Prediction of Diseases using Deep Learning: A Review

Raman Kumar

, Arvind Kumar Tiwari

Department of CSE, Kamla Nehru Institute of Technology, Sultanpur(India)

Keywords: Medical Image Analysis, Deep Learning, Long Short Term Memory.

Abstract: Deep learning is a prime focus area for medical image analysis in the recent time. With the technical

advancements in the last few decades there is a good amount of images in various databases and a number of

researchers have focused their researches in medical image analysis. This paper presents the deep learning

based approaches for the prediction of Gastrointestinal Diseases, Lung Disease, Breast Cancer Diagnosis and

Brain Diseases available in literature. This paper also presents the summary for the prediction of

Gastrointestinal Diseases, Lung Disease, Breast Cancer Diagnosis and Brain Diseases available in literature.

1 INTRODUCTION

Medical imaging is a term generally used to represent

a set of technologies that produce images of internal

body parts. The purpose of medical imaging is to

monitor the health and injuries of organs. Magnetic

Resonance Imaging (MRI), Ultrasound, Biopsy,

Endoscopy, X-Ray, Mammography are among most

commonly used medical imaging techniques. With

the technical advancements in the last few decades

the amount of images generated in the field of

medical has increased tremendously. These medical

images are stored in various databases along with the

images of healthy organs or tissues to compare. The

generated medical images differ in imaging technique,

resolution, dimensionality and quality. Generally

these medical images are analyzed manually by the

medical practitioner. Medical image analysis is a

complex and time consuming process with which

even professional struggle .One of the major problem

associated with manual image analysis is the image

can be analyzed differently by different experts based

on their knowledgeE.Sudheer Kumar and C.Shoba

Bindu,2019.This may result in different diagnosis

and treatment .Thus there is a need for automated

analysis of medical data . Bruijne has discussed major

challenges in medical image data as lack of labeled

data, variable imaging techniques (Marleen de

Bruijne,2016).

In early days of automated classification medical

images the images were classified based on the

predefined characteristics assigned previously by

experts to machine learning models. The drawback

of this system was that the machine learning

algorithms were not able ho classify the images when

they contain structural similarity. Then the focus of

image classification shifted from machine learning to

deep learning techniques. The automated analysis of

medical data can be done by using deep learning

techniques. Deep learning is a sub domain of machine

learning which uses techniques inspired from the

learning ability of the human brain. The architecture

of these deep learning techniques is little complex but

computationally stronger when compared with other

machine learning methods. Along with other

applications, deep learning is deployed at the front

lines of healthcare and has produced the influential

results by analyzing huge electronic data for

treatment of various diseases. Deep Learning

techniques have shown capacity to analyze the

medical data at much faster rate and with more

accuracy in comparison to manual methods. Deep

learning algorithms attempt to learn high level and

complex abstraction as representations of data by

utilizing the hierarchical learning process (Ricardo

Buettner,2020) .

2 DEEP LEARNING

TECHNIQUES

Deep learning is a branch of the machine learning

which primarily uses different kind of neural

networks for prediction .Some of the deep learning

methods used in this paper are as follows:

Kumar, R. and Tiwari, A.

Prediction of Diseases using Deep Learning: A Review.

DOI: 10.5220/0010564000003161

In Proceedings of the 3rd International Conference on Advanced Computing and Software Engineering (ICACSE 2021), pages 129-134

ISBN: 978-989-758-544-9

129

2.1 CNN

Convolutional Neural Networks (CNN) is neural

networks which contain convolution layers(filter

layers) which are used to extract useful information

from the input image.

2.2 RNN

Recurrent Neural Network is neural networks

characterized by presence of self-loops in the hidden

layers of the neural network. Recurrent Neural

Networks have the capability to use sequential data

and have predicted outputs based on input data along

with the information is previous layers.

2.3 LSTM

LSTM is special kind of recurrent neural networks

which are created to solve the problem of long term

dependencies. LSTM have chain like structure, along

with the repeating units. LSTM have three gates

called forget gate, input gate and output gate which

decide how much information should be erased,

updated and provided as output.

