Kidney Stone Detection
Kidney Stone Detection
Kidney Stone Detection
Introduction
1.1 Medical Image
With increasing use of medical imgs like [Computed topography (CT) and Magnetic
resonance imaging (MRI) for diagnosis,ultra sound imgs] treatmentplanning and clinical
studies, it has become almost compulsory to use computers to assist radiological experts in
clinicaldiagnosis, treatment planning.Ultrasound imaging techniques play a crucial role in
emergency diagnostic method. It is widelyused due to its noninvasive inexpensive availability
and non radiation exposure. Medical img reconstruction for deeper understanding of clinical
abnormalities including brain tumour ,breast cancer &kidney-stone disease etc.Various types
of kidney diseases are listed under chronic Kidney diseases which maycause severe health
problem. This work emphasizes the most prevalent diseases occurring in thekidney region for
detection. It mainly emphasized on kidney stone, kidney cyst and renal cellcancer.
1
of the best imaging diagnostic techniques when compared to other imaging modalities such
asMRI, CT scan and X-ray, since of it isavailable at less expense with no harmful radiation
exposure and its smart portability. Various types of kidney diseases are listed under chronic
Kidney diseases which maycause severe health problem. This work emphasizes the most
prevalent diseases occurring in thekidney region for detection. It mainly emphasized on
kidney stone, kidney cyst and renal cellcancer[33].
KIDNEY-STONE is a solid piece of material formed due to minerals in urine. These stones
are formed by combination of genetic and environmental factors. It is also caused due to
overweight, certain foods, some medication and not drinking enough of water. Any persons
can be susceptible to kidney-stone, even infants[1].
The five common types of stones are calcium oxalate, calcium phosphate, uric acid, struvite,
and cysteine . Calcium oxalate and calcium phosphate is the mostly occurred type of kidney
stones. Many methods are used for diagnosing this kidney stone such as blood test, urine test,
scanning. Scanning also differs in CT scan, Ultrasound scan and Doppler scan. Now days a
field of automation came into existence which also being used in medical field .There are
many types of tools(computer-assisted) likeultrasound , CT scan( computed tomography), and
X-rays that deliver the most exact diagnostic tools for kidney-stone screening and diagnosis.
To convert an img into digital form and perform some operations by using img processing
method,and we can get an enhanced img or to extract some useful information from it.
KIDNEY-STONEdetec-tion is challenging because the ultra sound img have low resolution
means img have poor quality and contain speckle noise,that is problematic to examine by
human as well as machine.But medical cannot afford low accuracy as we all familiar .so that
is why,we will choose to improve classification technique in order to analyze best kidney-
stonedetection. The proposed techniques will be start with img acquisition which will be used
to take img from the external source of system. After first step we will move to the filtering
like median etc. By using filtering technique we can remove noise from the img. Generally
medical imgs have low quality particularly in contrast form. Therefore we will be need to
enhance the img.The following img is enhanced by applying histequaliz(histogram-
equalization). The preprocessed img is achieved with level set segmentation to detect the
stone region.For getting better results,we will apply segmentation process two times; first
segment for kidney portion and then apply segment for stone portion, respectively.
2
Fig1.1.kidney with stone
The aim behind this operation is to find the out the best quality, so that the identification
becomes eas-ier. Medical imaging is one of the fundamental imaging, because they are used
in more sensitive field which is a medical field and it must be accurate.
3
Fig1.2.Block diagram of the system
4
Chapter 2
Literature Review
2.1 Introduction
This section discusses the background knowledge and concepts of the domain, followed by
the related work. The related work discusses the existing approaches or algorithms in brief.
The case studies of the different research papers (on similar work space) are included
for the literature review purpose.
5
and Correlation are used. The KNN classifier is also used to classification based on training
img dataset.
*Zilong Hu, Jinshan Tang Cluster(2016) Proposed thet speckle reduction is the major task
on ultrasound imgs due to high multiplicative noises created by back scattered waves. A
cluster based anisotropic diffusion filtering technique for reducing the multiplicative noises in
ultrasound imgs which have the significant outcomes [17]
6
*Sneha A Mane, S R Chougule(2014) proposed that two neural network algorithms viz
Radial basis function and Learning vector quantization are used for diagnosis purpose Also a
comparison is made between the two algorithms using MATLAB software. The main purpose
is to find the best tool for medical diagnosis to reduce diagnosis time and increase efficiency
with accuracy[4].
