IMPROVING PREDICTION PERFORMANCE OF

DIFFERENT STAGES OF SEPSIS USING MEACHINE

LEARNING ALGORITHM

A PROJECT REPORT

Submitted by

VISHVASUNDAR S 312321104196
SAKTHI VAIBHAV M 312321104144

in partial fulfilment of the award of the degree of

BACHELOR OF ENGINEERING
in
COMPUTER SCIENCE AND ENGINEERING

St. JOSEPH’S COLLEGE OF ENGINEERING

(An Autonomous Institution)
OMR, Chennai 600 119
ANNA UNIVERSITY :: CHENNAI 600 025
APRIL 2024
 ANNA UNIVERSITY, CHENNAI

BONAFIDE CERTIFICATE

Certified that this project report “IMPROVING PREDICTION

PERFORMANCE OF DIFFERENT STAGES OF SEPSIS USING
MEACHINE LEARING ALGORITHM” is the bonafide work of
VISHVASUNDAR S (312321104196) AND SAKTHI VAIBHAV M
(312321104144) who carried out the work under my guidance. Certified further
that to the best of my knowledge the work reported herein does not form part of
any other thesis or dissertation on the basis of which a degree or award was
conferred on an earlier occasion on this or any other candidate.

SIGNATURE SIGNATURE

HEAD OF THE DEPARTMENT SUPERVISOR

Dr.V Muthulakshmi, M.E., Ph.D., Mrs.Jeipratha. P. N, M.E.,(Ph.D),
Professor & Head of Department Assistant Professor,
Dept. of Computer Science and Engineering, Dept. of Computer Science and Engineering,,
St. Joseph’s college of Engineering, St. Joseph’s college of Engineering,
OMR,Chennai-600 119 OMR Chennai-600 119

Submitted to Project and Viva Examination held on __________________

INTERNAL EXAMINER EXTERNAL EXAMINER

II
 ACKNOWLEDGEMENT

At the outset, we would like to express our sincere gratitude to our beloved
Dr. B. Babu Manoharan M.A., M.B.A., Ph.D., Chairman, St. Joseph’s Group of
Institutions for his constant guidance and support to the student community and the
Society.

We would like to express our hearty thanks to our respected

Managing Director Mr. B. Shashi Sekar, M.Sc. for his kind encouragement and
blessings.

We wish to express our sincere thanks to the

Executive Director Mrs. S. Jessie Priya M.Com. for providing ample facilities in
the institution.
We express sincere gratitude to our beloved
Principal Dr. Vaddi Seshagiri Rao M.E., M.B.A., Ph.D., F.I.E. for his
inspirational ideas during the course of the project.
We express our sincere gratitude to our beloved
Dean (Research) Dr. A. Chandrasekar M.E., Ph.D., Dean (Student Affairs)
Dr. V. Vallinayagam M.Sc., M.Phil., Ph.D., and Dean (Academics)
Dr. G. Sreekumar M.Sc., M.Tech., Ph.D., for their inspirational ideas during the
course of the project.

We wish to express our sincere thanks to Dr. V Muthulakshmi, M.E., Ph.D.,

Head of the Department, Department of Computer Science and Engineering,
St. Joseph’s College of Engineering for her guidance and assistance in solving the
various intricacies involved in the project.

We would like to acknowledge our profound gratitude to our supervisor

Mrs. Jeipratha. P.N, M.E.,(Ph.D), for her expert guidance and connoisseur
suggestion to carry out the study successfully.

Finally, we thank the Faculty Members and our Family, who helped and
encouraged us constantly to complete the project successfully.

III
 ABSTRACT

Sepsis, which is known to be as Septicemia, is a life-threatening condition that

arises when the body's response to infection causes injury to its tissues and organs,

later on leads to death. Sepsis is an inflammatory immune response

triggered by an infection. Undetected sepsis can progress to severe sepsis and

septic shock. The risk of death from sepsis is as high as 30%, as high as 50% from

severe sepsis, and up to 80% from septic shock. Sepsis requires immediate

treatment with intravenous fluids and antimicrobials. Delay in predicting and

identifying the various stages is the major concern. It may at least take two to three

days to get the result. By the time the organs starts to dysfunction or tissue

hypoperfusion. Children, particularly newborns and young infants, can be more

susceptible in developing Sepsis. In existing system, the medical teams diagnose

the disease from the patient’s medical history, symptoms, a physical exam and test.

This approach promotes the early prediction of various sepsis levels and could act

as prognostic stratification system.

IV
 TABLE OF CONTENTS

S.NO TITLE PAGE NO.

