AU2021103132A4 - An automatic tumor detection system based on local linear wavelet artificial neural network with hybrid optimization - Google Patents
An automatic tumor detection system based on local linear wavelet artificial neural network with hybrid optimization Download PDFInfo
- Publication number
- AU2021103132A4 AU2021103132A4 AU2021103132A AU2021103132A AU2021103132A4 AU 2021103132 A4 AU2021103132 A4 AU 2021103132A4 AU 2021103132 A AU2021103132 A AU 2021103132A AU 2021103132 A AU2021103132 A AU 2021103132A AU 2021103132 A4 AU2021103132 A4 AU 2021103132A4
- Authority
- AU
- Australia
- Prior art keywords
- neural network
- artificial neural
- local linear
- optimization
- tumors
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
- 206010028980 Neoplasm Diseases 0.000 title claims abstract description 88
- 238000001514 detection method Methods 0.000 title claims abstract description 39
- 238000005457 optimization Methods 0.000 title claims abstract description 35
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 31
- 230000011218 segmentation Effects 0.000 claims abstract description 27
- 208000003174 Brain Neoplasms Diseases 0.000 claims abstract description 18
- 238000000605 extraction Methods 0.000 claims abstract description 16
- 241000254158 Lampyridae Species 0.000 claims abstract description 11
- 238000007781 pre-processing Methods 0.000 claims abstract description 11
- 230000003211 malignant effect Effects 0.000 claims abstract description 7
- 201000011510 cancer Diseases 0.000 claims abstract description 6
- 230000035945 sensitivity Effects 0.000 claims abstract description 4
- 239000013598 vector Substances 0.000 claims description 5
- 239000002245 particle Substances 0.000 claims description 4
- 210000004556 brain Anatomy 0.000 abstract description 27
- 238000002595 magnetic resonance imaging Methods 0.000 abstract description 17
- 238000000034 method Methods 0.000 abstract description 13
- 210000002569 neuron Anatomy 0.000 abstract description 7
- 238000004458 analytical method Methods 0.000 abstract description 5
- 230000002159 abnormal effect Effects 0.000 abstract description 4
- 238000001356 surgical procedure Methods 0.000 abstract description 4
- 210000000056 organ Anatomy 0.000 abstract description 3
- JLNTWVDSQRNWFU-UHFFFAOYSA-N OOOOOOO Chemical compound OOOOOOO JLNTWVDSQRNWFU-UHFFFAOYSA-N 0.000 abstract description 2
- 210000004027 cell Anatomy 0.000 abstract description 2
- 210000003169 central nervous system Anatomy 0.000 abstract description 2
- 230000004222 uncontrolled growth Effects 0.000 abstract description 2
- 238000013473 artificial intelligence Methods 0.000 description 5
- 238000012360 testing method Methods 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 240000007594 Oryza sativa Species 0.000 description 1
- 235000007164 Oryza sativa Nutrition 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 230000010261 cell growth Effects 0.000 description 1
- 238000002591 computed tomography Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 235000009566 rice Nutrition 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/05—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves
- A61B5/055—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/40—Detecting, measuring or recording for evaluating the nervous system
- A61B5/4058—Detecting, measuring or recording for evaluating the nervous system for evaluating the central nervous system
- A61B5/4064—Evaluating the brain
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/03—Recognition of patterns in medical or anatomical images
- G06V2201/031—Recognition of patterns in medical or anatomical images of internal organs
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- Public Health (AREA)
- Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Veterinary Medicine (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Neurology (AREA)
- Heart & Thoracic Surgery (AREA)
- Artificial Intelligence (AREA)
- Physiology (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Mathematical Physics (AREA)
- Fuzzy Systems (AREA)
- Evolutionary Computation (AREA)
- Psychology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Neurosurgery (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- High Energy & Nuclear Physics (AREA)
- Psychiatry (AREA)
- Radiology & Medical Imaging (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
AN AUTOMATIC TUMOR DETECTION SYSTEM BASED ON LOCAL
LINEAR WAVELET ARTIFICIAL NEURAL NETWORK WITH
HYBRID OPTIMIZATION
ABSTRACT
The Human Brain is most complex, complicated organ with billions of neurons and
command control of the Central Nervous System. The Human Brain contains neuron
and non-neuron cells, abnormal and uncontrolled growth of these cells indicates tumors.