3 RELATED WORK

This paper presents the deep learning based

approaches for the prediction of Gastrointestinal

Diseases, Lung Disease, Breast Cancer Diagnosis and

Brain Diseases.

3.1 Gastrointestinal Diseases

Gastrointestinal (GI) diseases are the diseases related

with the digestive system. The organs studied within

the gastrointestinal domain are liver, pancreas, small

intestine large intestine, rectum and anus. Aman

Srivastava et al. (2019) proposed a CNN model to

predict celiac and environmental enteropathy using

biopsy images and obtained accuracy of 97.6%.

Samira Lafraxo et al. (2020) used CNN model to

abnormalities recognition on endoscopic images

obtained from KVASIR dataset and obtained 96.89%

accuracy.

Table 1: Summary of deep learning methods in GI disease

Prediction

Author/Year

Techniq

ues

Modality Source Accuracy

Chen-Ying

Hung

,2019

DNN

Elelectroni

cHealth

Record

(multimod

ality)

EHR of

Taichung

Veterans

General

Hospital

87.6%

Kuntesh

Jani,2019

CNN Endoscopy

Images

from CE

videos

95.11%

Pradipta

Sasmal,2018

CNN Endoscopy

CVC

clinic

database&

Hamlyn

Centre

Laproscop

ic/Endosc

opic

dataset

99.85%

Aman

Srivastava,2

019

CNN Biopsy

WSI

images

93%

Alexy A

Shvets,

2018

CNN Endoscopy

WCE

images

75.35%

Franklin

Sierra,

2020

CNN

Colonosco

Dataset

consists of

76 NBI

video

images

90.79%

Yaxing Cao,

2018

DCNN Endoscopy

WCE

images

98.37%

Qiang Wang

,2019

CNN Endoscopy

Chinese

PLA

General

Hospital

datase

96.1%

Chathurika

Gamage,201

CNN Endoscopy

KVASIR

dataset

97.38%

Spiros V.

Georgakopo

ulos,2018

CNN Endoscopy

KID

database

90.2%

Tonmoy

Ghosh,

2016

CNN Endoscopy

KID

database

94.42%

3.2 Lung Disease

Lung diseases are related with the respiratory

function of the lungs. Lungs contract and expand with

the help of diaphragm. This contraction and

expansion helps the lungs to inhale fresh air

ICACSE 2021 - International Conference on Advanced Computing and Software Engineering

130

containing oxygen and exhale air containing carbon

dioxide. Amrit Sreekumar et al.(2020) used a 3DCNN

model to detect presence of lung cancer using CT

scan images obtained from LIDC-IDRI dataset.They

obtained the accuracy of 86% . Rahul Hooda et al

(2017) proposed a Tuberculosis prediction method

using CXR images with CNN.Used images are from

two datasets Montgomery and Shenzhen.

Table 2. Summary of deep learning methods in lung disease

Prediction

Author/Year Techniques Modality Source Accuracy

Sheikh

Rafiul

Islam, 2019

CNN CXR Kaggle 97.34%

Karan

Jakhar, 2018

DCNN CXR

Chest

XRay

data

84%

Diksha

Mhaske,201

CNN-

LSTM

CT scan

LIDC-

IDRI

97%

S. Rajarama

2019

Ensemble CXR

Kaggle

pneumo

nia

detectio

challeng

e datase

98.7%

Ahmad

P.Tafti,

2018

3DCNN CT scan

DSB

2017,

MLCIA

datasets

83.75%

Ruchika

Tekade,201

3DCNN CT scan

LIDC-

IDRI,

LUNA

2016,

Kaggle

data

science

Bowl

2017

datasets

95.66%

Matko Saric,

2019

CNN

Histopath

ology

WSI 75.41%

3.3 Breast Cancer Diagnosis

Breast cancer is one of the most frequent occurring

cancer in females. Breast cancer is second largest

cause of deaths in females after skin cancer. Ankit

Titoriya et al.(2019) used CNN to predict the breast

cancer using the histopathology images. The data set

used is BreakHis dataset and they obtained accuracy

of 93.8% in classifying.