*Rahman & Uddin(2013) have proposed a technique toreduction of noise like speckle and
partition byusing segmentation from US img.they provide img quality enhancement and also
detects problem in the kidney region [1].
*Tijjani and Sani(2012) provide an overview of the ANN based approaches to predicting
kidney problem through comparing mental behavior of the patient using matlab software [7].
8
Chapter 3
Proposed Methodology
Using these steps the proposed system gives a robust detection procedure.Figure 4 explains
the flow diagram for the detection of stone with the help of the entropy based segmentation.
To implement the proposed method publicly,we will use MATLAB.
The proposed techniques start with img acquisition which is used to take img from the
external source of system.In this portion we used to acquire the image using image
acquisition toolbox command in MATLAB. In order to read the image in the format of matrix
we used above mentioned toolbox. The main role of this tool-box commands to make the
image readable to the machine and can be apply other operation on the same provided format.
For this work, at first theacquired dicom images are converted into jpeg format for processing.
3.2.Image Restoration
Img Restoration is the process of gaining the original or eye-catchingimg from the degraded
img.Digital img restoration is a field of engineering that studies methods used to recover
original scene from the degraded imgs and observations. Techniques used for img restoration
9
are oriented towards modeling the degradations, usually blur and noise and applying various
filters like median filter which is actually used to remove speckle noise for the img , to
obtain an approximation of the original scene. There are a variety of reasons that could cause
degradation of an img and img restoration is one of the key fields in today's Digital Img
Processing due to its wide area of applications.
3.3.Image Enhancement
The main aim of img enhance is a processing on an IMG in order to make it more appropriate
for certain applications. Img enhance mainly sharpens img features such as boundaries, edges
or contrast and reduces the ringing artifacts. The enhancement improves the quality of the
imgs so that the information contained in them could be extracted in a meaningful sense. Img
enhance techniques improve the quality of an img as perceived by the human. .It becomes
difficult for human eyes to percept such imgs because of the increase in dynamic range of
sensing by the human eye. Many important areas such as vision, remote sensing, dynamic
scene analysis, autonomous navigation, and biomedical img analysis [33] require img
modification to show the information in a better way and reveal the important content. Img
enhance can be applied to different areas of science & engineering. Except for illumination
conditions, the quality of IMGs is also affected by external and environmental disturbances
such as ambient pressure and temperature fluctuations. Thus, img enhance is necessary. You
can apply img processing in every field where imgs are to be implicit and analyzed. Such as
img analysis in medical field, satellite img analysis, etc. By using .Img enhance techniques we
can modify the img components to increase clarity, sharpness and details through the visual
analysis and interpretation. It appearances also to transform the graphical impact in a way that
heightens the information parts of the img.
There are many img enhance procedures that we can get best quality of digital imgs without
affecting any destruction to it. Several of the img enhance techniques contain of enhancement
of contrast, intensity, hue, saturation transformations, gray level slicing, and edge
enhancement[32].
10
Fig 3.1. Submission specific img enhance
3.4.Morphological&ImgSegmentation
By using morphological operations like erosion and dilation we will find the boundaries of
theimg or what we need to detect.
Morphological operations will be applied to remove the unwanted regions. Morphological
method are based on two main operations, which are dilation and erosion.
The dilation equation is as follows:
where is the empty set and is the reflection of the structuring element B.
The erosion equation is defined as:
To detect the stone in the kidney, wewill apply the segmentation technique. By
usingsegmentation, we partition an img into distinct regions. The result of segmentation is set
of segments that collectively cover entire img or set of contours extracted from img
11
Img Segmentation has several techniques for dividing the imgs into segments. Each technique
has its own features. All techniques of img segmentation is broadly divided into three parts.
These are as follows:
Segmentation will be logical implementation in our research, which is generally used to find
the region of interest on the behalf of some characteristic of the imgs. By using the
characteristic of imgs and its features we are able to find the region where kidney may be
suspected
12
3.5. Region of interest
To identify the kidney-stone, kidney region is focused. We use the area criteria to select
thelabel that represents the stone region and to eliminate the unlikely labels (tape artifacts and
highintensity labels). Finally, to obtain the effective kidney stone region, the result of this step
wasmultiplied with the original img. Thus kidney-stone is detected [9].The result of renal
calculiimgs are used for the future analysis. From the renal calculi img, the calculi regions
areextracted.
