SlideShare a Scribd company logo
Lung Conditions Prognosis Using CNN Model
BY HARSHIT JAIN
INDEX
INTRODUCTION
• Different sorts of disorders that impair the lungs' normal function are referred to
as Lung Diseases. They have an impact on pulmonary and respiratory processes,
including breathing and lung health.
• There are several lung conditions which are brought on by bacteria, viruses, or
other fungal infections, environmental changes. Other variables such as asthma,
carcinoma, and mesothelioma, also contribute to lung disorders .
• The subsequent disorders for which our study, Lung Condition Prognosis have
provided are Infiltration, Pneumonia, Hernia, Atelectasis, Cardiomegaly, Effusion,
Mass, Nodule, Consolidation and Pneumothorax.
PROBLEM AND THEORIES
PROBLEMS
• We are using CNN and not Artificial Intelligence because one problem with using
artificial intelligence in medicine is that there isn't enough data.
• The medical industry makes substantial use of machine learning techniques.
Finding hidden patterns in enormous amounts of data that is utilized for clinical
diagnostics has a lot of potential with data mining.
• Health businesses may use data mining to analyze data systematically, find
inefficiencies, and pinpoint best practices that enhance patient care while
reducing costs.
• Identification of lung disorders is one of the largest challenges, and several
researchers are working to assist physicians by creating sophisticated algorithms
for making medical judgments.
PROBLEM AND THEORIES
THEORY
• Deep learning strategies categories were learned gradually owing to their hidden layer
design by initially producing low-level categories like letters, then high-level categories
like words, and finally high-level categories like sentences. The network's neurons and
nodes generated a complete representation of the picture, with each representing a
different aspect of the whole.
• The main benefit of deep learning algorithms was that they tried to gradually learn high-
level qualities from data. As a result, hard-core feature extraction and domain expertise
were no longer necessary
• We have used CNN for the tissue pattern classification using mammographic images and
it shows the outstanding performance.
• The main objective was to create a prediction engine that will enable consumers to
determine if they have lung illness while sitting at home.
METHODS
• The first step was to create a custom X-Ray dataset for the 10 lung diseases which we collected
them from different labs and hospitals.
• After that, we used different object extraction models to extract the lungs from the X-Ray images
i.e., the required region and remove the unwanted regions in order to decrease the expense, and
time, and to improve the model's ability to forecast lung ailments.
• The task was to detect the diseases from abstracted lung images for which some deep learning
algorithms were used. We used CNN in which the feature extraction was done before flattening the
image. In this, the features were extracted, and the dimensions were reduced.
• Template matching was done which is a technique used in CNN to find a small part of the image. A
filter was generated here which was moving and generally of a 3x3 matrix, and that filter did match
with every pixel.
METHEDOLOGY FLOW CHART
• Firstly, the input data is taken and after that
preprocessing is performed on the data
inputted after preprocessing, which is
removal of missing values, noises in the data
etc.
• After that, the data bifurcation is performed
that is dividing the dataset into test and
train data and this has been done as we
cannot train the model on single dataset and
if we did so then it will not be able to assess
the performance of the model. Therefore,
there is a need to separate the data into
train test and validation datasets.
• After that the model is trained on the train
dataset and the resulting model M1 will be
used to validate the test dataset and the
final decision will be made i.e., finding or no
finding.
RESULT
DATASET
• The dataset used here for the implementation of this research work consists of data images of
lungs collected from different hospitals.
• In this research work, 10 different diseases have been tested. The data images have been divided
into two parts as one part is used for training and another part is used for testing process.
• The training data set consists of 90 images of every disease we are testing, and the testing dataset
consists of 10 images of each disease which in total makes 100 images of each disease utilized in
carrying out the research work.
RESULT
EXPERIMENTAL ANALYSIS
• The experiment carried out for this work was based on the self-developed CNN model.
• The developed CNN model consists of an input layer of size 600 × 600 × 3, dropout layer, average
pooling layer and dense layer.
• For every layer ‘relu’ activation function is used.
• After passing all images to the model, a total of 64,952,958 parameters are extracted.
• Among these parameters, 64,801,534 are used for training purposes and the remaining 151,424
are treated as non-trainable features.
RESULT
TRAINING AND VALIDATION LOSS CURVE OF USED MODEL.
RESULT
TRAINING AND VALIDATION ACCURACY CURVE
RESULT
• After the execution, the model is evaluated for 35 epochs with early-stopping
features. Therefore, the experiment stops at epoch number 12.
• The obtained results are measured in terms of Training loss , training accuracy,
validation loss and Validation accuracy.
• The obtained accuracy of the model after 13th epoch is like 90.6 % of training
accuracy, 0.1838 is training loss, 82.6 % of validation accuracy and 0.2396 is
validation error.
• The error and validation accuracy can be further enhanced by using various
parameters tunning. This can be performed in the next work.
CONCLUSION
• This work presents the working of different CNN for the automated detection of eleven different
lung diseases using chest X-Ray images.
• The self-designed CNN model has been used for the study and performance of the model is
computed in terms of the training accuracy, testing accuracy, training loss and validation loss.
• This study aimed to achieve accurate and error-free prediction of diseases while using minimal
manpower and small model architectures.
• To increase the accuracy of the work, the segmentation of the lung’s X-Ray images was carried
out which was a crucial step in order to reach precision using radiographs. It eliminated the noisy
data which was not required for the prediction of diseases.
• The study done has proved to be helpful in the fast and accurate detection of lung diseases and
has the potential to save many lives that are lost due to incorrect and delayed diagnoses.
REFERENCES
1. Frade, J., Pereira, T., Morgado, J. et al. Multiple instances learning for lung patho-physiological findings
detection using CT scans. Med Biol Eng Comput 60, 1569–1584 (2022). https://doi.org/10.1007/s11517-
022-02526-y
2. AliboniAlibi L, Pennati F, Gelmini A, Colombo A, Cioni A, Milanese G, Sverzellati N, Magnani S, Vespro V,
Blasi F, Aliverti A. Detection and classification of bronchiectasis through convolutional neural networks.
Journal of Thoracic Imaging. 2022 Mar 13; 37(2):100-8.
3. Chauhan, R., Ghanshala, K. K., & Joshi, R. C. (2018, December). Convolutional neural network (CNN) for
image detection and recognition. In 2018 First International Conference on Secure Cyber Computing and
Communication (ICSCCC) (pp. 278-282). IEEE.
4. Tiwari, P., Pant, B., Elarabawy, M. M., Abd-Elnaby, M., Mohd, N., Dhiman, G., & Shar-ma, S. (2022). Cnn
based multiclass brain tumor detection using medical imaging. Computa-tional Intelligence and
Neuroscience, 2022.
5. Bian Y, Zheng Z, Fang X, Jiang H, Zhu M, Yu J, Zhao H, Zhang L, Yao J, Lu L, Lu J. Ar-tificial Intelligence to
Predict Lymph Node Metastasis at CT in Pancreatic Ductal Adenocar-cinoma. Radiology. 2022 Sep 6:220329.
THANK YOU

