AU2021101851A4 - A deep learning based system for automatic segmentation and quantification of covid-19 in CT images - Google Patents
A deep learning based system for automatic segmentation and quantification of covid-19 in CT images Download PDFInfo
- Publication number
- AU2021101851A4 AU2021101851A4 AU2021101851A AU2021101851A AU2021101851A4 AU 2021101851 A4 AU2021101851 A4 AU 2021101851A4 AU 2021101851 A AU2021101851 A AU 2021101851A AU 2021101851 A AU2021101851 A AU 2021101851A AU 2021101851 A4 AU2021101851 A4 AU 2021101851A4
- Authority
- AU
- Australia
- Prior art keywords
- images
- covid
- deep learning
- quantification
- based system
- 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
- 208000025721 COVID-19 Diseases 0.000 title claims abstract description 21
- 238000013135 deep learning Methods 0.000 title claims abstract description 16
- 238000011002 quantification Methods 0.000 title claims abstract description 11
- 230000011218 segmentation Effects 0.000 title claims abstract description 11
- 238000012549 training Methods 0.000 claims abstract description 6
- 238000012360 testing method Methods 0.000 claims abstract description 3
- 210000004072 lung Anatomy 0.000 claims description 15
- 238000002591 computed tomography Methods 0.000 claims description 8
- 241000711573 Coronaviridae Species 0.000 claims description 3
- 238000013136 deep learning model Methods 0.000 abstract 1
- 238000000034 method Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 238000012216 screening Methods 0.000 description 3
- 238000013526 transfer learning Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000000241 respiratory effect Effects 0.000 description 2
- 238000010200 validation analysis Methods 0.000 description 2
- 241000700605 Viruses Species 0.000 description 1
- 210000004712 air sac Anatomy 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 239000012530 fluid Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000003757 reverse transcription PCR Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
- G06T7/0014—Biomedical image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Quality & Reliability (AREA)
- Radiology & Medical Imaging (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
A deep learning based system for automatic segmentation and quantification of covid-19 in
CT images
The present invention relates to a deep learning based system for automatic segmentation and
quantification of covid-19 in CT images. This invention uses pre-trained deep learning models.
These pre-trained models have to be retrained on the new dataset to do the prediction. Once the
training and testing are done on these datasets then we can use these newly trained models, pass
new images to them and do the predictions. The final result would notify us whether the image is
from an infected class or a non-infected class.
1/3
"Pv w~iltoo ^ 1% Ill.OM 10 10
Fiur
Figre
Fi 0re4
Description
1/3
"Pv w~iltoo ^ 1% Ill.OM 10 10
Fiur
Figre
Fi 0re4
A deep learning based system for automatic segmentation and quantification of covid-19 in CT images
Technical field of invention:
Present invention in general relates to field of information technology and more particularly a deep learning based system for automatic segmentation and quantification of covid-19 in CT images.
Background ofthe invention:
The background information herein below relates to the present disclosure but is not necessarily prior art.
Currently, there are no machines available that could tackle the problem of mass screening of COVID-19 infected CT images of lungs, but, In the studies of Shan F[1], he has mentioned that without any assistance of deep learning it takes 211.3±52.6 minutes to contour COVID 19 infection regions on one CT scan. So, by using transfer learning models the required time would be drastically reduced.
.0 Objective of the invention
An objective of the present invention is to attempt to overcome the problems of the prior art and provide a deep learning based system for automatic segmentation and quantification of covid-19 in CT images.
These and other objects and characteristics of the present invention will become apparent from the further disclosure to be made in the detailed description given below.
Summary of the invention:
Accordingly following invention provides a deep learning based system for automatic segmentation and quantification of covid-19 in CT images. In the present scenario, there is no technology to help medical professionals to do a mass screening. With the help of this technology, there would be no physical contact as well as social distancing norms would be followed. Medical professionals would not require to go near the patient like they do while collecting samples for the RT-PCR technique. For this invention, all we need is the CT machine to do the CT scanning of lungs and a Computer to read/process the image and predict the results. Once the image is scanned, read, and processed then it will be passed to our transfer learning model to do the prediction.
