AI and ML

L1
Dr. Sujata Kulkarni
Associate Professor, EXTC

19-09-2024 1
 Teaching Scheme (Hrs/week) Credits Assigned

Course L T P O E L T P Total
(Category) Course Name
Code

2 - 2 2 6 2 - 1 3

Examination Scheme
PE-IV
Component ISE MSE ESE Total
Artificial Intelligence In
Healthcare
Theory 50 50 100 200

IT24
Laboratory 50 -- 50 100

19-09-2024 2
19-09-2024 3
19-09-2024 4
19-09-2024 5
19-09-2024 6
References

19-09-2024 7
19-09-2024 8
Course Plan
Credits No. of Hr Total hrs
2/week 2 2 Hr/week 26-28 hrs
First 3 Module are based on fundamental 2 hrs/week 14hrs
statistics and Machine Learning

Next 3 modules are based Healthcare Data 2hrs/week 14hrs

representation and analysis

Laboratory 1 2Hrs 26-30hrs

Application based Laboratory 2hrs 6hrs

Assignment based on Theory 2hrs 8hrs

/presentation based on Laboratory 2hrs 8hrs
19-09-2024 9
 Evaluation Scheme Continuous evaluation Marks
Theory
Continuous evaluation : 60 to 80 % ISE1 Assignments/tutorial/ 10
End semester evaluation : 20-40 % Tutorial /quiz 10

MSE 30
ESE 100
Laboratory- ISE1 40
Performance of Experiment as per Rubrics mentioned
Data creation, Analysis, Performance , documentation
ISE2 Debate cum discussion/Oral 30

LAB ESE : Practical Exam 30

19-09-2024 10
 Thank you

19-09-2024 11
 Introduction to Machine
Learning
L1
Dr. Sujata Kulkarni

19-09-2024 12
 ISTE approved a one week STTP on "Machine Learning and
19-09-2024 13
Data Science Application"
19-09-2024 14
ISTE approved a one week STTP on "Machine Learning and Data Science Application"
 ISTE approved a one week STTP on "Machine Learning and
19-09-2024 15
Data Science Application"
 ISTE approved a one week STTP on "Machine Learning and
19-09-2024 16
Data Science Application"
 ISTE approved a one week STTP on "Machine Learning and
19-09-2024 17
Data Science Application"
Importance of ML

ISTE approved a one week STTP on "Machine Learning and

19-09-2024 18
Data Science Application"
 ISTE approved a one week STTP on "Machine Learning and
19-09-2024 19
Data Science Application"
Manual coding

ISTE approved a one week STTP on "Machine Learning and

19-09-2024 20
Data Science Application"
19-09-2024 21
Types of ML Algorithm
➢ Supervised Learning

➢ Unsupervised Learning

➢ Reinforcement Learning

19-09-2024 22
19-09-2024 23
19-09-2024 24
Hybrid Model

19-09-2024 25
19-09-2024 26
19-09-2024 27
19-09-2024 28
 • OTT

• High quality streaming

• ML to provide movie
recommendations and using AI
to guarantee high-quality
streaming even at reduced
bandwidths.

• Netflix advertisement

19-09-2024 29
 “tag suggestions” or “face
recognition” turned on in
Facebook (this means you have
provided permission for Facial
Recognition), then the Machine
Learning System analyses the
pixels of the face in the image and
creates a template which is a
string of numbers.

The Instagram machine learning algorithms use certain user activities for insights that Instagram refers to
as "signals". Signals include what a person posted, when and how often a person posts content, user
preferences towards particular types of content, etc.

19-09-2024 30
Ex NLP
Understand the human language and then reply to them appropriately according to the tone of the user and
the context he is referring to it.
These conversational AI bots are more advanced than regular chatbots that are programmed with answers to
certain questions.

19-09-2024 31
 Similarity-based: Filters compare incoming emails with pre-existing emails stored in servers.
Sample-based: Templates of legitimate and non-legitimate spam emails allow AI to assess new
emails.
Use specific filtering methods to identify the content of emails or their senders and then flag the
email as spam. EX Gmail

19-09-2024 32
19-09-2024 33
➢ Machine Learning:

It is field of study that gives the computers capability to learn without

being explicitly programmed.

Example :

• Online Shopping

• Web Browsing

Explicitly: Without human interference

Machine learned the program and give output i.e computer learn itself

19-09-2024 34
• Traditional Program

• Machine Learning

Example: Student Exam

Data, output,Programm

Model: Student Brain

• Train Phase
• Test Phase

Objective of ML (less error margin)

Ideal output=actual output
19-09-2024 35
19-09-2024 36
19-09-2024 37
19-09-2024 38
19-09-2024 39
19-09-2024 40
 Applications
➢ Medicine: Diagnosis a diseases

➢ Vision

➢ NLP

➢ Speech Recognition

➢ Business Intelligence
▪ Forecasting product sale
▪ Promotion opportunity

➢ Others
▪ Fraud detection
▪ Defaulter on a home mortgage etc
19-09-2024 41
1.Define Machine Learning. Discuss with examples why machine learning is important.