ABSTRACT i

LIST OF FIGURES viii

LIST OF ABBREVATIONS ix

1 INTRODUCTION 1

1.1 GENERAL INTRODUCTION - SEPSIS 1

1.2 DOMAIN OVERVIEW 2

1.2.1 ARTIFICIAL INTELLIGENCE 2

1.2.2 UNDERSTANDING THE IMPORTANCE OF AI 2

1.3 FUTURE TECHNOLOGY OF INTELLIGENCE 3

1.3.1 USE OF MULTIPLE TECHNOLOGIES IN ML 4

1.3.1.1 PERSONALIZED COMPUTING ENVIRONMENT 4

1.3.1.2 TUNED RECOMMENDATION ENGINE 4

1.3.1.3 CLASSIFICATION AND REGRESSION 4

FOR PREDICTIVE ANALYSIS

1.3.1.4 CLUSTER ANALYSIS 4

1.4 MACHINE LEARNING - ELEMENTS 5

1.5 CHARACTERSTICS OF MACHINE LEARNING 5

1.6 ADVANTAGES OF MACHINE LEARNING 5

1.6.1 IDENTIFYING TRENDS AND PATTERNS 5

V
1.6.2 LACK OF HUMAN INTERVENTION (AUTOMATION) 5

VI
 1.6.3 CONTINUOUS IMPROVEMENT 6

1.6.4 HANDLING MULTI-DIMENSIONAL AND MULTI- 6

VARIENT DATA

1.6.5 WIDE APPLICATIONS 6

2 LITERATURE SURVEY 7

3 SYSTEM CONCEPT 16

3.1 EXISTING SYSTEM 16

3.1.1 DISADVANTAGES 16

3.2 PROPOSED SYSTEM 17

3.2.1 ADVANTAGES 17

4 AIM AND SCOPE OF THE PRESENT INVESTIGATION 18

4.1 FUTURE SCOPE 18

4.2 NEED OF THE PROJECT 18

4.2.1 PROBLEM DEFINITION 18

4.2.2 MODULES 18

4.2.2.1 DATA COLLECTION AND PREPROCESSING 18

4.2.2.2 SEPSIS AND SEVERE SEPSIS MODEL 19

4.2.2.3 SEPTIC SHOCK AND INTERFACE WITH UI 19

4.2.3 WEB DEVELOPMENT MODULES 19

4.2.3.1 SEPSIS PREDICTION 20

4.2.3.2 SEVERE SEPSIS PREDICTION 20

4.2.3.3 SEPTIC SHOCK PREDICTION 20

VI
 4.2.3.4 HOSPITAL SUGGESTION 20

4.2.4 MODEL BUILDING 20

4.2.4.1 SUPERVISED LEARNING 21

4.2.4.2 UNSUPERVISED LEARNING 22

4.2.4.3 SEMI-SUPERVISED LEARNING 22

4.2.4.4 REINFORCEMENT LEARNING 23

5 IMPLEMENTATION 24

5.1 METHODLOGIES 24

5.1.1 RANDOM FOREST 24

5.1.2 NEURAL NETWORK 24

5.2 ACCURACY CALCULATION 25

5.2.1 CONFUSION MATRICES 25

5.2.2 FEASIBLITY STUDY 27

5.2.2.1 ECONIMICAL FEASIBILITY 27

5.2.2.2 TECHINICAL FEASIBLITY 27

5.2.2.3 SOCIAL FEASIBLITY 27

5.2.3 REQUIREMENT SPECIFICATION 28

5.2.3.1 HARDWARE REQUIREMENTS 28

5.2.3.2 SOFTWARE REQUIREMENTS 28

6 PERFORMANCE ANALYSIS 29

6.1 SYSTEM ARCHITECTURE 29

6.2 PYTHON 29

VII
 6.3 ANACONDA 30

6.4 NUMPY 32

6.5 PANDAS 33

6.6 SEPSIS MODEL 35

6.7 FLASK CODE 37

6.8 UML DIAGRAMS 39

6.8.1 USECASE DIAGRAM FOR SEPSIS PREDICTION 40

6.8.2 CLASS DIAGRAM FOR SEPSIS PREDICTION 40

6.8.3 SEQUENCE DIAGRAM 41

6.8.4 ACTIVITY DIAGRAM 41

6.8.5 DATAFLOW DIAGRAM 43

7 SYSTEM ANALYSIS 44

7.1 UNIT TESTING 44

7.2 INTEGRATION TESTING 44

7.3 FUNCTIONAL TEST 45

7.4 WHITE BOX TESTING 45

7.5 BLACK BOX TESTING 45

8 SCREENSHOTS 46

9 CONCLUSIONS 52

10 REFERENCES 53

VIII
 LIST OF FIGURES

FIG NO. TITLE PAGE NO.

3.2.1 CLASSIFICATION 21

3.2.2 REGRESSION 21

5.5.1 SYSTEM ARCHITECTURE 29

xi