Brief description of drawing:
This invention is described by way of example with reference to the following drawing where,
Figure 1 of sheet 1 illustrates accuracy & loss graph of VGG-16 of dataset No: 1. Figure 2 of sheet 1 illustrates accuracy & loss graph of Inceptionv3 of dataset No: 1. Figure 3 of sheet 1 illustrates Accuracy & loss graph of Xception of dataset No: 1. Figure 4 of sheet 2 illustrates Accuracy & loss graph of VGG-19 of dataset No: 1. Figure 5 of sheet 2 illustrates Accuracy & loss graph of VGG-16 of dataset No: 2. Figure 6 of sheet 2 illustrates Accuracy & loss graphs of Inceptionv3 of dataset No:2. Figure 7 of sheet 3 illustrates Accuracy & loss graphs of Xception of dataset No:2. Figure 7 of sheet 3 illustrates Accuracy & loss graph of VGG-19of dataset No:2.
In order that the manner in which the above-cited and other advantages and objects of the invention are obtained, a more particular description of the invention briefly described above will be referred, which are illustrated in the appended drawing. Understanding that these drawing depict only typical embodiment of the invention and therefore not to be considered limiting on its scope, the invention will be described with additional specificity and details through the use of the accompanying drawing.
Detailed description of the invention:
The present invention relates to a deep learning based system for automatic segmentation and quantification of covid-19 in CT images. SAARS-COV-2 is a respiratory illness caused by the novel Coronavirus (COVID-19) disease. The virus goes into the lungs through the respiratory tracks and damages the walls and linings of the air sacs in our lungs, as our body tries to fight it, our lungs become more inflamed and fill with fluid. This makes it harder to breathe. So, at the early stages, deep learning applications can be used for screening and prediction at a rapid rate for diagnosing the lungs of patients. This paper uses Transfer learning methods. Four pre-trained models were used in this study - VGG-16, VGG-19, Inceptionv3, Xception.
This invention addresses challenges while using pre-trained models in the real world. Also, high accuracies were achieved on these models.
Material:
The training and test data were fetched from www.Kaggle.com. The data is publicly available for everyone to use. These datasets contain Computed tomography images of lungs of both normal as well as COVID-19 infected lungs of patients. There were two datasets used in this study. The first dataset consists of 599 images, out of which 299 belong to COVID-19 infected cases, and the rest were normal. The second dataset consists of 2481 images, which belong to the 1252 Covid-19 infected images.
Methods:
In this study, we have used pre-trained deep learning models- VGG16, VGG-19, Xception, InceptionV3. These models helped to classify images of lungs infected by coronavirus against normal images. We have mentioned VGG-16 Architecture because for this problem statement we got the best classification result from the VGG-16 model (training accuracy of 99.00% for dataset 1 was achieved). And this model can further be improvised to get higher classification accuracy.
TABLE 1: DATASET Dataset Number of COVID-19 infected CT Images Number of Normal CT Images of oflungs lungs 1 299 300 2 1252 1229
TABLE 2: ACCURACY AND VALIDATION RESULT OF DATASET NO. 1 Dataset 1 SR.No Deep learning Training Validation models Accuracy Accuracy 1 VGG16 99.00% 99.50% 2 InceptionV3 97.50% 100.00% 3 Xception 96.16% 97.33% 4 VGG19 95.99% 98.83%
Best method of performance of the invention:
Step1: Get the CT scan image of lungs from CT scanning machine Step2: Pass the CT image to the afore mentioned models. Step 3: Do the prediction Step 4: Result is displayed
The many features and advantages of the invention are apparent from the detailed specification, and thus, it is intended by the appended claims to cover all such features and advantages of the invention which fall within the true spirit and scope of the invention. Further, since numerous modifications and variations will readily occur to those skilled in the art, it is not desired to limit the invention to the exact construction and operation illustrated and described, and accordingly, all suitable modifications and equivalents may be resorted to, falling within the scope of the invention.
Claims (3)
1. A deep learning based system for automatic segmentation and quantification of covid 19 in CT images wherein the training and test data fetched; these datasets contain Computed tomography images of lungs of both normal as well as COVID-19 infected lungs of patients; there were two datasets used in this study; the first dataset consists of 599 images, out of which 299 belong to COVID-19 infected cases, and the rest were normal; the second dataset consists of 2481 images, which belong to the 1252 Covid-19 infected images.
2. The deep learning based system for automatic segmentation and quantification of covid-19 in CT images as claimed in claim 1 wherein have used pre-trained deep learning models- VGG16, VGG-19, Xception, InceptionV3; these models helped to classify images of lungs infected by corona virus against normal images have mentioned VGG-16 Architecture because for this problem statement we got the best classification result from the VGG-16 model (training accuracy of 99.00% for dataset 1 was achieved). And this model can further be improvised to get higher classification accuracy.
