DEEP

LEARNING
Ria Kalra (Dept. Information Technology)
[Pranveer Singh Institute of Technology]
Kanpur , Uttar Pradesh
● Deep learning, a subfield of
artificial intelligence, has
revolutionized the field of image
classification. Deep learning
models use a hierarchy of layers
to automatically learn features
from the images without the need
for hand-crafted feature
extraction.
 A convolutional neural network (CNN or ConvNet) is a network architecture for deep
learning that learns directly from data. CNNs are particularly useful for finding patterns
in images to recognize objects, classes, and categories. They can also be quite effective for
classifying audio, time-series, and signal data.
TOOLS and PLATFORMS
● TensorFlow gives you the flexibility and control with
features like the Keras Functional API and Model
Subclassing API for creation of complex topologies.

● Colab allows anybody to write and execute arbitrary

python code through the browser, and is especially well
suited to machine learning, data analysis and education.

● Automate the training of your own custom machine

learning models. Simply upload images and train custom
image and video models with AutoML's easy-to-use
graphical interface; optimize your models for accuracy,
latency, and size; and export them to your application in
the cloud
HOW TO INITIATE A DEEP LEARNING
PROJECT ?
Define the task
Data Collection
Model training and exploration
Keep track of your model configuration and experiment
metadata
Model Refinement
Testing and evaluation
Model Deployment
PNEUMONI
A
DETECTIO
N SYSTEM
Using Chest X-ray
Images
 ABSTRACT
Developing an automatic system for detecting pneumonia would be beneficial for
treating the disease without any delay particularly in remote areas. Due to the success of
deep learning algorithms in analyzing medical images, Convolutional Neural Networks
(CNNs) have gained much attention for disease classification. In addition, features
learned by pre-trained CNN models on large-scale datasets are much useful in image
classification tasks. In this work, we appraise the functionality of pre-trained CNN
models utilized as feature-extractors followed by different classifiers for the classification
of abnormal and normal chest X-Rays. We analytically determine the optimal CNN
model for the purpose. Statistical results obtained demonstrates that pre-trained CNN
models employed along with supervised classifier algorithms can be very beneficial in
analyzing chest X-ray images, specifically to detect Pneumonia.
 INTRODUCTION
PNEUMONIA has been one of the most dangerous diseases across the world .
Researchers are trying to develop automated and feasible detection systems with
the help of deep neural networks, machine learning techniques, etc. In this paper, a
deep learning-based PNEUMONIA detection system is proposed that contributes
to detecting PNEUMONIA automatically using chest X-ray images. The system
introduces classifiers, namely, a multiclass classifier that automatically predicts
pneumonia, and a binary classifier that predicts pneumonia classes. Using transfer
learning, a deep CNN model is proposed to extract distinct and high-level features
from X-ray images in collaboration with the pertained model VGG-19.
 OBJECTIVE
By efficiently training through a
relatively image data set, our fine-tuned
models show high performance in the
classification of pneumonia. Our
conviction is that the proposed
computer-aided diagnosis mechanism
could outstandingly improve the
diagnosis of pneumonia cases. This is
very helpful, especially in the remote
areas.
METHODOLOGY
The proposed system extracts
distinct features from chest X-ray
images. After collecting the dataset
from different sources, the dataset is
preprocessed. Using transfer
learning, pertained model VGG-19 is
trained for the model evaluation.
Few new layers are added with the
base model to avoid the overfitting
problem of training the model.
 Data augmentation helped the proposed
model to perform better with exceptional
features of the training images.
Convolutional neural network (VGG-19) The
features of the image are extracted through
a series of convolutional layers. Using
transfer learning Transfer learning is a
method where a model trained for one
classification problem is used in training for
another classification problem.
DEMONSTARTION
MODEL EVALUATION
The metrics are evaluated separately
for each class label. To evaluate
these metrics, four basic terms are:
true positives (TP)
true negatives (TN)
false positives (FP)
false negatives (FN)
Accuracy=TP + TNTP + TN + FP +
FN 0.0<Accuracy
RESULT
On the basis of comparison with
other Models
Vgg19 found to be most efficient in
predicting the result with 90%
accuracy.
CONCLUSIO
N
The highest accuracy reported in the
above-mentioned proposed model
Vgg19 in classifying the normal vs.
pneumonia patients using X-ray images
using deep learning algorithms is
90.84%.

Successfully building Deep

Learning diagnosis systems
while maintaining people’s
privacy is a frontier that the
healthcare industry has
already established.
FUTURE SCOPE
Early detection of pneumonia increases
the survival rate of patients. It is one of
the medical systems that help to train
radiologists to improve their accuracy of
diagnosing pneumonia.
A approach for the automatic detection of
pneumonia in energy-efficient medical
systems is important :
--To improve the quality
--Reduced cost
--Time response
A number of experiments were
performed on a dataset of X ray images
and evaluated by a set of metrics. All
results confirmed the effectiveness and
efficiency of the proposed approach.
In future work, a large-sized dataset of
different lung diseases will be collected
to train the model to improve the
healthcare sectors. Moreover, an energy
evaluation measure will be used to
compute a real result of the energy
consumption of the proposed approach.
REFERENCES
DATASET
https://www.kaggle.com/datasets/paultimothymooney/chest-xray-pneumonia

MODEL Images and other

keras.io information
techvidvan.com
towardsdatascience.com
PLATFORM
www.researchgate.net
colab.research.google.com www.foreseemed.com
github.com
THANK YOU
Ria Kalra (IT) PSIT Kanpur, Uttar Pradesh (6387241340), [email protected]
ROC
CURVE
CONFUSION MATRIX
A confusion matrix, also
known as an error matrix, is
a summarized table used to
assess the performance of a
classification model. The
number of correct and
incorrect predictions are
summarized with count
values and broken down by
each class.
 OUR TEAM

RISHABH
RIA KALRA GULATI

2101640130061 (IT- 2101640130063

II) (IT-II)
TEAM LEADER
 THANK
youDO YOU HAVE ANY
QUESTIONS?
[email protected]
6387075810
PDS.com
CREDITS: This presentation template was created by Slidesgo, and includes
icons by Flaticon, and infographics & images by Freepik

Please keep this slide for attribution

 Model Deployment

CREDITS: This presentation template was created by Slidesgo, and includes

icons by Flaticon, and infographics & images by Freepik