3.6.Feature extraction
The ultrasound medical images based on texture feature. The objective of the features
extraction is to capture important characteristics of region under investigation in the kidney
image [2].We will extract the features to calculate different parameters likemean, variance,
standard devia-tion, entropy, and energyfrom the detected kidney stone image.
13
The homogeneous image shows lower entropy value, whereas,
heterogeneous region results in a
higher entropy value
3.7.Flow Chart
14
Start
ImgAquisition
Img
restoration&filteration
Img Enhancement
Morphological&
Segmentation
Find Region
Of Interest
No stone in Kidney
Detect
Interest
Feature Extraction
Calculate parameter
Performance
END
Fig4.1.Flow Chart Of Proposedmethodology
15
As showing in block diagram,the proposed techniques start with img acquisition which is
used to take img from the external source of system. After first step we move to the median
filter which is actually used to remove noise for the img. Sharpening of the img is done with
the help of un-sharp masking. Generally medical imgs are poor quality especially in contrast
form. Therefore it is required to enhance the img.
Img enhancement (IE) is essential for improving the perception of img information. IE
transforms imgs to provide better representation of the subtle detail about the imgs. Also a
classifier based on distance and dimension seems to be a very good approach for ultrasound
kidney imgs classification but for the sake of better performance we used two more classifier
in order to acquire good accuracy.
Finally in order to find the kidney-stone or region of interest we use entropy based
segmentation and morphological operations.Finally calculated different parameters like
standard deviation, entropy, thresholding, energy and homogeneity for the original img.
Img Segmentation for medical imgs has been very useful for clinicians and pathologists as it
allows the extraction of highly specific anatomical structures. This task is the first and also
one of the most critical steps of ultra sound img analysis (3). The main goal of this technique
is not only to extract information from an img but also object representation, and feature
measurement. Img segmentation technology may be performed manually or with the help of
specific algorithms. Multiple techniques such as watershed algorithm, thresholding algorithm
etc. have been applied to tackle this problem.KIDNEY-STONE detector proves to be a major
challenge for detecting the kidney stone disease. The main purpose is to find the best tool for
medical diagnosis to reduce diagnosis time and increase efficiency with accuracy and to find
out the best quality, so that the identification becomes eas-ier. Medical imaging is one of the
fundamental imaging, because they are used in more sensitive field which is a medical field
and it must be accurate.
Chapter 4
17
SIMULATION RESULTS
18
19
Input Image Pre-processed Image
Morphological Segmentation
20
True Positive Classifying Without stone as without stone
True Negative Classifying Stone as Stone
False Positive Classifying as Stone which is without stone
False Negative Classifying as Without Stone which actual result is stone
Segmentation Thresholding as (δ=10)
True False Positive(Type I False Negative(Type II
S.1.NO Positive(TP) True Negative(TN) Error)FP Error)FN
1 1 0 0 0
2 1 0 0 0
3 1 0 0 0
4 1 0 0 0
5 1 0 0 0
6 1 0 0 0
7 1 0 0 0
8 1 0 0 0
9 1 0 0 0
10 1 0 0 0
11 1 0 0 0
12 1 0 0 0
13 1 0 0 0
14 1 0 0 0
15 1 0 0 0
16 0 1 0 0
17 0 0 1 0
18 0 0 1 0
19 0 0 1 0
20 0 1 0 0
21 0 1 0 0
22 0 1 0 0
23 0 1 0 0
24 0 1 0 0
25 0 1 0 0
26 0 1 0 0
27 0 1 0 0
28 0 0 1 0
29 0 1 0 0
30 0 1 0 0
15 11 4 0
21
P(Condition Positive) 15
N(Condition negative) 15
TP 15
TN 11
FP 4
FN 0
TPR(SENSITIVITY 1
TNR(SPECIFICITY) 0.733333333
TP+FP 19