2. Discuss with examples some useful applications of machine learning.

3. Explain how some areas/disciplines that influenced the machine learning.

4. What do you mean by a well–posed learning problem? Explain the important features that are
required to well–define a learning problem.

5. Define learning program for a given problem. Describe the following problems with respect to
Tasks, Performance and Experience:
a. Checkers Learning Problems
b. Handwritten Recognition Problem
c. Robot Driving Learning Problem

6. Describe in detail all the steps involved in designing a learning system.

7. Discuss the perspective and issues in machine learning.

19-09-2024 42
 Stages /Process of ML

19-09-2024 43
 ISTE approved a one week STTP on "Machine Learning and
19-09-2024 44
Data Science Application"
Steps of Machine Learning

ISTE approved a one week STTP on "Machine Learning and

19-09-2024 45
Data Science Application"
19-09-2024 46
19-09-2024 47
19-09-2024 48
19-09-2024 49
19-09-2024 50
19-09-2024 51
19-09-2024 52
19-09-2024 53
19-09-2024 54
 ISTE approved a one week STTP on "Machine Learning and
19-09-2024 55
Data Science Application"
 Techniques:
Better Feature means: ▪ Imputation
▪ Flexibility ▪ Handling Outlier
▪ Simple model ▪ Log Transform
▪ Better result ▪ Encoding
▪ Scaling

19-09-2024 56
ISTE approved a one week STTP on "Machine Learning and Data Science Application"
 ISTE approved a one week STTP on "Machine Learning and
19-09-2024 57
Data Science Application"
19-09-2024 ISTE approved a one week STTP on "Machine Learning and Data Science Application" 58
 ISTE approved a one week STTP on "Machine Learning and
19-09-2024 59
Data Science Application"
 Test Your Skill
on real world Application

ISTE approved a one week STTP on "Machine Learning and

19-09-2024 60
Data Science Application"
Email Filtering satellite images to
identify climate
anomalies.
Credit Scoring

Spam Filtering
House Price Prediction
Voice Recognition Credit Scoring
Face Recognition

customer segmentation,
recommender systems,
target marketing campaigns Game Playing
Robot Navigation

19-09-2024 61
 ISTE approved a one week STTP on "Machine Learning and
19-09-2024 62
Data Science Application"
19-09-2024 63
19-09-2024 64
ISTE approved a one week STTP on "Machine Learning and Data Science Application"
Questions based on Learned topic
What is machine learning?
2. Define supervised learning
3. Define unsupervised learning
4. Define semi supervised learning
5. Define reinforcement learning
6. Example of algorithm of each learning type.

19-09-2024 65
19-09-2024 66
19-09-2024 67
19-09-2024 68
 DEMO

19-09-2024 69
19-09-2024 70
19-09-2024 71
 THANK YOU

19-09-2024 72
Artificial Intelligence for Healthcare

By :
Dr. Sujata Kulkarni
➢ What is AI?
➢ AI in Healthcare
➢ Use cases
➢ ML and DL
Use cases

➢ Managing Medical Data

➢ Early Detection
➢ Medical assistant
▪ AI in healthcare uses Cognitive Technology to unwind huge amount of medical records and perform
powerful diagnosis.
▪
Example Nuance is production Service provider that uses AI and ML in order to predict the intent of
particular user .
MIT Team develop NN Volex
Morph for diagosis

Train 7000 MRI

Saved Life of Scott Gillan
Summary
➢ Benefits of AI in Healthcare:
1. Improved Accuracy: Enhanced diagnostic accuracy through machine learning algorithms.
2. Cost Reduction: Lower costs due to automated processes and efficient resource utilization.
3. Faster Decision Making: Quick analysis of complex data sets aids in faster decision-making.
4. Better Patient Outcomes: Personalized treatment and continuous monitoring lead to improved patient
outcomes.

➢ Challenges in AI Integration:
▪ Data Privacy: Ensuring patient data privacy and compliance with regulations.
▪ Ethical Considerations: Addressing ethical issues related to AI decision-making in healthcare.
▪ Implementation Costs: High initial costs for AI integration and infrastructure.
▪ Skill Gap: Need for skilled professionals to develop, implement, and manage AI systems.

➢ Future of AI in Healthcare:
AI is expected to continue evolving, offering more sophisticated tools for diagnostics, treatment, and patient
care. Emphasis on integrating AI seamlessly with existing healthcare systems for enhanced functionality.
How is technology changing the healthcare sector? - YouTube