Table 3. Summary of deep learning methods in Breast

Cancer Diagnosis

Author/Year Techniques Modality Source Accuracy

Hari Krishna

Tiammana,

2020

CNN

Mammogr

aphy

WBCD 97.94%

Pritam

Sarkar,

2019

DNN

Mammogr

aphy,

scans,MRI

WBC

DIAGN

OSTIC,

WBC

Original

datasets

99.52%

Naresh

Khuriwal,20

CNN

Histopath

ology

MIAS

dataset

98%

Jasmir,

2018

MLP Oncology

Medical

Center

universit

institute

Oncolog

y dataset

96.5%

Ahmed

Hijab, 2019

CNN

Ultrasoun

d images

Data

was

collecte

d at

Baheya

Foundat

ion for

Treatme

nt of

Breast

Cancer

97.39%

Benzheng

Wei,

2017

BiCNN

Histopath

ology

BreaKH

dataset

97%

Sidharth S

Prakash,202

DNN

Mammogr

aphy

WBCD 99%

Nur Syahmi

Ismail,

2019

CNN

Mammogr

aphy

IRMA

dataset

94%

Mahboubeh

Jannesari,

2018

CNN

Histopath

ology

Tissue

Micro

Array(T

MA),Br

eaKHis

datase

98.7%

PhuT.

Nguyen,

2019

CNN

Histopath

ology

BreaKH

dataset

73.68%

Prediction of Diseases using Deep Learning: A Review

131

3.4 Brain Diseases

Brain is the most important part of the human body

which controls the functionality of all other organs.

Brain provides the living organisms the ability to

learn, think and make decisions. Marek Wodkinski et

al.(2019) proposed RNN-CNN based method to

convert voice recordings into spectrogram and then

use it to identify the presence of Parkinson's

disease.The observed accuracy is 90%.

Table 4 . Summary of deep learning methods in

classification of brain diseases

Author/Year Techniques Modality Source Accuracy

Amin Ul

Haq,

2018

DNN

Voice

sample

PD dataset 98%

Gaurav

Shalin,

2020

CNN F scan

Data

recorded

University

of Ottawa

95.1%

Pir

Mohammad

Shah,

2018

CNN MRI scans PPMI 96%

Mohammad

Shaban,

2020

DCNN

Handwriti

drawings

Kaggle

handwriti

ng datase

94%

Pedram

Khatamino,

2018

DCNN

Handwriti

drawings

dataset

79.64%

Gunawarden

2017

CNN MRI ADNI 84.4%

Ahmad

Waleed

Salehi,

2020

CNN MRI ADNI 99%

Ibtissam

Bakkouri,

2019

3DCNN MRI ADNI 93%

4 CONCLUSION

This paper provides an overview of various deep

learning technologies used by various researchers in

the medical field. The deep learning techniques have

shown unique capabilities in analyzing different

kinds images in medical field .Deep learning

techniques have the potential to reduce the efforts

medical personals by accurate and faster analysis of

medical images , this may help the proper treatment

of the patients.This paper presented the deep learning

based approaches for the prediction of

Gastrointestinal Diseases, Lung Disease, Breast

Cancer Diagnosis and Brain Diseases available in

literature. This paper also presented the summary for

the prediction of Gastrointestinal Diseases, Lung

Disease, Breast Cancer Diagnosis and Brain Diseases.

REFERENCES

Ahmad P. Tafti, Fereshteh S. Bashiri, Eric Larose, and

Peggy Peissig, 2018,“Diagnostic Classifification of

Lung CT Images Using Deep 3D Multi-Scale

Convolutional Neural Network”, IEEE International

Conference on Healthcare Informatics (ICHI).

Ahmad Waleed Salehi,Preety Baglat;Brij Bhushan

Sharma;Gaurav Gupta;Ankita Upadhya, 2020,“a CNN

Model: Earlier Diagnosis and Classification of

Alzheimer Disease Using MRI”, International

Conference on Smart Electronics and Communication

(ICOSEC).

Ahmed Hijab ,Muhammad a. Rushdi ,Mohammed M.

Gomaa ,Ayman Eldeib, 2019, “Breast Cancer

Classification in Ultrasound Images Using Transfer

Learning”, International Conference on Advances in

Biomedical Engineering (ICABME).