Precision 0.789473684
TP+TN 26
P+N 30
ACCURACY 0.866666667
2TP 30
2TP+FP+FN 34
F1-SCORE 0.882352941
22
Segmentation Thresholding as (δ=15)
True True False Positive(Type I False Negative(Type
S.1.NO Positive(TP) Negative(TN) Error)FP II Error)FN
1 1 0 0 0
2 1 0 0 0
3 1 0 0 0
4 1 0 0 0
5 1 0 0 0
6 1 0 0 0
7 1 0 0 0
8 1 0 0 0
9 1 0 0 0
10 1 0 0 0
11 1 0 0 0
12 1 0 0 0
13 1 0 0 0
14 1 0 0 0
15 1 0 0 0
16 0 1 0 0
17 0 1 0 0
18 0 1 0 0
19 0 0 1 0
20 0 1 0 0
21 0 1 0 0
22 0 1 0 0
23 0 1 0 0
24 0 1 0 0
25 0 1 0 0
26 0 1 0 0
23
27 0 1 0 0
28 0 1 0 0
29 0 1 0 0
30 0 1 0 0
15 14 1 0
P(Condition Positive) 15
N(Condition negative) 15
TP 15
TN 14
FP 1
FN 0
TPR(SENSITIVITY 1
TNR(SPECIFICITY) 0.933333333
TP+FP 16
Precision 0.9375
TP+TN 29
P+N 30
ACCURACY 0.966666667
2TP 30
2TP+FP+FN 31
F1-SCORE 0.967741935
24
Segmentation Thresholding as (δ=20)
False
False Positive(Type I Negative(Type
S.1.NO True Positive(TP) True Negative(TN) Error)FP II Error)FN
1 1 0 0 0
2 1 0 0 0
3 1 0 0 0
4 1 0 0 0
5 1 0 0 0
6 1 0 0 0
7 1 0 0 0
8 1 0 0 0
9 1 0 0 0
10 1 0 0 0
11 1 0 0 0
12 1 0 0 0
13 1 0 0 0
14 1 0 0 0
15 1 0 0 0
16 0 1 0 0
17 0 1 0 0
18 0 1 0 0
19 0 1 0 0
20 0 1 0 0
21 0 1 0 0
22 0 1 0 0
23 0 1 0 0
24 0 1 0 0
25 0 1 0 0
25
26 0 1 0 0
27 0 1 0 0
28 0 1 0 0
29 0 1 0 0
30 0 1 0 0
15 15 0 0
P(Condition Positive) 15
N(Condition negative) 15
TP 15
TN 15
FP 0
FN 0
TPR(SENSITIVITY 1
TNR(SPECIFICITY) 1
TP+FP 15
Precision 1
TP+TN 30
P+N 30
ACCURACY 1
2TP 30
2TP+FP+FN 30
F1-SCORE 1
3 20 1 1 1 1
26
Performance Evaluation
Threshold Vs Accuracy
27
Threshold Vs F1-score
28
CHAPTER 5
Conclusion
The segmentation to detect renal calculi will be proposed in this work. We will use edge
detection method for segmenting the renal img.. The segmentation method will be
implemented in MATLAB software. MATLAB software integrates computation,
visualization and programming in a user friendly environment where problems and solutions
are expressed in familiar mathematical notation. It serves as a complete environment for high-
level programming, as well as interactive data analysis. Many imgs with different size will be
collected from various centers and will be used for this study. Experimental results for the
detection of various types of kidney imgs using entropy based segmentation features will be
show better performance.. The obtained result will be segments the calculi part accurately
from the scan imgs which are given as input and For future implementation, this feature will
be subjected to some other Img processing method to gain an improved accuracy.
References
29
[1] Shukla A. et al , “Diagnosis of Thyroid Disorders using Artificial Neural Networks”, Department of
Information Communication and Technology, ABV-Indian Institute of Technology.
[2] T. Rahman and M. S. Uddin, “Speckle noise reduction and segmentation of kidney regions from ultrasound
img,” in Proceedings of the 2nd International Conference on Informatics,Electronics and Vision (ICIEV ’13),
pp. 1–5, IEEE, Dhaka, Bangladesh, May 2013.
[3] Rouhani M.et al, “The comparison of several ANN Architecture on thyroid disease”, Islami Azad University,
Gonabad branch Gonabad ,2010
[4]Sneha A Mane1, S R Chougule2 1M. E E& TC, BharatiVidyapeet’s College of Engg, Kolhapur, Shivaji
University, Kolhapur, Maharashtra, India 2The Principal, BharatiVidyapeet’s College of Engg, Kolhapur,
Shivaji University, Kolhapur, Maharashtra, India
[5] Duryeal A.P. et al, “Optimization of Histotripsy for Kidney Stone EROSION”, Department of Biomedical
Engineering, University of Michigan, Ann Arbor, MI 2Department of Urology, University of Michigan, Ann
Arbor, MI 2010
[6] ChahatChaudhary, Mahendra Kumar Patil, Review OfImg enhance Techniques using Histogram
Equalization, International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Volume 2, Issue 5, May 2013 ISSN 2319 – 4847.