More Related Content

Similar to Lung Conditions Prognosis Using CNN Model.pptx

research on journaling
research on journalingresearch on journaling
research on journaling
graphicdesigner79
 
AN AUTOMATED FRAMEWORK FOR DIAGNOSING LUNGS RELATED ISSUES USING ML AND DATA ...
AN AUTOMATED FRAMEWORK FOR DIAGNOSING LUNGS RELATED ISSUES USING ML AND DATA ...AN AUTOMATED FRAMEWORK FOR DIAGNOSING LUNGS RELATED ISSUES USING ML AND DATA ...
AN AUTOMATED FRAMEWORK FOR DIAGNOSING LUNGS RELATED ISSUES USING ML AND DATA ...
IRJET Journal
 
IRJET - Lung Disease Prediction using Image Processing and CNN Algorithm
IRJET -  	  Lung Disease Prediction using Image Processing and CNN AlgorithmIRJET -  	  Lung Disease Prediction using Image Processing and CNN Algorithm
IRJET - Lung Disease Prediction using Image Processing and CNN Algorithm
IRJET Journal
 
Prediction of Lung Cancer Using Image Processing Techniques: A Review
Prediction of Lung Cancer Using Image Processing Techniques: A ReviewPrediction of Lung Cancer Using Image Processing Techniques: A Review
Prediction of Lung Cancer Using Image Processing Techniques: A Review
aciijournal
 
Prediction of Lung Cancer Using Image Processing Techniques: A Review
Prediction of Lung Cancer Using Image Processing Techniques: A ReviewPrediction of Lung Cancer Using Image Processing Techniques: A Review
Prediction of Lung Cancer Using Image Processing Techniques: A Review
aciijournal
 
Prediction of lung cancer using image
Prediction of lung cancer using imagePrediction of lung cancer using image
Prediction of lung cancer using image
aciijournal
 