Alexey a. Shvets,Vladimir I. Iglovikov,Alexander

Rakhlin,Alexandr a. Kalinin ,2018 “Angiodysplasia

Detection and Localization Using Deep Convolutional

Neural Networks”,IEEE International Conference on

Machine Learning and Applications .

Aman Srivastava, Saurav Sengupta, Sung-Jun Kang, Karan

Kant,Marium Khan, S. Asad Ali, Sean R. Moore,

Beatrice C. Amadi, Paul Kelly, Sana Syed, Donald E.

Brown,2019, “Deep Learning for Detecting Diseases in

Gastrointestinal Biopsy Images”, Systems and

Information Engineering Design Symposium (SIEDS).

Aman Srivastava, Saurav Sengupta, Sung-Jun Kang, Karan

Kant,Marium Khan, S. Asad Ali, Sean R. Moore,

Beatrice C. Amadi, Paul Kelly,Sana Syed, Donald E.

Brown, 2019 “ Deep Learning for Visual Recognition

of Environmental Enteropathy and Celiac

Disease”,IEEE EMBS International Conference on

Biomedical & Health Informatics (BHI).

Amin Ul Haq;Jianping Li;Muhammad Hammad

Memon;Jalaluddin Khan;Salah Ud Din;Ijaz

Ahad;Ruinan Sun;Zhilong Lai,2018, “Comparative

Analysis of The Classification Performance of Machine

Learning Classifiers and Deep Neural Network

Classifier for Prediction of Parkinsons

Disease”,International Computer Conference on

Wavelet Active Media Technology and Information

Processing (ICCWAMTIP).

Amrit Sreekumar, Karthika Rajan Nair, Sneha Sudheer,

Ganesh Nayar H and Jyothisha J Nair ,2020,

“Malignant Lung Nodule Detection Using Deep

Learning”,International Conference on

Communication and Signal Processing (ICCSP).

Ankit Titoriya,Ankit Titoriya ,2019, “Breast Cancer

Histopathology Image Classification Using

ICACSE 2021 - International Conference on Advanced Computing and Software Engineering

132

Alexnet”,International Conference on Information

Systems and Computer Networks (ISCON) .

Benzheng Weil, Zhongyi Han,Xueying He,Yilong

Yin,2017, “Deep Learning Model based Breast Cancer

Histopathological Image Classification”,IEEE

International Conference on Cloud Computing and Big

Data Analysis (ICCCBDA).

Chathurika Gamage,Isuru Wijesinghe,Charith

Chitraranjan,Indika Perera,2019, “ GI-Net: Anomalies

Classification in Gastrointestinal Tract Through

Endoscopic Imagery With Deep Learning”,Moratuwa

Engineering Research Conference (Mercon) .

Chen-Ying Hung, Ching-Heng Lin, Chi-Sen Chang, Jeng-

Lin Li, and Chi-Chun Lee ,2019 “Predicting

Gastrointestinal Bleeding Events from Multimodal in-

Hospital Electronic Health Records Using Deep Fusion

Networks”, Annual International Conference of The

IEEE Engineering in Medicine and Biology Society

(EMBC).

Diksha Mhaske ,Kannan Rajeswari Ruchita Tekade,2019 ,

“Deep Learning Algorithm for Classification and

Prediction of Lung Cancer Using CT Scan

Images”, 5th International Conference on Computing,

Communication, Control and Automation (ICCUBEA)

E.Sudheer Kumar and C.Shoba Bindu,2019, “Medical

Image Analysis Using Deep Learning : a Systematic

Literature Review”, in Emerging Technologies in

Computer Engineering - Microservices in Big Data

Analytics(ICETCE) ,Pp. 81–97.

Franklin Sierra, Yesid Gutiérrez and Fabio Martínez,2020,

“an Online Deep Convolutional Polyp Lesion

Prediction Over Narrow Band Imaging (NBI)”,Annual

International Conference of The IEEE Engineering in

Medicine & Biology Society (EMBC).

Gaurav Shalin, Scott Pardoel,Julie Nantel, Edward D.