The Magnetic Resonance Imaging (MRI) is performed to acquire Human Brain Images
for analysis and classification of the tumor present by the radiologist. The detection of
the Tumors in brain by the radiologist with MR Images is attentive and time taken
process. An Artificial Neural Network facilitates Detection, Segmentation,
Classification and Extraction of the Tumor area automatically which is necessary for
performing tumor removal surgery. The present invention disclosed herein is an
Automatic Tumor Detection System based on Local Linear Wavelet Artificial Neural
Network with Hybrid Optimization comprising of MRI Dataset (201), Preprocessing
(202), Segmentation (203), Box Bounding (204), GLCM-LBP Feature Extraction (205),
Hybrid Firefly Optimization (206), Local Linear Wavelet ANN (207), and Performance
Metrics; can detect human brain tumors automatically using Artificial Neural Network.
In the present invention disclosed, the Enhanced FCM with Cluster Map is used in
Segmentation, GLCM-LBP is used as features extraction, the firefly features
optimization is to select optimized features and Local Linear Wavelet ANN for
classification. The present invention disclosed here can able detect the multiple tumors
automatically and classify the tumors as Malignant or non- Malignant tumors. The
present invention can select deterministic and optimized features, decomposes the
optimized features set into small local wavelets for classification. The present invention
shows better performance in the form of Classification Accuracy of 99.52%, Sensitivity
of 98.65%, and Specificity of 1.0, the system can able to detect the tumor automatically
in 9.82352 Seconds.
1/2
AN AUTOMATIC TUMOR DETECTION SYSTEM BASED ON LOCAL
LINEAR WAVELET ARTIFICIAL NEURAL NETWORK WITH
HYBRID OPTIMIZATION
DRAWINGS
101 102 10
BRAIN MR fre-'rocessinA;
IMAGE Automatic Se6mentation;
MR IMAGE TumorIdentificationS stem FeatureSelection;
-CUII\O Feature Extraction;
STIO 104 Featurenhancement;
SYSTEMFeatureClassification
- Operations
OOOOOOO
Practitioner 10
Figure 1: General Automatic Tumor Identification System
20120 V r i Frr
MRI DATASET PREPROCESSING SEGMENTATION BOX BOUNDING
PERFORMANCE LOCAL LINEAR HYBRID FIREFLY LIGLCM-LBP FEATURE
METRICS WAVELET ANN OPTIMIZATION EXTRACTION
Figure 2: An Automatic Tumor Detection System based on Local Linear Wavelet
Artificial Neural Network with Hybrid Optimization.
Description
1/2
101 102 10
BRAIN MR fre-'rocessinA; IMAGE Automatic Se6mentation; MR IMAGE TumorIdentificationS stem FeatureSelection; -CUII\O Feature Extraction; STIO 104 Featurenhancement; SYSTEMFeatureClassification
- Operations OOOOOOO
Practitioner 10
Figure 1: General Automatic Tumor Identification System
20120
MRI DATASET V r iFrr PREPROCESSING SEGMENTATION BOX BOUNDING
Figure 2: An Automatic Tumor Detection System based on Local Linear Wavelet Artificial Neural Network with Hybrid Optimization.
[0001] The present invention relates to the technical field of Artificial Intelligence.
[0002] Particularly, the present invention is related to an Automatic Tumor Detection System based on Local Linear Wavelet Artificial Neural Network with Hybrid Optimization of the broader field of Biomedical Image Processing in Artificial Intelligence.
[0003] More particularly, the present invention is relates to an Automatic Tumor Detection System based on Local Linear Wavelet Artificial Neural Network with Hybrid Optimization can detect the human brain tumors automatically from the MRI images. The system can able detect the multiple tumors automatically and classify the tumors as Malignant or non- Malignant tumors with the help of Local Linear Wavelet Artificial Neural Network (LLW-ANN).
[0004] The Computer Tomography is the one of the non-invasive method of acquiring the Brain Tumor Images, used as preliminary method to find the tumors present in the human brain. The tumor is an abnormal cell growth occurs anywhere in the human body, it should be detected and treated otherwise it may threaten the life of the patient. When the Tumors are located in the human brain, further analysis is required to perform the surgical removal. The advanced method of acquiring the Brain Tumor Images is Magnetic Resonance Imaging (MRI). The MRI scanning is the further option available to detect the Tumors present in the Human Brain. The Human brain is sensitive and sophisticated organ; even a small abnormal tissue in it can shows degradation of the human activity leads to loss of life. The accurate detection methods are required to identify and classify the brain tumors.