AN EFFECTIVE AND EFFICIENT FEATURE SELECTION METHOD FOR LUNG CANCER DETECTION
AN EFFECTIVE AND EFFICIENT FEATURE SELECTION METHOD FOR LUNG CANCER DETECTIONAN EFFECTIVE AND EFFICIENT FEATURE SELECTION METHOD FOR LUNG CANCER DETECTION
AN EFFECTIVE AND EFFICIENT FEATURE SELECTION METHOD FOR LUNG CANCER DETECTION
ijcsit
 
X-Ray Disease Identifier
X-Ray Disease IdentifierX-Ray Disease Identifier
X-Ray Disease Identifier
IRJET Journal
 
Digital flouroscopy
Digital flouroscopyDigital flouroscopy
Digital flouroscopy
HassanNaseer10
 
YOLOv8-Based Lung Nodule Detection: A Novel Hybrid Deep Learning Model Proposal
YOLOv8-Based Lung Nodule Detection: A Novel Hybrid Deep Learning Model ProposalYOLOv8-Based Lung Nodule Detection: A Novel Hybrid Deep Learning Model Proposal
YOLOv8-Based Lung Nodule Detection: A Novel Hybrid Deep Learning Model Proposal
IRJET Journal
 
Madhavi tippani
Madhavi tippaniMadhavi tippani
Madhavi tippani
Madhavi Tippani
 
ICU Patient Deterioration Prediction : A Data-Mining Approach
ICU Patient Deterioration Prediction : A Data-Mining ApproachICU Patient Deterioration Prediction : A Data-Mining Approach
ICU Patient Deterioration Prediction : A Data-Mining Approach
csandit
 
ICU PATIENT DETERIORATION PREDICTION: A DATA-MINING APPROACH
ICU PATIENT DETERIORATION PREDICTION: A DATA-MINING APPROACHICU PATIENT DETERIORATION PREDICTION: A DATA-MINING APPROACH
ICU PATIENT DETERIORATION PREDICTION: A DATA-MINING APPROACH
cscpconf
 
A Review On Lung Cancer Detection From CT Scan Images Using CNN
A Review On Lung Cancer Detection From CT Scan Images Using CNNA Review On Lung Cancer Detection From CT Scan Images Using CNN
A Review On Lung Cancer Detection From CT Scan Images Using CNN
Don Dooley
 
covid 19 detection using lung x-rays.pptx.pptx
covid 19 detection using lung x-rays.pptx.pptxcovid 19 detection using lung x-rays.pptx.pptx
covid 19 detection using lung x-rays.pptx.pptx
Diya940551
 
Lung Nodule detection System
Lung Nodule detection SystemLung Nodule detection System
Lung Nodule detection System
Editor IJMTER
 
Multiple Analysis of Brain Tumor Detection Based on FCM
Multiple Analysis of Brain Tumor Detection Based on FCMMultiple Analysis of Brain Tumor Detection Based on FCM
Multiple Analysis of Brain Tumor Detection Based on FCM
IRJET Journal
 
Artificial Intelligence in pathology
Artificial Intelligence in pathologyArtificial Intelligence in pathology
Artificial Intelligence in pathology
nehaSingh1543
 
Lung Cancer Detection with Flask Integration
Lung Cancer Detection with Flask IntegrationLung Cancer Detection with Flask Integration
Lung Cancer Detection with Flask Integration
IRJET Journal
 
A Progressive Review on Early Stage Breast Cancer Detection
A Progressive Review on Early Stage Breast Cancer DetectionA Progressive Review on Early Stage Breast Cancer Detection
A Progressive Review on Early Stage Breast Cancer Detection
IRJET Journal
 

Similar to Lung Conditions Prognosis Using CNN Model.pptx (20)

research on journaling
research on journalingresearch on journaling
research on journaling
 
AN AUTOMATED FRAMEWORK FOR DIAGNOSING LUNGS RELATED ISSUES USING ML AND DATA ...
AN AUTOMATED FRAMEWORK FOR DIAGNOSING LUNGS RELATED ISSUES USING ML AND DATA ...AN AUTOMATED FRAMEWORK FOR DIAGNOSING LUNGS RELATED ISSUES USING ML AND DATA ...
AN AUTOMATED FRAMEWORK FOR DIAGNOSING LUNGS RELATED ISSUES USING ML AND DATA ...
 
