Table 3 Disease states and recommended actions in each state.

From: Computational medication regimen for Parkinson’s disease using reinforcement learning

Disease state

Estimated Total UPDRS at \({t}_{2}\)

Actions

clinicians

AI

State

Subtype

Total UPDRS at \({t}_{1}\)

% UPDRS change

Age

L

D

O

L

D

O

0

AR

( , 8]

  

10.2

  

×

  

×

1

AR

[9, 11]

[− 25.6, )

 

11.9

 

×

    

2

AR

[12, 14]

[− 25.6, )

 

14.7

 

×

×

×

×

×

3

AR

[15,17]

[− 12.7, )

 

17.0

 

×

  

×

 

4

AR

[9,14]

( , − 25.6]

 

17.1

×

 

×

×

 

×

5

AR

[18,19]

[− 12.7, )

 

19.3

 

×

×

 

×

×

6

AR

[20,24]

[16.2, )

 

20.0

×

 

×

×

×

 

7

AR

[15,19]

( , − 12.7]

 

21.9

 

×

×

×

×

 

8

AR

[20,24]

(− 14.6, 16.228]

[65, )

23.4

×

×

×

×

×

×

9

AR

[20,24]

(− 14.6, 16.228]

( , 64]

24.0

×

×

  

×

 

10

AR

[25,33]

(39.6, )

 

25.1

 

×

×

 

×

×

11

AR

[20,24]

( , − 14.6]

 

26.6

×

×

×

×

×

×

12

AR

[25,28]

( − 7.28, 11.8]

 

27.5

×

×

×

 

×

 

13

AR

[25,33]

( 11.8, 39.6]

 

27.7

×

×

 

×

×

 

14

AR

[25,28]

( , − 7.28]

 

29.5

×

×

×

×

 

×

15

AR

[29,33]

(− 3.1, 11.8]

 

30.5

×

×

×

×

×

×

16

AR

[34,39]

(10.6, )

 

31.3

×

×

×

×

×

×

17

AR

[29,33]

( , − 3.1]

 

32.5

×

×

×

×

×

×

18

AR

[34,39]

( , 10.6]

 

36.8

 

×

×

 

×

×

19

AR

[40,45]

  

37.9

×

×

×

 

×

 

20

AR

[46, )

  

46.5

×

×

×

×

×

×

21

Mixed

( ,16]

  

13.5

 

×

×

×

  

22

Mixed

[17,24]

  

21.3

×

×

×

×

  

23

Mixed

[25, )

  

32.1

×

 

×

×

 

×

24

TD

( , 14]

  

11.8

 

×

  

×

 

25

TD

[15,19]

  

19.0

×

 

×

×

×

 

26

TD

[20,28]

  

23.0

×

 

×

×

 

×

27

TD

[29, )

  

33.1

×

×

×

×

×

×

  1. L Levodopa, D dopamine agonist, O other medicine, AR Akinetic-rigid, TD tremor-dominant.
Back to article page