[0005] Generally, the patient suspected having tumor will go for acquiring the MRI scan images referred by the Practitioner. The practitioner or the radiologist will acquire the MR (Magnetic Resonance) images, analyzed to determine the tumors present. The adequate knowledge is required to the Practitioner to determine the present of tumors, sometimes may be misinterpretation occurs in deciding the tumors present in the human brain. The latest advancements in the Artificial Intelligence brings the Neural Networks works same as human brain neurons, can be trained and used for testing the tumors present automatically to reduce the human effort. The Human Brain is most complex, complicated organ with billions of neurons and command control of the Central Nervous System. The Human Brain contains neuron and non-neuron cells, abnormal and uncontrolled growth of these cells indicates tumors. The Magnetic Resonance Imaging (MRI) is performed to acquire Human Brain Images for analysis and classification of the tumor present by the radiologist. The detection of the Tumors in brain by the radiologist with MR Images is attentive and time taken process. An Artificial Neural Network facilitates Detection, Segmentation, Classification and Extraction of the Tumor area automatically which is necessary for performing tumor removal surgery.
[0006] The existing inventions can detect the brain tumors with less performance, more detection time, less number of the tumors in an image. There is a need to develop a new novel system that can detect the Brain Tumors present in the MR images automatically, required a comfortable Graphical User Interface (GUI) to be operated even by the normal practitioner without that much huge knowledge. The present invention disclosed herein facilitates the easy detection with more accuracy, can detect multiple tumors, and takes less time to detect and classify. The present invention disclosed herein is having the applications such as Health Care Hospitals, Object detection, and object recognition.
[0007] The MRI Machine acquired Brain Images to be analyzed by the Practitioner
(103) or by the radiologist to detect the brain tumors present if any. If brain tumors were present then the Practitioner (103) or Radiologist should classify the detected tumor by making further analysis to perform the surgical removal. Early detection is important if it is the Malignant, Practitioner (103) or Radiologist should have adequate knowledge to classify and locate the tumor. In contrast to this situation, an Artificial Intelligence facilitates the automatic detection and classification of the brain tumors from the acquired Brain MR images. The main embodiments of the present disclosure, Referring to Figure 2, illustrates the embodiments of the disclosure that is an Automatic Tumor Detection System based on Local Linear Wavelet Artificial Neural Network with Hybrid Optimization comprising of MRI Dataset (201), Preprocessing (202), Segmentation (203), Box Bounding (204), GLCM-LBP Feature Extraction (205), Hybrid Firefly Optimization (206), Local Linear Wavelet ANN (207), and Performance Metrics (208); can detect human brain tumors automatically using Artificial Neural Network. The present system can detect the brain tumors from the brain MR images available publicly. Initially if any sort of noise and discontinuities present in the input MR images are preprocessed. In preprocessing, Fuzzy weighted- Non Local Means (FW-NLM) filter is used by keeping the Rician and Gaussian noises in minds which are most commonly occurring noises while capturing the MR images. The Segmentation is performed with Enhanced Fuzzy C Mean (EFCM) and Cluster Feature Map. The Segmented MR image shows clearly the spatial location and the location is box binding to understand the tumor region. The features are extracted with the statistical features selector and visual extractors such as GLCM and LBP in the present invention. The feature vectors generated in the present invention are optimized with the Hybrid Firefly Optimization and then tumors are classified by the Local Linear Wavelet ANN. The present invention disclosed here can able detect the multiple tumors automatically and classify the tumors as Malignant or non- Malignant tumors. The present invention can select deterministic and optimized features, decomposes the optimized features set into small local wavelets for classification. The present invention shows better performance in the form of Classification Accuracy of 99.52%, Sensitivity of 98.65%, and Specificity of 1.0, the system can able to detect the tumor automatically in 9.82352 Seconds.
[0008] The Summary of the Invention, as well as the attached sketches and the Detailed Description of the Invention, describe the present invention in various levels of detail, and the inclusion or omission of components, sections, or other things in this Summary of the Invention is not intended to limit the scope of the present disclosure. The summary of the Invention can be read with the detailed description for better understanding of the current disclosure.