IRJET - Lung Disease Prediction using Image Processing and CNN Algorithm
IRJET -  	  Lung Disease Prediction using Image Processing and CNN AlgorithmIRJET -  	  Lung Disease Prediction using Image Processing and CNN Algorithm
IRJET - Lung Disease Prediction using Image Processing and CNN Algorithm
 
Prediction of Lung Cancer Using Image Processing Techniques: A Review
Prediction of Lung Cancer Using Image Processing Techniques: A ReviewPrediction of Lung Cancer Using Image Processing Techniques: A Review
Prediction of Lung Cancer Using Image Processing Techniques: A Review
 
Prediction of Lung Cancer Using Image Processing Techniques: A Review
Prediction of Lung Cancer Using Image Processing Techniques: A ReviewPrediction of Lung Cancer Using Image Processing Techniques: A Review
Prediction of Lung Cancer Using Image Processing Techniques: A Review
 
Prediction of lung cancer using image
Prediction of lung cancer using imagePrediction of lung cancer using image
Prediction of lung cancer using image
 
AN EFFECTIVE AND EFFICIENT FEATURE SELECTION METHOD FOR LUNG CANCER DETECTION
AN EFFECTIVE AND EFFICIENT FEATURE SELECTION METHOD FOR LUNG CANCER DETECTIONAN EFFECTIVE AND EFFICIENT FEATURE SELECTION METHOD FOR LUNG CANCER DETECTION
AN EFFECTIVE AND EFFICIENT FEATURE SELECTION METHOD FOR LUNG CANCER DETECTION
 
X-Ray Disease Identifier
X-Ray Disease IdentifierX-Ray Disease Identifier
X-Ray Disease Identifier
 
Digital flouroscopy
Digital flouroscopyDigital flouroscopy
Digital flouroscopy
 
YOLOv8-Based Lung Nodule Detection: A Novel Hybrid Deep Learning Model Proposal
YOLOv8-Based Lung Nodule Detection: A Novel Hybrid Deep Learning Model ProposalYOLOv8-Based Lung Nodule Detection: A Novel Hybrid Deep Learning Model Proposal
YOLOv8-Based Lung Nodule Detection: A Novel Hybrid Deep Learning Model Proposal
 
Madhavi tippani
Madhavi tippaniMadhavi tippani
Madhavi tippani
 
ICU Patient Deterioration Prediction : A Data-Mining Approach
ICU Patient Deterioration Prediction : A Data-Mining ApproachICU Patient Deterioration Prediction : A Data-Mining Approach
ICU Patient Deterioration Prediction : A Data-Mining Approach
 
ICU PATIENT DETERIORATION PREDICTION: A DATA-MINING APPROACH
ICU PATIENT DETERIORATION PREDICTION: A DATA-MINING APPROACHICU PATIENT DETERIORATION PREDICTION: A DATA-MINING APPROACH
ICU PATIENT DETERIORATION PREDICTION: A DATA-MINING APPROACH
 
A Review On Lung Cancer Detection From CT Scan Images Using CNN
A Review On Lung Cancer Detection From CT Scan Images Using CNNA Review On Lung Cancer Detection From CT Scan Images Using CNN
A Review On Lung Cancer Detection From CT Scan Images Using CNN
 
covid 19 detection using lung x-rays.pptx.pptx
covid 19 detection using lung x-rays.pptx.pptxcovid 19 detection using lung x-rays.pptx.pptx
covid 19 detection using lung x-rays.pptx.pptx
 
Lung Nodule detection System
Lung Nodule detection SystemLung Nodule detection System
Lung Nodule detection System
 
Multiple Analysis of Brain Tumor Detection Based on FCM
Multiple Analysis of Brain Tumor Detection Based on FCMMultiple Analysis of Brain Tumor Detection Based on FCM
Multiple Analysis of Brain Tumor Detection Based on FCM
 
Artificial Intelligence in pathology
Artificial Intelligence in pathologyArtificial Intelligence in pathology
Artificial Intelligence in pathology
 
Lung Cancer Detection with Flask Integration
Lung Cancer Detection with Flask IntegrationLung Cancer Detection with Flask Integration
Lung Cancer Detection with Flask Integration
 
A Progressive Review on Early Stage Breast Cancer Detection
A Progressive Review on Early Stage Breast Cancer DetectionA Progressive Review on Early Stage Breast Cancer Detection
A Progressive Review on Early Stage Breast Cancer Detection
 

Recently uploaded

POST OPERATIVE OLIGURIA and its management
POST OPERATIVE OLIGURIA and its managementPOST OPERATIVE OLIGURIA and its management
POST OPERATIVE OLIGURIA and its management
touseefaziz1
 
New Drug Discovery and Development .....
New Drug Discovery and Development .....New Drug Discovery and Development .....
New Drug Discovery and Development .....
NEHA GUPTA
 
