Automatic Hookworm Detection in Wireless Capsule Endoscopy Images

ABSTRACT:

Wireless capsule endoscopy (WCE) has become a widely used diagnostic technique to examine inflammatory bowel diseases and disorders. As one of the most common human helminths, hookworm is a kind of small tubular structure with grayish white or pinkish semi-transparent body, which is with a number of 600 million people infection around the world. Automatic hookworm detection is a challenging task due to poor quality of images, presence of extraneous matters, complex structure of gastrointestinal, and diverse appearances in terms of color and texture. This is the first few works to comprehensively explore the automatic hookworm detection for WCE images. To capture the properties of hookworms, the multi scale dual matched filter is first applied to detect the location of tubular structure. Piecewise parallel region detection method is then proposed to identify the potential regions having hookworm bodies. To discriminate the unique visual features for different components of gastrointestinal, the histogram of average intensity is proposed to represent their properties. In order to deal with the problem of imbalance data, Rusboost is deployed to classify WCE images. Experiments on a diverse and large scale dataset with 440K WCE images demonstrate that the proposed approach achieves a promising performance and outperforms the state-of-the-art methods. Moreover, the high sensitivity in detecting hookworms indicates the potential of our approach for future clinical application.

PROJECT OUTPUT VIDEO:

EXISTING SYSTEM:

• In recent years, research on WCE becomes a hot topic. Computer aided detection systems based on different techniques have been extensively conducted, which bring endoscopists great convenience and efficiency.

• Comprehensive surveys on WCE research can be found in existing papers, which summarize the latest development from different aspects. The research on WCE covers a wide range of topics, including image enhancement, tissue segmentation or matching, video segmentation, motility event detection, video summarization, and so on.

• A majority of the research on WCE focuses on pathological abnormality detection

DISADVANTAGES OF EXISTING SYSTEM:

• The existing work is evaluated in a relative small and balanced dataset images, which is not applicable for practical scenarios with unbalance dataset. The characteristics of hookworms are quite different from bleeding, ulcers and polyps.

• The boundary and body of hookworms are also different from the patterns of existing pathologies. It remains unclear whether existing approaches for other lesion detection are also effective for hookworm detection.

PROPOSED SYSTEM:

• A novel automatic hookworm detection approach is proposed for WCE images, which analyzes the morphology and geometrical characteristic of hookworms. The contributions of this work are as follow:

• To the best of our knowledge, this is the first few works to comprehensively explore the automatic hookworm detection for WCE images.

• The piecewise parallel region detection (PPRD) and the uncurled tubular region (UTR) are novelly proposed to detect the parallel regions and represent the extracted regions, respectively.

• To discriminate the unique features for different components of GI, such as hookworms, bubbles, and folds, the histogram of average intensity (HAI) is proposed to represent their properties.

• The experiments are performed on the datasets for WCE detection, WCE images of 11 patients. The results demonstrate that the proposed approach outperforms the state of-the-art approaches.

ADVANTAGES OF PROPOSED SYSTEM:

• Experiments from different aspects demonstrate that the proposed method is a robust classification tool for hookworm detection, which achieves promising performance.

• Reduces the number of images a clinician needs to review.

MODULES:

1. Image Acquisition

2. Preprocessing

3. Segmentation

4. Feature Extraction

5. Classification

MODULE DESCRIPTION:

1. Image Acquisition:

Images are acquired from Gallery.

2. Preprocessing:

The objective of the preprocessing phase is to apply possible image enhancement techniques to obtain the required visual quality of the ultrasound images.

Image enhancement techniques,

1. Guided Filter Process

1. Guided Filter Process:

• To have clear details and conspicuous texture information, the guided filter is first adopted for WCE image enhancement due to its good performance and efficiency.

3. Segmentation:

i. Multiscale Matched Filter Method

ii. Parallel Region Detection

i. Mulitscale Matched Filter Method:

After preprocessing stage, Dual Matched Filter is employed to detect tubular region in the WCE images. The matched filter is an effective approach, which is originally pro-posed to detect retinal vessels. The matched filter is a Gaussian-shaped template, which is based on prior information that the cross-section of a vessel is Gaussian-shaped. Since the vessels appear darker than other retinal surfaces, the tubular region will produce a higher response when the matched filter convolves with the image.

ii. Parallel Region Detection:

After tubular region detection using multi-scale dual matched filter, the detection results contain potential tubular regions of hookworms as well as some non-tubular region-s, such as bubbles and intestine folds, which have similar structure as hookworms. The edges of hookworm bodies usually demonstrate parallel shapes, which is a useful cue to distinguish tubular structure and non-tubular one. Therefore, Piecewise Parallel Region Detection (PPRD) is proposed to detect the parallel edges.

First, for each binary tubular region, the Canny detector is adopted to detect the edges of tubular structure in the gray image. Short edges with small gaps are connected together to form a long edge by performing edge linking. The isolated dots are treated as noises and removed.

4. Feature Extraction:

i. DWT:

In this stage, DWT is used for extract the features from segmented image. In case of 2D images, the DWT is applied to each dimension separately. As a result, there are 4 sub-band (LL, LH, HH, and HL) images at each scale.

ii. Feature Reduction using PCA

Excessive features increase computation times and storage memory. Furthermore, they sometimes make classification more complicated, which is called the curse of dimensionality. It is required to reduce the number of features. PCA is an efficient tool to reduce the dimension of a data set consisting of a large number of interrelated variables while retaining most of the variations. It is achieved by transforming the data set to a new set of ordered variables according to their variances or importance.

iii. Final Features:

After reduce the feature dimensionality, extract final feature such as mean, standard deviation and energy. These are our final feature vector.

5. Classification:

The classification process is done over the segmented images. The main novelty here is the adoption of Random Forest. RF classifier is applied over the segmented images and the classification is done.

Random Forest:

One of the foremost common ways or frameworks utilized by knowledge scientists at the ‘rose knowledge science skilled follow cluster’ is Random Forests. The Random Forests formula is one in every of the simplest among classification algorithms -able to classify giant amounts of information with accuracy. Random Forests are associate degree ensemble learning technique (also thought of as a kind of nearest neighbor predictor) for classification associate degreed regression that construct variety of call trees at coaching time and outputting the category that’s the mode of the categories output by individual trees (Random Forests may be a trademark of Leo Bremen and Adele monger for an ensemble of call trees).

SYSTEM REQUIREMENTS:

HARDWARE REQUIREMENTS:

• System           : Pentium Dual Core.

• Hard Disk      : 120 GB.

• Monitor          : 15’’ LED

• Input Devices : Keyboard, Mouse

• Ram                : 1GB.

SOFTWARE REQUIREMENTS:

• Operating system : Windows 7.

• Coding Language : MATLAB

• Tool                      : MATLAB R2013A

REFERENCE:

Xiao Wu_, Member, IEEE, Honghan Chen, Tao Gan, Junzhou Chen, Member, IEEE, Chong-Wah Ngo, Member, IEEE, and Qiang Peng, “Automatic Hookworm Detection in Wireless Capsule Endoscopy Images”, IEEE TRANSACTIONS ON MEDICAL IMAGING. VOL. X, NO. X, MARCH 2016.