai, Tamil Nadu, India
Epithelioid haemangioendothelioma (EH) is a rare malignant vascular tumour occurring mainly in the liver and lungs, with bones being a rare site and primarily seen in the adult population. This case presents a male patient in his 40s who presented to the outpatient department with a chief issue of a painless swelling over the inguinal region for 4 months, gradually increasing in size, along with a history of a gradually enlarging, painless mass on his left knee over the past 5 years. Despite occasional discomfort during physical activities, the mass exhibited no associated trauma, fever, weight loss or systemic symptoms.
View Article and Find Full Text PDFCarcinoma en Cuirasse as the heralding sign of underlying HER-2 positive male breast cancer.
BMJ Case Rep
July 2024
Medical Oncology and Hematology, All India Institute of Medical Sciences-Rishikesh, Dehradun, Uttarakhand, India.
Carcinoma en cuirasse (CeC) is an uncommon presentation of metastatic cutaneous carcinoma, most often originating from breast carcinoma. We present a case study of a man in his 30s exhibiting progressive skin thickening over the left chest, alongside appetite and weight loss. On examination, the patient had painless skin induration and palpable, matted, hard, immobile and non-tender axillary, cervical and inguinal lymphadenopathy.
View Article and Find Full Text PDFReal-Time Multi-Person Video Synthesis with Controllable Prior-Guided Matting.
Sensors (Basel)
April 2024
School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing 100876, China.
In order to enhance the matting performance in multi-person dynamic scenarios, we introduce a robust, real-time, high-resolution, and controllable human video matting method that achieves state of the art on all metrics. Unlike most existing methods that perform video matting frame by frame as independent images, we design a unified architecture using a controllable generation model to solve the problem of the lack of overall semantic information in multi-person video. Our method, called ControlMatting, uses an independent recurrent architecture to exploit temporal information in videos and achieves significant improvements in temporal coherence and detailed matting quality.
View Article and Find Full Text PDFA Video-based Automated Tracking and Analysis System of Plaque Burden in Carotid Artery Using Deep Learning: A Comparison with Senior Sonographers.
Curr Med Imaging
August 2024
Department of Ultrasound, Shenzhen People's Hospital (The Second Clinical Medical College, Jinan University, The First Affiliated Hospital, Southern University of Science and Technology), Shenzhen 518020, Guangdong, China.
Background And Objective: The incidence of stroke is rising, and it is the second major cause of mortality and the third leading cause of disability around the globe. The goal of this study was to rapidly and accurately identify carotid plaques and automatically quantify plaque burden using our automated tracking and segmentation US-video system.
Methods: We collected 88 common carotid artery transection videos (11048 frames) with a history of atherosclerosis or risk factors for atherosclerosis, which were randomly divided into training, test, and validation sets using a 6:3:1 ratio.
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!