The Normal Electrocardiogram - Part I of II
The Normal Electrocardiogram - Part I of IIThe Normal Electrocardiogram - Part I of II
The Normal Electrocardiogram - Part I of II
MedicoseAcademics
 
NVBDCP.pptx Nation vector borne disease control program
NVBDCP.pptx Nation vector borne disease control programNVBDCP.pptx Nation vector borne disease control program
NVBDCP.pptx Nation vector borne disease control program
Sapna Thakur
 
Ophthalmology Clinical Tests for OSCE exam
Ophthalmology Clinical Tests for OSCE examOphthalmology Clinical Tests for OSCE exam
Ophthalmology Clinical Tests for OSCE exam
KafrELShiekh University
 
Charaka Samhita Sutra sthana Chapter 15 Upakalpaniyaadhyaya
Charaka Samhita Sutra sthana Chapter 15 UpakalpaniyaadhyayaCharaka Samhita Sutra sthana Chapter 15 Upakalpaniyaadhyaya
Charaka Samhita Sutra sthana Chapter 15 Upakalpaniyaadhyaya
Dr KHALID B.M
 
New Directions in Targeted Therapeutic Approaches for Older Adults With Mantl...
New Directions in Targeted Therapeutic Approaches for Older Adults With Mantl...New Directions in Targeted Therapeutic Approaches for Older Adults With Mantl...
New Directions in Targeted Therapeutic Approaches for Older Adults With Mantl...
i3 Health
 
Alcohol_Dr. Jeenal Mistry MD Pharmacology.pdf
Alcohol_Dr. Jeenal Mistry MD Pharmacology.pdfAlcohol_Dr. Jeenal Mistry MD Pharmacology.pdf
Alcohol_Dr. Jeenal Mistry MD Pharmacology.pdf
Dr Jeenal Mistry
 
How to Give Better Lectures: Some Tips for Doctors
How to Give Better Lectures: Some Tips for DoctorsHow to Give Better Lectures: Some Tips for Doctors
How to Give Better Lectures: Some Tips for Doctors
LanceCatedral
 
24 Upakrama.pptx class ppt useful in all
24 Upakrama.pptx class ppt useful in all24 Upakrama.pptx class ppt useful in all
24 Upakrama.pptx class ppt useful in all
DrSathishMS1
 
Ocular injury ppt Upendra pal optometrist upums saifai etawah
Ocular injury  ppt  Upendra pal  optometrist upums saifai etawahOcular injury  ppt  Upendra pal  optometrist upums saifai etawah
Ocular injury ppt Upendra pal optometrist upums saifai etawah
pal078100
 
Lung Cancer: Artificial Intelligence, Synergetics, Complex System Analysis, S...
Lung Cancer: Artificial Intelligence, Synergetics, Complex System Analysis, S...Lung Cancer: Artificial Intelligence, Synergetics, Complex System Analysis, S...
Lung Cancer: Artificial Intelligence, Synergetics, Complex System Analysis, S...
Oleg Kshivets
 
TEST BANK for Operations Management, 14th Edition by William J. Stevenson, Ve...
TEST BANK for Operations Management, 14th Edition by William J. Stevenson, Ve...TEST BANK for Operations Management, 14th Edition by William J. Stevenson, Ve...
TEST BANK for Operations Management, 14th Edition by William J. Stevenson, Ve...
kevinkariuki227
 
Physiology of Special Chemical Sensation of Taste
Physiology of Special Chemical Sensation of TastePhysiology of Special Chemical Sensation of Taste
Physiology of Special Chemical Sensation of Taste
MedicoseAcademics
 
How STIs Influence the Development of Pelvic Inflammatory Disease.pptx
How STIs Influence the Development of Pelvic Inflammatory Disease.pptxHow STIs Influence the Development of Pelvic Inflammatory Disease.pptx
How STIs Influence the Development of Pelvic Inflammatory Disease.pptx
FFragrant
 
Superficial & Deep Fascia of the NECK.pptx
Superficial & Deep Fascia of the NECK.pptxSuperficial & Deep Fascia of the NECK.pptx
Superficial & Deep Fascia of the NECK.pptx
Dr. Rabia Inam Gandapore
 
Couples presenting to the infertility clinic- Do they really have infertility...
Couples presenting to the infertility clinic- Do they really have infertility...Couples presenting to the infertility clinic- Do they really have infertility...
Couples presenting to the infertility clinic- Do they really have infertility...
Sujoy Dasgupta
 