Lemaire,Jonathan Kofman, 2020 “Prediction of

Freezing of Gait in Parkinson’s Disease from Foot

Plantar -Pressure Arrays Using Convolutional Neural

Network”, Annual International Conference of The

IEEE Engineering in Medicine & Biology Society

(EMBC).

Hari Krishna Timmana,Rajabhushanam C,2020, “Breast

Malignant Detection Using Deep Learning

Model”,International Conference on Smart Electronics

and Communication (ICOSEC ).

Ibtissam Bakkouri , Karim Afdel , Jenny Benois-Pineau

and Gwenaelle Catheline ,2019, “Recognition of

Alzheimer’s Disease on Smri based on 3D Multi-Scale

CNN Features and a Gated Recurrent Fusion

Unit”International Conference on Content-based

Multimedia Indexing (CBMI).

Jasmir ,Siti Nurmaini ,Reza Firsandaya Malik,Dodo Zaenal

Abidin ,Ahmad Zarkasi ,Yesi Novaria Kunang ,Firdaus ,

2018,“Breast Cancer Classification Using Deep

Learning”, International Conference on Electrical

Engineering and Computer Science (ICECOS).

K a N N P Gunawardena , R N Rajapakse,N D Kodikara,

2017,“Applying Convolutional Neural Networks for

Pre-Detection of Alzheimer’s Disease from Structural

MRI Data”,International Conference on Mechatronics

and Machine Vision in Practice (M2VIP).

Karan Jakhar,Nishtha Hooda,2018,“Big Data Deep

Learning Framework Using Keras: a Case Study of

Pneumonia Prediction”, International Conference on

Computing Communication and Automation (ICCCA).

Kuntesh Jani ,Rajeev Srivastava ,Subodh Srivastava ,2019

“Computer Aided Medical Image Analysis for Capsule

Endoscopy Using Conventional Machine Learning and

Deep Learning”, International Conference on Smart

Computing & Communications (ICSCC) .

Mahboubeh Jannesari,Mehdi Habibzadeh,Hamidreza

Aboulkheyr,Pegah Khosravi,Olivier Elemento,Mehdi

Totonchi,Iman Hajirasouliha,2018, “Breast Cancer

Histopathological Image Classifification: a Deep

Learning Approach”, IEEE International Conference

on Bioinformatics and Biomedicine (BIBM).

Marek Wodzinski,Andrzej Skalski, Daria

Hemmerling,Juan Rafael Orozco Arroyave Elmar

Noth, ,2019,“Deep Learning Approach To Parkinson’s

Disease Detection Using Voice Recordings and

Convolutional Neural Network Dedicated To Image

Classification”,Annual International Conference of

The IEEE Engineering in Medicine and Biology Society

(EMBC).

Marleen De Bruijne,2016, “Machine Learning Approaches

in Medical Image Analysis: from Detection To

Diagnosis”, Medical Image Analysis ,Pp.94-97

Matko ˇSari´C, Mladen Russo, Maja Stella, Marjan Sikora,

2019,“CNN-based Method for Lung Cancer Detection

in Whole Slide Histopathology Images”,

International

Conference on Smart and Sustainable Technologies

(Splitech).

Mohamed Shaban,2020, “Deep Convolutional Neural

Network for Parkinson’s Disease based Handwriting

Screening”,IEEE International Symposium on

Biomedical Imaging Workshops (ISBI Workshops).

Naresh Khuriwal ,Nidhi Mishra ,2018, “Breast Cancer

Detection from Histopathological Images Using Deep

Learning”,International Conference and Workshops on

Recent Advances and Innovations in Engineering

(ICRAIE) .

Nur Syahmi Ismail,Cheab Sovuthy, 2019 “Breast Cancer

Detection based on Deep Learning Technique”,

International UNIMAS STEM 12th Engineering

Conference (Encon).

Pedram Khatamino, İsmail Cantürk, Lale Özylmaz, 2018,

“a Deep Learning-CNN based System for Medical

Diagnosis: an Application on Parkinson’s Disease

Handwriting Drawings” ,International Conference on

Control Engineering & Information Technology (CEIT).