[7] Tijjani Adam and U. HAshim And U.S.Sani , “Designing of Artificial neural network model for the
prediction of kidney problem symptoms through patients mental behavior for pre clinical medical
diagnosis”,ICBE Feb. 2012
[8] K. Viswanath and R. Gunasundari, “Kidney stone detection from ultrasound imgs by Level Set Segmentation
and multilayer perceptron ANN,” in Proceedings of the InternationalConference on Communication and
Comuting (IMCIET-ICCE’14), pp. 38–48, Elsevier, 2014.
[9] V. P. G. P. Rathi and S. Palani, “Detection and characterization of brain tumor using segmentation based on
HSOM, wavelet packet feature spaces and ANN,” in Proceedings of the 3rdInternational Conference on
Electronics Computer Technology(ICECT ’11), vol. 6, pp. 274–277, IEEE, Kanyakumari, India,April 2011.
[10] N. Koizumi, J. Seo, D. Lee et al., “Robust kidney stone tracking for a non-invasive ultrasound theragnostic
system—servoing performance and safety enhancement,” in Proceedings of theIEEE International
Conference on Robotics and Automation(ICRA ’11), pp. 2443–2450, Shanghai, China, May 2011.
[11] Hong Song, Wei Kang QianZhang and Shuliang Wang, “Kidney segmentation in CT sequences using
SKFCM and improved GrowCut algorithm”, IEEE International Conference on Bioinformatics and
Biomedicine, 2014.
30
[12] Mahdi Marsousi, Konstantinos N. Plataniotis and StergiosStergiopoulos, “Shape-Based Kidney Detection
and Segmentation in Three-Dimensional Abdominal Ultrasound Imgs”, IEEE, 2014.
[13] SamanEbrahimi and Vladimir Y. Mariano, “Img Quality Improvement in Kidney Stone Detection on
Computed Tomography Imgs”, Journal of Img and Graphics, 2015.
[14] Srimadevi.S, Sathiya.A, Saranya.B and Arulmozhi.K, “Improved Edge Detection Scheme for Segmentation
in MRI Kidney Img Registration”, International Journal of Modern Trends in Engineering and Research,
2015.
[16] Monika Pathak, HarshSadawarti, SukhdevSingh”Features extraction and classification for detection of
kidney stone region in ultrasound imgs” International Journal of Multidisciplinary Research and
Development(2016)
[17].Zilong Hu, Jinshan Tang Cluster” Driven Anisotropic Diffusion for Speckle Reduction in Ultrasound
Imgs,” In Proc. IEEE International conference for img processing, Aug 2016 , pp.2325-2329.
31
[24].VarshaSaroha ,RakeshVerma, Dr. K.K Saini“ Enhancement of Kidney Stone Imgs using different Filtering
Technique’’International Journal of scientific research and management (IJSRM) ||Volume||4||Issue||08||
Pages||4587-4591||2016||
[26]EKO SUPRIYANTO, NURUL AFIQAH TAHIR, SYED MOHD NOOH “Automatic Ultrasound Kidney’s
Centroid Detection System” Advanced Diagnostics and Progressive Human Care Research Group Research
Alliance Biotechnology
[27].Anushalin. P. S, 2 Samson Isaac. J’’Ultrasound Img Analysis of Kidney Stone using Wavelet Transform”
International Journal of Bio Sciences and Engineering, Vol 1(1), 39-49, August 2014
[28].T.RathaJeyalakshmi and K.Ramar“A Modified Method for Speckle Noise Removal in Ultrasound Medical
Imgs” International Journal of Computer and Electrical Engineering, Vol. 2, No. 1, February, 2010
[32] DibyaJyoti Bora, “Importance of Img enhance Techniques in Color IMG Segmentation: A Comprehensive
and Comparative Study”, Indian J.Sci.Res. 15 (1): 115-131, 2017
32
33