HOT NEW PRODUCT! BIG SALES FAST SHIPPING NOW FROM CHINA!! EU KU DB BK substit...
HOT NEW PRODUCT! BIG SALES FAST SHIPPING NOW FROM CHINA!! EU KU DB BK substit...HOT NEW PRODUCT! BIG SALES FAST SHIPPING NOW FROM CHINA!! EU KU DB BK substit...
HOT NEW PRODUCT! BIG SALES FAST SHIPPING NOW FROM CHINA!! EU KU DB BK substit...
GL Anaacs
 
Cervical & Brachial Plexus By Dr. RIG.pptx
Cervical & Brachial Plexus By Dr. RIG.pptxCervical & Brachial Plexus By Dr. RIG.pptx
Cervical & Brachial Plexus By Dr. RIG.pptx
Dr. Rabia Inam Gandapore
 
Maxilla, Mandible & Hyoid Bone & Clinical Correlations by Dr. RIG.pptx
Maxilla, Mandible & Hyoid Bone & Clinical Correlations by Dr. RIG.pptxMaxilla, Mandible & Hyoid Bone & Clinical Correlations by Dr. RIG.pptx
Maxilla, Mandible & Hyoid Bone & Clinical Correlations by Dr. RIG.pptx
Dr. Rabia Inam Gandapore
 

Recently uploaded (20)

POST OPERATIVE OLIGURIA and its management
POST OPERATIVE OLIGURIA and its managementPOST OPERATIVE OLIGURIA and its management
POST OPERATIVE OLIGURIA and its management
 
New Drug Discovery and Development .....
New Drug Discovery and Development .....New Drug Discovery and Development .....
New Drug Discovery and Development .....
 
The Normal Electrocardiogram - Part I of II
The Normal Electrocardiogram - Part I of IIThe Normal Electrocardiogram - Part I of II
The Normal Electrocardiogram - Part I of II
 
NVBDCP.pptx Nation vector borne disease control program
NVBDCP.pptx Nation vector borne disease control programNVBDCP.pptx Nation vector borne disease control program
NVBDCP.pptx Nation vector borne disease control program
 
Ophthalmology Clinical Tests for OSCE exam
Ophthalmology Clinical Tests for OSCE examOphthalmology Clinical Tests for OSCE exam
Ophthalmology Clinical Tests for OSCE exam
 
Charaka Samhita Sutra sthana Chapter 15 Upakalpaniyaadhyaya
Charaka Samhita Sutra sthana Chapter 15 UpakalpaniyaadhyayaCharaka Samhita Sutra sthana Chapter 15 Upakalpaniyaadhyaya
Charaka Samhita Sutra sthana Chapter 15 Upakalpaniyaadhyaya
 
New Directions in Targeted Therapeutic Approaches for Older Adults With Mantl...
New Directions in Targeted Therapeutic Approaches for Older Adults With Mantl...New Directions in Targeted Therapeutic Approaches for Older Adults With Mantl...
New Directions in Targeted Therapeutic Approaches for Older Adults With Mantl...
 
Alcohol_Dr. Jeenal Mistry MD Pharmacology.pdf
Alcohol_Dr. Jeenal Mistry MD Pharmacology.pdfAlcohol_Dr. Jeenal Mistry MD Pharmacology.pdf
Alcohol_Dr. Jeenal Mistry MD Pharmacology.pdf
 
How to Give Better Lectures: Some Tips for Doctors
How to Give Better Lectures: Some Tips for DoctorsHow to Give Better Lectures: Some Tips for Doctors
How to Give Better Lectures: Some Tips for Doctors
 
24 Upakrama.pptx class ppt useful in all
24 Upakrama.pptx class ppt useful in all24 Upakrama.pptx class ppt useful in all
24 Upakrama.pptx class ppt useful in all
 
Ocular injury ppt Upendra pal optometrist upums saifai etawah
Ocular injury  ppt  Upendra pal  optometrist upums saifai etawahOcular injury  ppt  Upendra pal  optometrist upums saifai etawah
Ocular injury ppt Upendra pal optometrist upums saifai etawah
 
Lung Cancer: Artificial Intelligence, Synergetics, Complex System Analysis, S...
Lung Cancer: Artificial Intelligence, Synergetics, Complex System Analysis, S...Lung Cancer: Artificial Intelligence, Synergetics, Complex System Analysis, S...
Lung Cancer: Artificial Intelligence, Synergetics, Complex System Analysis, S...
 
TEST BANK for Operations Management, 14th Edition by William J. Stevenson, Ve...
TEST BANK for Operations Management, 14th Edition by William J. Stevenson, Ve...TEST BANK for Operations Management, 14th Edition by William J. Stevenson, Ve...
TEST BANK for Operations Management, 14th Edition by William J. Stevenson, Ve...
 
