Changes for page Step 4: Claims

Summary

• Page properties (2 modified, 0 added, 0 removed)

Details

Page properties

Author

...	...	@@ -1,1 +1,1 @@
1		-XWiki.ShambhawiPal
	1	+XWiki.varunsingh3000

Content

...	...	@@ -33,8 +33,12 @@
33	33
34	34	**Measurement of negative consequences:**
35	35
36		-1. To measure if the patient is frustrated with the tools, interaction of the patient with the robot can be monitored. Emotion recognition can be used to see the patient's attitude towards the robot. (Emotions such as Anger, fear, disgust, happiness, sadness, surprise and neutral can be analyzed)
37		-1. {{mention reference="XWiki.varunsingh3000" style="FULL_NAME" anchor="XWiki-varunsingh3000-okraab"/}} we need to disciss the rest of the points
	36	+1. To measure if the patient is frustrated with the tools, the interaction of the patient with the robot can be monitored. Emotion recognition can be used to see the patient's attitude towards the robot.(Emotions such as Anger, fear, disgust, happiness, sadness, surprise and neutral can be analyzed)
	37	+1. This can be achieved using emotion recognition when the PwD shows negative emotions like Anger, Fear, Disgust and Sadness for the majority of the session.
	38	+1. If the response time for a certain question/prompt is too long (find the general response time for PwD) then automatically call the caretaker as this would indicate that something has gone wrong.
	39	+1. After the session, the robot can ask for feedback regarding the preference of whether they liked painting with their caretaker or the robot. E.g. Do you love me? **(Robot likeability feedback: robot likeability in comparison to their caretaker)**
	40	+1. There would be a button on the screen displaying a statement like "Click here to make me act more human", which would keep a count of how many times the button was pressed.  **(**Measurement: **Humaness of the robot**)
	41	+1. TBD
38	38	)))
39	39	|(((
40	40	[[image:14.png]]
...	...	@@ -43,7 +43,7 @@
43	43	)))|For each measurement, what are the benchmarks (criteria)? (i.e., what are desired values?)|(((
44	44	Benchmark criteria for positive consequences:
45	45
46		-1. Amount of time saved by the caretaker per day: We aim to save atleast 45 minutes of the caretaker's time per patient per day.
	50	+1. Amount of time saved by the caretaker per day: We aim to save at least 45 minutes of the caretaker's time per patient per day.
47	47	1. Painting session feedback: We aim to get a rating of more than 3.
48	48	1. Sense of identity feedback: We aim to get a rating of more than 3.
49	49	1. Conversations with relative feedback: We aim to get a rating of more than 3.
...	...	@@ -52,6 +52,13 @@
52	52	1. Relative's ease of use feedback: We aim to get a rating of more than 3.
53	53
54	54	Benchmark criteria for negative consequences:
	59	+
	60	+1. Preference regarding the tool and interface: We aim to have this criterion triggered less than 3 times during a session.
	61	+1. Patient's state of mind: We aim to have this criterion triggered less than 3 times during a session.
	62	+1. Session emergency: Since the response time exceeded a certain threshold, we call the caretaker. (We aim to have this criterion triggered less than 2 times in a session.)
	63	+1. Opinion of the patient regarding the change towards a robot: We aim to get a rating of more than 3. (We aim to have this criterion triggered less than 2 times in a session.)
	64	+1. Humaness of the robot: We aim to get a rating of less than 3.
	65	+1. TBD
55	55	)))
56	56	|(((
57	57
...	...	@@ -64,9 +64,24 @@
64	64
65	65	1. The robot will help the patient paint. Moreover, the robot will be humble and patient with the Pwd, this will create a sense of comfort for the Pwd and then they will be willing to spend more time with the robot. Thus the caretaker can get the extra time in this case.
66	66	1. To get a good feedback of the painting session, the robot will try to keep up a healthy conversation with the patient and be supportive of their thoughts and actions. With respect to painting, the robot can also guide the patient by showing a picture of the painting that the patient would like to draw, for example an apple. The robot can also discuss different colors that can be used at any given time to help the patient. The robot can also make the patient see the outside weather conditions and make them draw the things outside. In this way, the robot patient interaction improves and the patient also gets more involved.
67		-1. {{mention reference="XWiki.varunsingh3000" style="FULL_NAME" anchor="XWiki-varunsingh3000-z0j9ws"/}} need to refer to a paper on how art improves condition of patients with dementia
68		-1. same as 3
69		-1. same as 3
	78	+1. The agent first interacts with the daughter to get information about the patient and then the agent helps the patient draw the similar item. For instance, if the patient loves playing basketball, then the robot can make the patient draw a basketball. This helps the patient identify the things that they like and help in restoring parts of their identity.
	79	+1. According to [[https:~~/~~/www.ncbi.nlm.nih.gov/pmc/articles/PMC6025004/>>https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6025004/]] art helps in improving the quality of life of people with dementia. Thus the overall conversations and relations between the father and daughter will improve too.
70	70	1. To improve the feedback on Relative's state of mind, the robot only asks the daughter for certain memories or information about the patient. The robot then conveys all the information to the patient while helping them paint. Thus, the daughter does not have to worry about her father to a certain extent.
71		-1. To improve the ease of use for the relative, we have a small conversation with the daughter where we narrow down the possibilities of patients' likes and dislikes. For example, the robot can directly ask the daughter about the hobbies of the patient. Since the topics are already narrowed down, the daughter does not have to think much in providing the information.
	81	+1. To improve the ease of use for the relative, we have a small conversation with the daughter where we narrow down the possibilities of patients' likes and dislikes. For example, the robot can directly ask the daughter about the hobbies of the patient. Since the topics are already narrowed down, the daughter does not have to think much about providing the information.
	82	+
	83	+Demonstration of the negative consequences listed above:
	84	+
	85	+ 1. To measure if the patient is frustrated with the tools, the interaction of the patient with the robot can be monitored. Emotion recognition can be used to see the patient's attitude towards the robot. We calculate the number of times negative emotions are captured and if it exceeds the threshold of 10, then we can call the caretaker.
	86	+
	87	+ 2. To robot might not be aware of the patient's state of mind outside of the sessions, so to predict their demeanour at the moment we can use emotion recognition here as well. Emotion recognition can be used to see the patient's attitude towards the session. We calculate the number of times negative emotions are captured and if it exceeds the threshold of 10, then we can call the caretaker.
	88	+
	89	+ 3. If the response time for a certain question/prompt is too long (find the general response time for PwD) then automatically call the caretaker as this would indicate that something has gone wrong.
	90	+
	91	+ 4. After the session, the robot can ask for feedback regarding the preference of whether they liked painting with their caretaker or the robot.
	92	+
	93	+ 5. There would be a button on the screen displaying a statement like "Click here to make me act more human", which would keep a count of how many times the button was pressed.
	94	+
	95	+ 6. TBD
	96	+
	97	+
72	72	)))