3. The deep learning based system for automatic segmentation and quantification of covid-19 in CT images as claimed in claim 1, comprises of following steps; a) Get the CT scan image of lungs from CT scanning machine; b) Pass the CT image to the afore mentioned models; c) Do the prediction; d) Result is displayed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
AU2021101851A AU2021101851A4 (en) | 2021-04-12 | 2021-04-12 | A deep learning based system for automatic segmentation and quantification of covid-19 in CT images |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
AU2021101851A AU2021101851A4 (en) | 2021-04-12 | 2021-04-12 | A deep learning based system for automatic segmentation and quantification of covid-19 in CT images |
Publications (1)
Publication Number | Publication Date |
---|---|
AU2021101851A4 true AU2021101851A4 (en) | 2021-06-03 |
Family
ID=76132814
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
AU2021101851A Ceased AU2021101851A4 (en) | 2021-04-12 | 2021-04-12 | A deep learning based system for automatic segmentation and quantification of covid-19 in CT images |
Country Status (1)
Country | Link |
---|---|
AU (1) | AU2021101851A4 (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114092819A (en) * | 2022-01-19 | 2022-02-25 | 成都四方伟业软件股份有限公司 | Image classification method and device |
CN116205289A (en) * | 2023-05-05 | 2023-06-02 | 海杰亚(北京)医疗器械有限公司 | Animal organ segmentation model training method, segmentation method and related products |
-
2021
- 2021-04-12 AU AU2021101851A patent/AU2021101851A4/en not_active Ceased
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114092819A (en) * | 2022-01-19 | 2022-02-25 | 成都四方伟业软件股份有限公司 | Image classification method and device |
CN116205289A (en) * | 2023-05-05 | 2023-06-02 | 海杰亚(北京)医疗器械有限公司 | Animal organ segmentation model training method, segmentation method and related products |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Öztürk et al. | Classification of Coronavirus (COVID‐19) from X‐ray and CT images using shrunken features | |
Umer et al. | COVINet: a convolutional neural network approach for predicting COVID-19 from chest X-ray images | |
Pandit et al. | Automatic detection of COVID-19 from chest radiographs using deep learning | |
Kumar et al. | Novel deep transfer learning model for COVID-19 patient detection using X-ray chest images | |
Ravi et al. | A multichannel EfficientNet deep learning-based stacking ensemble approach for lung disease detection using chest X-ray images | |
Meraj et al. | Detection of pulmonary tuberculosis manifestation in chest X-rays using different convolutional neural network (CNN) models | |
Shambhu et al. | Binary classification of covid-19 ct images using cnn: Covid diagnosis using ct | |
AU2021101851A4 (en) | A deep learning based system for automatic segmentation and quantification of covid-19 in CT images | |
Vieira et al. | Detecting pulmonary diseases using deep features in X-ray images | |
Osman et al. | SOM-LWL method for identification of COVID-19 on chest X-rays | |
Pant et al. | Pneumonia detection: An efficient approach using deep learning | |
US11157697B2 (en) | Method and apparatus for extracting diagnosis object from medical document | |
Mangeri et al. | Chest diseases prediction from X-ray images using CNN models: a study | |
Kim et al. | Evaluation of deep learning for COVID‐19 diagnosis: impact of image dataset organization | |
Rehman Khan et al. | Cloud‐Based Framework for COVID‐19 Detection through Feature Fusion with Bootstrap Aggregated Extreme Learning Machine | |
Hussain et al. | Automated Deep Learning of COVID-19 and Pneumonia Detection Using Google AutoML. | |
Agnihotri et al. | A novel lightweight deep learning model based on SqueezeNet architecture for viral lung disease classification in X-ray and CT images | |
Arsenovic et al. | Detection of COVID-19 cases by utilizing deep learning algorithms on X-ray images | |
Al-Karawi et al. | Artificial intelligence-based chest X-Ray test of COVID-19 patients | |
Kumar et al. | Machine learning-based scheme to identify covid-19 in human bodies | |
Avianty et al. | COVID-19 Prediction Based on DWT and Moment Invariant Features of Radiography Image Using the Artificial Neural Network Classifier | |
Ogundokun et al. | Application of pre-trained CNN methods to identify COVID-19 pneumonia from Chest X-Ray | |
Sevli | A deep learning‐based approach for diagnosing COVID‐19 on chest x‐ray images, and a test study with clinical experts | |
Kudum et al. | Determining the best convolutional neural network for identifying tuberculosis and pneumonia in chest x-rays | |
Chandra et al. | A Deep InceptionV3 Model for Detecting Tuberculosis Disease Using CXR Images |
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 |