[0009] The accompanying drawings are incorporated into and constitute part of this specification to help you better comprehend the innovation. When read in connection with the explanation, the drawing displays exemplary embodiments of the present disclosure and aids understanding of the disclosure's concepts. The drawings are for illustration purposes only and are not intended to limit the scope of the disclosure. The usage of the same reference numerals demonstrates that the elements are comparable but not identical. On the other side, different reference numerals might be used to define related components. Some embodiments might lack such elements and/or components, whereas others may use elements or components not depicted in the sketches.
[0010] Referring to Figure 1, illustrates General Automatic Tumor Identification System comprising of MR Image Acquisition System (101), Brain MR Image (102), Practitioner (103), Automatic Tumor Identification System (104), and Operations (105), in accordance with an exemplary embodiment of the present disclosure to understand the General Automatic Tumor Identification System with set of its operations to detect tumor. This drawing is considered to understand how the generally Tumor are identified by the practitioner and automatic system with machine learning operations, the invention is not limited to this drawing, and this illustration is given to aid comprehension of the disclosure and should not be construed as limiting the disclosure's breadth, scope, or applicability.
[0011] Referring to Figure 2, illustrates the present invention and main embodiment of current disclosure that is an Automatic Tumor Detection System based on Local Linear Wavelet Artificial Neural Network with Hybrid Optimization comprising of MRI
Dataset (201), Preprocessing (202), Segmentation (203), Box Bounding (204), GLCM LBP Feature Extraction (205), Hybrid Firefly Optimization (206), Local Linear Wavelet ANN (207), and Performance Metrics (208); can detect human brain tumors automatically using Artificial Neural Network, in accordance with main exemplary embodiment of the present disclosure, accompanied drawing and this drawing is provided to facilitate knowledge of the present disclosure; however, some aspects and/or components may not be present in embodiments, and others may be used in different forms than those indicated in the sketches. The usage of single language to describe a component or element might encompass a plural number of such components or elements, depending on the context, and vice versa.
[0012] Referring to Figure 3, illustrated with images showing Output of Segmentation with bounding box, in accordance with another exemplary embodiment of the present disclosure to understand segmentation of the present disclosure. This drawing is considered to understand the segmentation process used in the invention disclosed herein, the invention is not limited only to this drawing, and this illustration is provided to assist comprehension of the disclosure and should not be construed as restricting the depth, nature, or applicability of the disclosure.
[0013] Referring to Figure 4, illustrates the GUI of proposed Automatic Tumor Detection System, in accordance with another exemplary embodiment of the present disclosure to understand performance of the present disclosure. This Graphical User Interface (GUI) is considered to understand the present system and facilitates the easy operations, the invention is not limited only to this drawing as some of the other performance metrics can be added further, and this illustration is provided to assist comprehension of the disclosure and should not be construed as restricting the depth, nature, or applicability of the disclosure.
[0014] The invention will become more well-known as a result of the following extensive discussion, and objects other than those listed below will become clear. The annexed drawings are referred to in this description. When the following detailed explanation of the invention is considered, the invention will be better known, and objects other than those set forth above will become obvious. The invention associated drawings are referred to in this description. The present disclosure's Embodiments will now be identified using the corresponding drawings as a guide. Embodiments are given so that a skilled in the art can fully comprehend the current disclosure. Several specifics, relating to various components and processes, are set out to provide a complete understanding of embodiments of the present disclosure. The information given in the embodiments should not be construed to restrict the scope of the present disclosure, as will be evident to those skilled in the art. The order of steps revealed in the procedure and process of this disclosure should not be construed as requiring that they be performed in the order defined or illustrated. It should also be noted that additional or alternate steps should be taken.
[0015] Referring to Figure 1, illustrates General Automatic Tumor Identification System comprising of MR Image Acquisition System (101), Brain MR Image (102), Practitioner (103), Automatic Tumor Identification System (104), and Operations (105), in accordance with an exemplary embodiment of the present disclosure to understand the General Automatic Tumor Identification System with set of its operations to detect tumor. The MR Image Acquisition System (101) is an MRI machine generally used to acquire the patient's brain images. This MR Image Acquisition System (101) uses Magnetic rays with strong magnetic fields with its powerful magnets to capture the all the human body parts. In the present invention disclosed herein only Brain MR Images (102) of human Brain are considered. These acquired Brain MR Images (102) are analyzed by the healthcare practitioner (103) manually by the knowledge they have or radiologist can perform manual analysis. The time taken for identifying the tumors by the practitioner (103) or by the radiologist (103) depends on how they are expertise. The huge experience is required to detect or identify the tumors from the MR images by the MRI scan by the practitioner (103) or radiologist (103). In contradiction, the Automatic Tumor Identification System (104) can detect or identify the brain tumors from the MR images in less time with the set operations of machine learning in artificial intelligence. The operations (105) include filter preprocessing, segmentation operations for feature extraction, feature selection, feature enhancement, and classification.