Physiology of Special Chemical Sensation of Taste
Physiology of Special Chemical Sensation of TastePhysiology of Special Chemical Sensation of Taste
Physiology of Special Chemical Sensation of Taste
 
How STIs Influence the Development of Pelvic Inflammatory Disease.pptx
How STIs Influence the Development of Pelvic Inflammatory Disease.pptxHow STIs Influence the Development of Pelvic Inflammatory Disease.pptx
How STIs Influence the Development of Pelvic Inflammatory Disease.pptx
 
Superficial & Deep Fascia of the NECK.pptx
Superficial & Deep Fascia of the NECK.pptxSuperficial & Deep Fascia of the NECK.pptx
Superficial & Deep Fascia of the NECK.pptx
 
Couples presenting to the infertility clinic- Do they really have infertility...
Couples presenting to the infertility clinic- Do they really have infertility...Couples presenting to the infertility clinic- Do they really have infertility...
Couples presenting to the infertility clinic- Do they really have infertility...
 
HOT NEW PRODUCT! BIG SALES FAST SHIPPING NOW FROM CHINA!! EU KU DB BK substit...
HOT NEW PRODUCT! BIG SALES FAST SHIPPING NOW FROM CHINA!! EU KU DB BK substit...HOT NEW PRODUCT! BIG SALES FAST SHIPPING NOW FROM CHINA!! EU KU DB BK substit...
HOT NEW PRODUCT! BIG SALES FAST SHIPPING NOW FROM CHINA!! EU KU DB BK substit...
 
Cervical & Brachial Plexus By Dr. RIG.pptx
Cervical & Brachial Plexus By Dr. RIG.pptxCervical & Brachial Plexus By Dr. RIG.pptx
Cervical & Brachial Plexus By Dr. RIG.pptx
 
Maxilla, Mandible & Hyoid Bone & Clinical Correlations by Dr. RIG.pptx
Maxilla, Mandible & Hyoid Bone & Clinical Correlations by Dr. RIG.pptxMaxilla, Mandible & Hyoid Bone & Clinical Correlations by Dr. RIG.pptx
Maxilla, Mandible & Hyoid Bone & Clinical Correlations by Dr. RIG.pptx
 

