Cardiac CT Image Segmentation for Deep Learning–Based Coronary Calcium Detection Using K-Means Clustering and Grabcut Algorithm

Lee, Sung‐Jin; Lee, Ah Young; Hong, Min

doi:10.32604/csse.2023.037055

Cited by 5 publications

(3 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Extrapolation using linear regression is a widely used method in numerous research fields for predicting data outside the observed range. − In recent years, several studies have focused on developing advanced analysis models to enhance the accuracy of data prediction. The performance of the SoW is comparable to that of conventional measurement equipment obtained in another study, which measured blood flow using a Doppler wire in the coronary artery of experimental animals with the same type and similar weight and recorded a speed of 65–117 cm/s in the left main coronary artery.…”

Section: Resultsmentioning

confidence: 99%

Monolithic Integration of Ultraslim Flow Sensor and Medical Guidewire by Laser Filament Scanning Sintering for In Vivo Diagnostics of Cardiovascular Diseases

Cho,

Kim,

Back

et al. 2023

ACS Sens.

View full text Add to dashboard Cite

In this study, an ultraslim thermal flow sensor system integrated onto a 340 μm diameter medical guidewire was developed using a laser filament scanning sintering method for the early diagnosis of cardiovascular diseases. The proposed system is a calorimetric-based micro thermal flow sensor comprising a microheater and two thermistors. Prior to fabrication, the sensor design was optimized through flow simulation, and the patterned sensor was successfully implemented on a thin and curved surface of the medical guidewire using a laser patterning method with Ag nanoparticles. The performance of the ultraslim thermal flow sensor-onguidewire system (SoW) was evaluated under pulsatile flow by using an artificial heartbeat simulator with differentially induced fluid flow velocities of up to 60 cm/s. The resulting electrical signals generated by the temperature difference between the two thermistors caused by the fluid flow were measured across different velocity ranges. Based on the obtained data, a calibration curve was derived to establish the relationship between the fluid velocity and the sensor output voltage. Furthermore, the SoW was tested on living animals, whereby the measured blood flow velocities were 60−90 cm/s in the left coronary artery of pigs. This research demonstrates the potential of ultraslim microsensors, such as the developed thermal flow sensor system, for various industries, particularly in the medical field.

show abstract

Section: Resultsmentioning

confidence: 99%

Monolithic Integration of Ultraslim Flow Sensor and Medical Guidewire by Laser Filament Scanning Sintering for In Vivo Diagnostics of Cardiovascular Diseases

Cho,

Kim,

Back

et al. 2023

ACS Sens.

View full text Add to dashboard Cite

show abstract

“…The revised encoder in the form of U-Net encoder is utilized for segmentation and experimental study is performed on datasets like ARIA [38]. The authors employed grabcut technique with K-means clustering for enhancing the segmentation of CT scan images for various parts and deep learning techniques were used for detection of disease [39]. Authors suggested performing global average pooling, replacing the fully connected layer of the neural network to improvise the feature selection process, hence the image pre-processing procedure can be improved [40].…”

Section: Literature Reviewmentioning

confidence: 99%

An Improved Lung Cancer Segmentation Based on Nature-Inspired Optimization Approaches

Shamas,

Panda,

Sharma

et al. 2024

CMES

View full text Add to dashboard Cite

The distinction and precise identification of tumor nodules are crucial for timely lung cancer diagnosis and planning intervention. This research work addresses the major issues pertaining to the field of medical image processing while focusing on lung cancer Computed Tomography (CT) images. In this context, the paper proposes an improved lung cancer segmentation technique based on the strengths of nature-inspired approaches. The better resolution of CT is exploited to distinguish healthy subjects from those who have lung cancer. In this process, the visual challenges of the K-means are addressed with the integration of four nature-inspired swarm intelligent techniques. The techniques experimented in this paper are K-means with Artificial Bee Colony (ABC), K-means with Cuckoo Search Algorithm (CSA), K-means with Particle Swarm Optimization (PSO), and Kmeans with Firefly Algorithm (FFA). The testing and evaluation are performed on Early Lung Cancer Action Program (ELCAP) database. The simulation analysis is performed using lung cancer images set against metrics: precision, sensitivity, specificity, f-measure, accuracy, Matthews Correlation Coefficient (MCC), Jaccard, and Dice. The detailed evaluation shows that the K-means with Cuckoo Search Algorithm (CSA) significantly improved the quality of lung cancer segmentation in comparison to the other optimization approaches utilized for lung cancer images. The results exhibit that the proposed approach (K-means with CSA) achieves precision, sensitivity, and Fmeasure of 0.942, 0.964, and 0.953, respectively, and an average accuracy of 93%. The experimental results prove that K-means with ABC, K-means with PSO, K-means with FFA, and K-means with CSA have achieved an improvement of 10.8%, 13.38%, 13.93%, and 15.7%, respectively, for accuracy measure in comparison to K-means segmentation for lung cancer images. Further, it is highlighted that the proposed K-means with CSA have achieved a significant improvement in accuracy, hence can be utilized by researchers for improved segmentation processes of medical image datasets for identifying the targeted region of interest.

show abstract

“…Watershed-based segmentation methods are sensitive to noise and can lead to over-segmentation [5]. The GrabCut algorithm is an interactive segmentation technique that requires users to draw a rectangular bounding box on the image to indicate approximate regions of foreground and background [6]. The algorithm iteratively refines the segmentation results based on these markings.…”

Section: Introductionmentioning

confidence: 99%

A Method of Lung Organ Segmentation in CT Images Based on Multiple Residual Structures and an Enhanced Spatial Attention Mechanism

Wang,

Zhang,

Zhang

et al. 2023

Mathematics

View full text Add to dashboard Cite

Accurate organ segmentation is a fundamental step in disease-assisting diagnostic systems, and the precise segmentation of lung is crucial for subsequent lesion detection. Prior to this, lung segmentation algorithms had typically segmented the entire lung tissue. However, the trachea is also essential for diagnosing lung diseases. Challenges in lung parenchyma segmentation include the limited robustness of U-Net in acquiring contextual information and the small size of the trachea being mixed up with lung, making it difficult to identify and reconstruct the lungs. To overcome these difficulties, this paper proposes three improvements to U-Net: multiple concatenation modules to enhance the network’s ability to capture context, multi-scale residual learning modules to improve the model’s multi-scale learning capabilities, and an enhanced gated attention mechanism to enhance the fusion of various hierarchical features. The experimental results demonstrate that our model has achieved a significant improvement in trachea segmentation compared to existing models.

show abstract

Cardiac CT Image Segmentation for Deep Learning–Based Coronary Calcium Detection Using K-Means Clustering and Grabcut Algorithm

Cited by 5 publications

References 18 publications

Monolithic Integration of Ultraslim Flow Sensor and Medical Guidewire by Laser Filament Scanning Sintering for In Vivo Diagnostics of Cardiovascular Diseases

Monolithic Integration of Ultraslim Flow Sensor and Medical Guidewire by Laser Filament Scanning Sintering for In Vivo Diagnostics of Cardiovascular Diseases

An Improved Lung Cancer Segmentation Based on Nature-Inspired Optimization Approaches

A Method of Lung Organ Segmentation in CT Images Based on Multiple Residual Structures and an Enhanced Spatial Attention Mechanism

Contact Info

Product

Resources

About