[0016] The main embodiments of the present disclosure, Referring to Figure 2, illustrates the embodiments of the disclosure that is an Automatic Tumor Detection System based on Local Linear Wavelet Artificial Neural Network with Hybrid Optimization comprising of MRI Dataset (201), Preprocessing (202), Segmentation (203), Box Bounding (204), GLCM-LBP Feature Extraction (205), Hybrid Firefly Optimization (206), Local Linear Wavelet ANN (207), and Performance Metrics (208); can detect human brain tumors automatically using Artificial Neural Network. To provide a full understanding of embodiments of the present disclosure, numerous information relating to particular components and methods are provided. The information given in the embodiments should not be construed to restrict the scope of the present disclosure, as will be evident to those skilled in the art.
[0017] The MRI Dataset (201) contains MR images are taken from the publicly available datasets such as BRATS 2015, UCI Dataset and ADNI. Total of 275 images were taken with different graded brain MRI images. Most preferably images having multiple tumors are tested with the present system of the disclosure. In an aspect, Preprocessing (202) is to filter the MR images to remove unwanted noise present, mostly Rician and Gaussian noise are present in the acquired images and are to be removed by preprocessing (202). The Rician noise has the rice distribution with zero mean value; Gaussian noise is with Gaussian distribution with zero mean. The Fuzzy weighted- Non Local Means (FW-NLM) filter is used to remove the Rician and Gaussian noises present in the MR images in the present invention disclosed herein. Based on the Gaussian fuzzy membership values and intensity distances between the pixels, the Fuzzy weighted- Non Local Means (FW-NLM) filter averages the similar pixels. The patch and pattern redundancy are considered while averaging the similar pixels. The Segmentation (203) is performed in the present invention disclosed herein with the Enhanced FCM with Cluster Map. The Enhanced FCM with Cluster Map is the combination of the Enhanced Fuzzy C-Mean (EFCM) and Cluster Feature Map. This hybrid combination effectively segmenting the MR images tested on the present invention. The Enhanced Fuzzy C-Mean (EFCM) enhances the similarity measurement between the pixel intensities and with neighbourhood center it forms the cluster. In another aspect, this Enhanced Fuzzy C-Mean (EFCM) corrects the intensity inhomogeneity present in the MR Images. Another aspect of the present invention is Cluster Feature Map in segmentation (203) reduces the complexity in the MR data input. The Cluster Feature Map in segmentation (203) selects the deterministic features of the tumors; maximum valued features have been selected from all selected features by sorting them. The Cluster Feature Map is constructed with the maximum valued features; the Enhanced FCM with Cluster Map can segment the tumor intensities which are complex. The Box Bounding (204) after the Segmentation (203) is performed in the present disclosure to locate the tumor spatial location on the segmented MR image. The Box Bounding (204) facilitates the easy tumor detection with rectangular boxes on the Tumor areas. The GLCM (Gray Level Co-occurrence Matrix) and LBP (Local Binary Pattern) Feature Extraction (205) is used in the present invention for extracting the features. The statistical textural features are extracted in specified spatial relationship of the pixels by the GLCM; it measures the textural pixels pairs to know the similarity, intensity, Homogeneity, Correlation, Mean, Entropy and other features. The LBP labels the pixels features by thresholding and acts as visual descriptor for the further features selection. The GLCM-LBP Feature Extraction (205) generates the feature vectors. These feature vectors are optimized by the Hybrid Firefly Optimization (206). The Hybrid Firefly Optimization (206) updates the particles, nearest particles distances are measured to obtain the optimized features for improving the tumor detection in the present invention disclosed herein. The Local Linear Wavelet ANN (207) is an Artificial Neural Network constructed with different layers. This network in the present invention has one input layer, seven hidden layers, one output layer, and 3-level wavelet decomposition to select optimal wavelets. The Local Linear Wavelet ANN (207) decomposes the features into smaller wavelets by locally considering smaller wavelets. Due to the Local Linear Wavelet ANN (207), the present system can train and test the MR Images and then classify them into Malignant or non- Malignant tumors. The Performance Metrics (208) are evaluated with the system disclosed herein, showing the Classification Accuracy of 99.52%, Sensitivity of 98.65%, and Specificity of 1.0, the system can able to detect the tumor automatically in 9.82352 Seconds. The present invention disclosed herein can able detect the multiple tumors automatically and classify the tumors as Malignant or non- Malignant tumors. The present invention can select deterministic and optimized features, decomposes the optimized features set into small local wavelets for classification.
