Integrating machine learning with region-based active contour models in medical image segmentation

Pratondo, Agus; Chui, Chee-Kong; Ong, Sim‐Heng

doi:10.1016/j.jvcir.2016.11.019

Cited by 97 publications

(41 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The region based active contour often fails for images having inhomogeneous intensity. Machine learning algorithms are more capable of handling inhomogeneous intensity of images [3]. The image is restored by introducing the cost function that enhances the contrast and consists of information loss term [4].…”

Section: Previous Workmentioning

confidence: 99%

Development of Feature Based Classification of Fruit using Deep Learning

Yogesh*¹,

Dubey²,

Ratan³

2019

IJITEE

View full text Add to dashboard Cite

One of the national primary health care services in Malaysia is school health care. This care is very crucial as it ensures that, countrywide, the health of students from the age of five to fifteen is in a good condition. In Malaysia, nurses hold a major responsibility for delivering the school health service. However, there is no solid research investigating the nursing time required to deliver school health services. This paper presents a system dynamics model representing the specific school health services delivered by nurses. System Dynamics is a computer-aided approach to policy analysis and design. In this paper, the system dynamics model are represented by several causal loop diagrams which covers all the school health activities and is able to determine the projected total nurse time required in delivering the service. The baseline simulation result of the nurse time required for delivering school health services is about 1080000 hours in year 2030, which is equivalent to 680 full time equivalent (FTE) nurses. Furthermore, various what-if analyses are tested with the model, as it is important for policy makers to investigate various scenarios for an effective decision-making process. In other words, the theme of the study is to understand the implication of the changes in school population size and the modification of certain activities in the school health program on the nurse time spent delivering school health service by developing a dedicated forecasting system dynamics model for school health. The time horizon for the forecasting is from 2018 until 2030Fruit classification is a challenging task in image processing. Computer vision based classification method is agile and rigorous compared to human based approach. In this paper, a method is developed for feature classification using deep learning. The region with their own characteristics is classified based on deep learning convolutional neural network technique. Traditional method for diagnosis of fruit involves visual observations by experts. The interference of environmental factors needs to be considered during diagnosis process. Datasets such as VOC, PASCAL, ImageNet etc. are easily available that are used for training of several different types of objects. The proposed model introduces two pre-trained networks; AlexNet and GoogLeNet. For faster and optimized training, Rectified linear unit (ReLu) is used that maintain positive value and map negative values to zero. The model learns to perform classification directly from images. Neural network architecture is used for implementation of deep learning. Error in deep learning is minimized compared to machine learning. The high end GPU’s reduces the training time. A transfer learning technique is proposed to retrain the network that is capable of performing new recognition task.

show abstract

Section: Previous Workmentioning

confidence: 99%

Development of Feature Based Classification of Fruit using Deep Learning

Yogesh*¹,

Dubey²,

Ratan³

2019

IJITEE

View full text Add to dashboard Cite

show abstract

“…Several techniques have been proposed to segment medical images [16][17][18][19]. We present a novel automated bio-image segmentation method for cardiomyocyte images obtained by DH, based on both region and edge information, using k-means clustering and a marker-controlled watershed segmentation algorithm.…”

Section: Funding Informationmentioning

confidence: 99%

Automated quantitative analysis of multiple cardiomyocytes at the single‐cell level with three‐dimensional holographic imaging informatics

Moon

Jaferzadeh

Ahmadzadeh

et al. 2018

Journal of Biophotonics

View full text Add to dashboard Cite

Cardiomyocytes derived from human pluripotent stem cells are a promising tool for disease modeling, drug compound testing, and cardiac toxicity screening. Bio-image segmentation is a prerequisite step in cardiomyocyte image analysis by digital holography (DH) in microscopic configuration and has provided satisfactory results. In this study, we quantified multiple cardiac cells from segmented 3-dimensional DH images at the single-cell level and measured multiple parameters describing the beating profile of each individual cell. The beating profile is extracted by monitoring dry-mass distribution during the mechanical contraction-relaxation activity caused by cardiac action potential. We present a robust two-step segmentation method for cardiomyocyte low-contrast image segmentation based on region and edge information. The segmented single-cell contains mostly the nucleus of the cell since it is the best part of the cardiac cell, which can be perfectly segmented. Clustering accuracy was assessed by a silhouette index evaluation for k-means clustering and the Dice similarity coefficient (DSC) of the final segmented image. 3D representation of single of cardiomyocytes. The cell contains mostly the nucleus section and a small area of cytoplasm.

show abstract

“…We can find some recent works using neural networks [44][45][46], Gaussian mixture model [47,48], support vector machine [49][50][51], and support vector machine with -means family based training [52,53]. Even though -means algorithm is old, it is still used in image segmentation due to its ease of implementation [39,[52][53][54].…”

Section: Adapting Color Model To -Meansmentioning

confidence: 99%

Improved Unsupervised Color Segmentation Using a Modified HSV Color Model and a Bagging Procedure in K-Means++ Algorithm

Chavolla

Valdivia

Diaz

et al. 2018

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Accurate color image segmentation has stayed as a relevant topic between the researches/scientific community due to the wide range of application areas such as medicine and agriculture. A major issue is the presence of illumination variations that obstruct precise segmentation. On the other hand, the machine learning unsupervised techniques have become attractive principally for the easy implementations. However, there is not an easy way to verify or ensure the accuracy of the unsupervised techniques; so these techniques could lead to an unknown result. This paper proposes an algorithm and a modification to the color model in order to improve the accuracy of the results obtained from the color segmentation using the -means++ algorithm. The proposal gives better segmentation and less erroneous color detections due to illumination conditions. This is achieved shifting the hue and rearranging the equation in order to avoid undefined conditions and increase robustness in the color model.

show abstract

Integrating machine learning with region-based active contour models in medical image segmentation

Cited by 97 publications

References 24 publications

Development of Feature Based Classification of Fruit using Deep Learning

Development of Feature Based Classification of Fruit using Deep Learning

Automated quantitative analysis of multiple cardiomyocytes at the single‐cell level with three‐dimensional holographic imaging informatics

Improved Unsupervised Color Segmentation Using a Modified HSV Color Model and a Bagging Procedure in K-Means++ Algorithm

Contact Info

Product

Resources

About