Phu T. Nguyen ,Tuan T. Nguyen ,Ngoc C. Nguyen ,Thuong

T. Le ,2019, “Multiclass Breast Cancer Classification

Using Convolutional Neural Network” ,International

Symposium on Electrical and Electronics Engineering

(ISEE).

Pir Masoom Shah, Adnan Zeb, Uferah Shafi, Syed Farhan

Alam Zaidi, Munam Ali Shah ,2018,“Detection of

Parkinson Disease in Brain MRI Using Convolutional

Prediction of Diseases using Deep Learning: A Review

133

Neural Network”,International Conference on

Automation and Computing (ICAC).

Pradipta Sasmal,Yuji Iwahori,M.K. Bhuyan,Kunio

Kasugai,2019 “Classification of Polyps in Capsule

Endoscopic Images Using CNN”,IEEE Applied Signal

Processing Conference (ASPCON) .

Pritam Sarkar, Vandad Davoodnia, Ali Etemad,

2019,“Computer-Aided Diagnosis Using Class-

Weighted Deep Neural Network”,IEEE International

Conference on Machine Learning and Applications

(ICMLA).

Qiang Wang, Huijie Fan, Yandong Tang,2019 “Computer-

Aided WCE Diagnosis Using Convolutional Neural

Network and Label Transfer”,IEEE International

Conference on CYBER Technology in Automation,

Control, and Intelligent Systems .

Rahul Hooda, Sanjeev Sofat,Simranpreet Kaur, Ajay

Mittal,Fabrice Meriaudeau,2017, “Deep-learning: A

Potential Method for Tuberculosis Detection using

Chest Radiography”,IEEE International Conference on

Signal and Image Processing Applications ( ICSIPA ).

Ricardo Buettner,2020, “A Systematic literarure Review Of

Medical Image Analysis Using Deep Learning”,IEEE

Symposium on Industrial Electronics & Applications

(ISIEA).

Ruchika Tekade,Prof. Dr.K.Rajeshwari,2018 “Lung

Cancer Detection And Classification using Deep

Learning”,IEEE International Conference on

Computing Communication Control and Automation

(ICCUBEA).

S.Rajaraman, S. Sornapudi,M. Kohli, and S. Antani,2019

“Assessment of an ensemble of machine learning

models toward abnormality detection in chest

radiographs”, Annual International Conference of the

IEEE Engineering in Medicine and Biology Society

(EMBC).

Samira Lafraxo , Mohammed El Ansari,

2020 ,“GastroANet:Abonormalities Recognition in

Gastrointestinal Tract through Endoscopic Imagery

using Deep Learning Techniques” International

Conference on Wireless Networks and Mobile

Communications (WINCOM).

Sheikh Rafifiul Islam , Santi P. Maity, Ajoy Kumar Ray and

Mrinal Mandal, 2019,“Automatic Detection of

Pneumonia on Compressed Sensing Images using Deep

Learning”, IEEE Canadian Conference of Electrical

and Computer Engineering (CCECE) .

Sidharth S Prakash1, Visakha K, 2020, “Breast Cancer

Malignancy Prediction Using Deep Learning Neural

Networks”,International Conference on Inventive

Research in Computing Applications (ICIRCA) .

Spiros V. Georgakopoulos, Dimitris K. Iakovidis, Michael

Vasilakakis, Vassilis P. Plagianakos , Anastasios

Koulaouzidis,2016“Weakly-Supervised Convolutional

Learning for Detection of Inflammatory

Gastrointestinal Lesions” IEEE International

Conference on Imaging Systems and Techniques (IST) .

Tonmoy Ghosh, Linfeng Li, and Jacob Chakareski,

2018,“Effective Deep Learning for Semantic

Segmentation Based Bleeding Zone Detection in

Capsule Endoscopy Images”, IEEE International

Conference on Image Processing (ICIP).

Yaxing Cao, Wenming Yang, Kaiquan Chen, Yong Ren,

Qingmin Liao , 2018,“Capsule Endoscopy Image

Classification with Deep Convolutional Neural

Networks”,

IEEE International Conference on

Computer and Communications .

ICACSE 2021 - International Conference on Advanced Computing and Software Engineering

134