[0018] Referring to Figure 3, illustrated with images showing Output of Segmentation with bounding box, is another exemplary embodiment of the present disclosure to understand segmentation of the present disclosure. These visualized images are the qualitative output of the segmentation process used in the present disclosure. The Enhanced Fuzzy C-Mean (EFCM), Cluster Feature Map and Box Bounding produce these qualitative results for the segmentation. These drawing are considered to understand the segmentation process used in the invention disclosed herein, the invention is not limited only to this drawing, and this illustration is provided to assist comprehension of the disclosure and should not be construed as restricting the depth, nature, or applicability of the disclosure.
[0019] Referring to Figure 4, illustrates the GUI (Graphical User Interface) of proposed Automatic Tumor Detection System, in accordance with another exemplary embodiment of the present disclosure to understand performance of the present disclosure. The GUI shown in the Figure 4 can be loaded with any brain MR image for testing and detecting the brain tumor. All the features extracted and performance metrics are instantly displayed to know the testing and detection capacity of the present disclosure. This Graphical User Interface (GUI) is considered to understand the present system and facilitates the easy operations, the invention is not limited only to this drawing as some of the other performance metrics can be added further, and this illustration is provided to assist comprehension of the disclosure and should not be construed as restricting the depth, nature, or applicability of the disclosure.
[0020] Several specific details are set out in the following exemplary explanation in order to provide a more detailed understanding of embodiments of the invention. An artisan of ordinary skill, on the other hand, might notice that the existing innovation can be practiced without integrating any of the specific information mentioned herein. The main embodiments of the present disclosure are considered with the correct classification with automatic tumor detection. The subsequent description gives the details about the features optimization and performance evaluation of the system. To detect the brain tumors, the method and the way of the present embodiment is provided in the above layout and it shall not limit the scope of the present disclosure.
Claims (5)
1. An Automatic Tumor Detection System based on Local Linear Wavelet Artificial Neural Network with Hybrid Optimization comprising of MRI Dataset (201), Preprocessing (202), Segmentation (203), Box Bounding (204), GLCM-LBP Feature Extraction (205), Hybrid Firefly Optimization (206), Local Linear Wavelet ANN (207), and Performance Metrics; can detect human brain tumors automatically using Artificial Neural Network.
2. An Automatic Tumor Detection System based on Local Linear Wavelet Artificial Neural Network with Hybrid Optimization as claimed in claim 1, wherein the Fuzzy weighted- Non Local Means (FW-NLM) filter is used to remove the Rician and Gaussian noises present in the MR images, based on the Gaussian fuzzy membership values and intensity distances between the pixels, the Fuzzy weighted- Non Local Means (FW-NLM) filter averages the similar pixels.
3. An Automatic Tumor Detection System based on Local Linear Wavelet Artificial Neural Network with Hybrid Optimization as claimed in claim 1, wherein the combination of the Enhanced Fuzzy C-Mean (EFCM) and Cluster Feature Map is used in the Segmentation. The Cluster Feature Map in segmentation (203) selects the deterministic features of the tumors; maximum valued features have been selected from all selected features by sorting them. The Cluster Feature Map is constructed with the maximum valued features.
4. An Automatic Tumor Detection System based on Local Linear Wavelet Artificial Neural Network with Hybrid Optimization as claimed in claim 1, wherein the GLCM-LBP Feature Extraction (205) generates the feature vectors. These feature vectors are optimized by the Hybrid Firefly Optimization (206). The Hybrid Firefly
Optimization (206) updates the particles, nearest particles distances are measured to obtain the optimized features for improving the tumor detection.
5. An Automatic Tumor Detection System based on Local Linear Wavelet Artificial Neural Network with Hybrid Optimization as claimed in claim 1, wherein the system can able detect the multiple tumors automatically and classify the tumors as Malignant or non- Malignant tumors. The present invention shows better performance in the form of Classification Accuracy of 99.52%, Sensitivity of 98.65%, and Specificity of 1.0, the system can able to detect the tumor automatically in 9.82352 Seconds.
