Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Trends in Diagnostics

No description
by

Chang-Hoon Nam

on 9 April 2018

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Trends in Diagnostics

Trends in Diagnostics
Part2: Precision diagnostics
Part1: Current issue on medical diagnosis

AI vs Doctors
What is the
main topics

of current molecular diagnostics?
POC
(point of care)
By 2020
,
20%

of the computers will learn
not only to process
but also work and manage things like humans.
By 2017
,

65%

of data center capacity will be private.
By 2020
, around

50 billion devices

are going to be connected.
Supply chain will be revolutionized by 3D printing.
(from IoT world Europe 2015)
combined with
IoT
detection limit:
low amount of marker
sampling(non-invasive)
biomarker
development
Early detection
of cancer
real-time check
daily check
What is the diagnosis?
medical diagnosis



molecular diagnostics
the identification
of the nature or the cause
of certain phenomenon
the
process
of determining
which disease explains a person's symptoms
the collection of techniques
used to
diagnose and monitor disease
,
detect risk,

decide which therapies

will work best for individual patients.
1. Do you agree with
the view that AI will
replace
doctors?
2. What do you think
about philosopher
G Ryle's comment
on the knowledge?
In 1945, the British philosopher
Gilbert Ryle

gave an influential lecture
about two kinds of knowledge.
A child knows that a bicycle has two wheels,
that its tires are filled with air, and that you ride the contraption by pushing its pedals forward in circles.
Ryle termed this kind of knowledge—
the factual, propositional kind—“knowing what.”

But to learn to ride a bicycle involves
another realm of learning.
A child learns how to ride by falling off,
by balancing herself on two wheels,
by going over potholes.
Ryle termed this kind of knowledge—
implicit, experiential, skill-based—“knowing how.”
3. How can AI improve by itself
its ability to diagnose?
“Imagine an old-fashioned program to identify a dog,” he said. “A software engineer would write a thousand if-then-else statements: if it has ears, and a snout, and has hair, and is not a rat . . . and so forth, ad infinitum. But that’s not how a child learns to identify a dog, of course. At first, she learns by seeing dogs and being told that they are dogs. She makes mistakes, and corrects herself. She thinks that a wolf is a dog—but is told that it belongs to an altogether different category. And so she shifts her understanding bit by bit: this is ‘dog,’ that is ‘wolf.’

The machine-learning algorithm
,
like the child, pulls information from a training set that has been classified.
Here’s a dog, and here’s not a dog. It then extracts features from one set versus another. And, by testing itself against hundreds and thousands of classified images, it begins to create its own way to recognize a dog—again, the way a child does.”

It just knows how to do it.
Drivers
of
development
fixed marker
sensitivity
selectivity
portability
robustness
why is it
important?
MERS
MERS
In which case
is it necessary?
1. why is
the
early detection
of cancer
important?
survival rate
optimal therapy
2. Then why is it
so difficult

to develop early detection marker?
technology platform

no clear boundary

bioinformatic issue
Technology platform:
proteomic toolkits
MS
SRM
Multiplexed
ELISA
collision induced dissociation
particular mass selection
Bioinformatic
considerations
3. individual biomarker
vs

biomarker signature
LOOVC
Kullback-Leibler
divergence
Support
vector
machine
defining boundary
in a statistical way
determining
optimal number
of biomarker
Power of
biomarker signature
from
genomics
sporadic
breast tumor
70-gene signature
83% accuracy
poor
prognosis
decision
of
immediate
chemotherapy
FDA
approved
from combining
genomics and proteomics
prostate
cancer
PSA
232 SNPs
6 protein markers
5 clinical variables
ROC AUC:
0.74
PSA alone
ROC AUC: 0.56
Proteomic
biomarker signatures
ovarian
cancer
area under the curve
OVA1
CA125
transthyretin
APOA1
b2-microglobulin
transferrin
identified
by
SELDI-TOF &
immunoassay
ROC AUC:
0.9
ROC AUC
0.6-0.7: poor
0.7-0.8: fair
0.8-0.9: good
0.9-1.0: excellent
detect in an early stage
with
91.4% accuracy
CA125 alone
65.7%
Daily check
Full transcript