Lung Conditions Prognosis Using CNN Model.pptx

  • 1. Lung Conditions Prognosis Using CNN Model BY HARSHIT JAIN
  • 3. INTRODUCTION • Different sorts of disorders that impair the lungs' normal function are referred to as Lung Diseases. They have an impact on pulmonary and respiratory processes, including breathing and lung health. • There are several lung conditions which are brought on by bacteria, viruses, or other fungal infections, environmental changes. Other variables such as asthma, carcinoma, and mesothelioma, also contribute to lung disorders . • The subsequent disorders for which our study, Lung Condition Prognosis have provided are Infiltration, Pneumonia, Hernia, Atelectasis, Cardiomegaly, Effusion, Mass, Nodule, Consolidation and Pneumothorax.
  • 4. PROBLEM AND THEORIES PROBLEMS • We are using CNN and not Artificial Intelligence because one problem with using artificial intelligence in medicine is that there isn't enough data. • The medical industry makes substantial use of machine learning techniques. Finding hidden patterns in enormous amounts of data that is utilized for clinical diagnostics has a lot of potential with data mining. • Health businesses may use data mining to analyze data systematically, find inefficiencies, and pinpoint best practices that enhance patient care while reducing costs. • Identification of lung disorders is one of the largest challenges, and several researchers are working to assist physicians by creating sophisticated algorithms for making medical judgments.
  • 5. PROBLEM AND THEORIES THEORY • Deep learning strategies categories were learned gradually owing to their hidden layer design by initially producing low-level categories like letters, then high-level categories like words, and finally high-level categories like sentences. The network's neurons and nodes generated a complete representation of the picture, with each representing a different aspect of the whole. • The main benefit of deep learning algorithms was that they tried to gradually learn high- level qualities from data. As a result, hard-core feature extraction and domain expertise were no longer necessary • We have used CNN for the tissue pattern classification using mammographic images and it shows the outstanding performance. • The main objective was to create a prediction engine that will enable consumers to determine if they have lung illness while sitting at home.
  • 6. METHODS • The first step was to create a custom X-Ray dataset for the 10 lung diseases which we collected them from different labs and hospitals. • After that, we used different object extraction models to extract the lungs from the X-Ray images i.e., the required region and remove the unwanted regions in order to decrease the expense, and time, and to improve the model's ability to forecast lung ailments. • The task was to detect the diseases from abstracted lung images for which some deep learning algorithms were used. We used CNN in which the feature extraction was done before flattening the image. In this, the features were extracted, and the dimensions were reduced. • Template matching was done which is a technique used in CNN to find a small part of the image. A filter was generated here which was moving and generally of a 3x3 matrix, and that filter did match with every pixel.
  • 7. METHEDOLOGY FLOW CHART • Firstly, the input data is taken and after that preprocessing is performed on the data inputted after preprocessing, which is removal of missing values, noises in the data etc. • After that, the data bifurcation is performed that is dividing the dataset into test and train data and this has been done as we cannot train the model on single dataset and if we did so then it will not be able to assess the performance of the model. Therefore, there is a need to separate the data into train test and validation datasets. • After that the model is trained on the train dataset and the resulting model M1 will be used to validate the test dataset and the final decision will be made i.e., finding or no finding.
  • 8. RESULT DATASET • The dataset used here for the implementation of this research work consists of data images of lungs collected from different hospitals. • In this research work, 10 different diseases have been tested. The data images have been divided into two parts as one part is used for training and another part is used for testing process. • The training data set consists of 90 images of every disease we are testing, and the testing dataset consists of 10 images of each disease which in total makes 100 images of each disease utilized in carrying out the research work.
  • 9. RESULT EXPERIMENTAL ANALYSIS • The experiment carried out for this work was based on the self-developed CNN model. • The developed CNN model consists of an input layer of size 600 × 600 × 3, dropout layer, average pooling layer and dense layer. • For every layer ‘relu’ activation function is used. • After passing all images to the model, a total of 64,952,958 parameters are extracted. • Among these parameters, 64,801,534 are used for training purposes and the remaining 151,424 are treated as non-trainable features.
  • 10. RESULT TRAINING AND VALIDATION LOSS CURVE OF USED MODEL.
  • 12. RESULT • After the execution, the model is evaluated for 35 epochs with early-stopping features. Therefore, the experiment stops at epoch number 12. • The obtained results are measured in terms of Training loss , training accuracy, validation loss and Validation accuracy. • The obtained accuracy of the model after 13th epoch is like 90.6 % of training accuracy, 0.1838 is training loss, 82.6 % of validation accuracy and 0.2396 is validation error. • The error and validation accuracy can be further enhanced by using various parameters tunning. This can be performed in the next work.
  • 13. CONCLUSION • This work presents the working of different CNN for the automated detection of eleven different lung diseases using chest X-Ray images. • The self-designed CNN model has been used for the study and performance of the model is computed in terms of the training accuracy, testing accuracy, training loss and validation loss. • This study aimed to achieve accurate and error-free prediction of diseases while using minimal manpower and small model architectures. • To increase the accuracy of the work, the segmentation of the lung’s X-Ray images was carried out which was a crucial step in order to reach precision using radiographs. It eliminated the noisy data which was not required for the prediction of diseases. • The study done has proved to be helpful in the fast and accurate detection of lung diseases and has the potential to save many lives that are lost due to incorrect and delayed diagnoses.
  • 14. REFERENCES 1. Frade, J., Pereira, T., Morgado, J. et al. Multiple instances learning for lung patho-physiological findings detection using CT scans. Med Biol Eng Comput 60, 1569–1584 (2022). https://doi.org/10.1007/s11517- 022-02526-y 2. AliboniAlibi L, Pennati F, Gelmini A, Colombo A, Cioni A, Milanese G, Sverzellati N, Magnani S, Vespro V, Blasi F, Aliverti A. Detection and classification of bronchiectasis through convolutional neural networks. Journal of Thoracic Imaging. 2022 Mar 13; 37(2):100-8. 3. Chauhan, R., Ghanshala, K. K., & Joshi, R. C. (2018, December). Convolutional neural network (CNN) for image detection and recognition. In 2018 First International Conference on Secure Cyber Computing and Communication (ICSCCC) (pp. 278-282). IEEE. 4. Tiwari, P., Pant, B., Elarabawy, M. M., Abd-Elnaby, M., Mohd, N., Dhiman, G., & Shar-ma, S. (2022). Cnn based multiclass brain tumor detection using medical imaging. Computa-tional Intelligence and Neuroscience, 2022. 5. Bian Y, Zheng Z, Fang X, Jiang H, Zhu M, Yu J, Zhao H, Zhang L, Yao J, Lu L, Lu J. Ar-tificial Intelligence to Predict Lymph Node Metastasis at CT in Pancreatic Ductal Adenocar-cinoma. Radiology. 2022 Sep 6:220329.