1/2
AN AUTOMATIC TUMOR DETECTION SYSTEM BASED ON LOCAL LINEAR WAVELET ARTIFICIAL NEURAL NETWORK WITH HYBRID OPTIMIZATION
DRAWINGS 2021103132
Figure 1: General Automatic Tumor Identification System
Figure 2: An Automatic Tumor Detection System based on Local Linear Wavelet Artificial Neural Network with Hybrid Optimization.
2021103132 2/2
Figure 3: Output of Segmentation with bounding box.
Figure 4: GUI of proposed Automatic Tumor Detection System
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| AU2021103132A AU2021103132A4 (en) | 2021-06-05 | 2021-06-05 | An automatic tumor detection system based on local linear wavelet artificial neural network with hybrid optimization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| AU2021103132A AU2021103132A4 (en) | 2021-06-05 | 2021-06-05 | An automatic tumor detection system based on local linear wavelet artificial neural network with hybrid optimization |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| AU2021103132A4 true AU2021103132A4 (en) | 2021-10-07 |
Family
ID=77922891
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| AU2021103132A Ceased AU2021103132A4 (en) | 2021-06-05 | 2021-06-05 | An automatic tumor detection system based on local linear wavelet artificial neural network with hybrid optimization |
Country Status (1)
| Country | Link |
|---|---|
| AU (1) | AU2021103132A4 (en) |
-
2021
- 2021-06-05 AU AU2021103132A patent/AU2021103132A4/en not_active Ceased
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kaldera et al. | Brain tumor classification and segmentation using faster R-CNN | |
| Akbar et al. | Automated techniques for blood vessels segmentation through fundus retinal images: A review | |
| Akram et al. | Detection of neovascularization in retinal images using multivariate m-Mediods based classifier | |
| CN111028206A (en) | Prostate cancer automatic detection and classification system based on deep learning | |
| Nigri et al. | Explainable deep CNNs for MRI-based diagnosis of Alzheimer’s disease | |
| Coupé et al. | LesionBrain: an online tool for white matter lesion segmentation | |
| EP2201525B1 (en) | Visualization of temporal data | |
| CN105787918B (en) | Analysis and processing method for magnetic resonance image of acute ischemic stroke | |
| Archa et al. | Segmentation of brain tumor in MRI images using CNN with edge detection | |
| US20200074631A1 (en) | Systems And Methods For Identifying Implanted Medical Devices | |
| Di Leo et al. | An improved procedure for the automatic detection of dermoscopic structures in digital ELM images of skin lesions | |
| CN109949288A (en) | Tumor type determines system, method and storage medium | |
| Suwalska et al. | CMB-HUNT: automatic detection of cerebral microbleeds using a deep neural network | |
| Viswanath et al. | Empirical evaluation of bias field correction algorithms for computer-aided detection of prostate cancer on T2w MRI | |
| AU2021103132A4 (en) | An automatic tumor detection system based on local linear wavelet artificial neural network with hybrid optimization | |
| Tara | Comparison among Supervised Classifiers for Classification of Brain Tumor | |
| Rode et al. | Image Segmentation with Priority Based Apposite Feature Extraction Model for Detection of Multiple Sclerosis in MR Images Using Deep Learning Technique. | |
| Nugroho et al. | Quad Convolutional Layers (QCL) CNN Approach for Classification of Brain Stroke in Diffusion Weighted (DW)-Magnetic Resonance Images (MRI). | |
| Heyi et al. | Development of a retinal image segmentation algorithm for the identifying prevalence markers of diabetic retinopathy using a neural network | |
| Sathya et al. | Development of CAD system based on enhanced clustering based segmentation algorithm for detection of masses in breast DCE-MRI | |
| Qu et al. | Positive unanimous voting algorithm for focal cortical dysplasia detection on magnetic resonance image | |
| Verma et al. | Detection of Brain Tumor in MR Images using Multi-objective Particle Swarm Optimization with Support Vector Machine Classifier with RBF kernel | |
| Fenneteau et al. | A size-adaptative segmentation method for better detection of multiple sclerosis lesions | |
| Chaphekarande et al. | Machine learning based brain mri estimation method | |
| Kochar | A Survey on Brain Tumor Detection and Classification System based on Artificial Neural Network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| FGI | Letters patent sealed or granted (innovation patent) | ||
| MK22 | Patent ceased section 143a(d), or expired - non payment of renewal fee or expiry |