Question 1:
The text file "q3_train.csv" contains 1001 lines with data for Question 3. The first line contains column headers that may be interpreted as follows:
id: observation identifier.
t1: measurement on test 1; t2: measurement on test 2.
t3: measurement on test 3; t4: measurement on test 4.
t5: measurement on test 5; t6: measurement on test 6.
t7: measurement on test 7; t8: measurement on test 8.
d: binary output variable set to 1 if product is defective and 0 otherwise.
The next 1000 lines contain1000examples, for which the values of the above features are specified.
The table below reproduces the first two observations.
|
id
|
t1
|
t2
|
t3
|
t4
|
t5
|
t6
|
t7
|
t8
|
d
|
|
1
|
84
|
8
|
64
|
6
|
94
|
36
|
51
|
21
|
1
|
|
2
|
39
|
67
|
61
|
77
|
80
|
35
|
89
|
80
|
1
|
Use the given examples to come up with a small set of rules that correctly classify the output variable "d" based on input variable values (t1, t2, t3, t4, t5, t6, t7, and t8).
Specify the rules and comment on your classification accuracy.
Then use the rules to predict the output class d for the following 20 test cases (presented in the file "q3_test.csv"):
|
Test_case
|
t1
|
t2
|
t3
|
t4
|
t5
|
t6
|
t7
|
t8
|
d
|
|
1
|
8
|
86
|
55
|
53
|
36
|
12
|
82
|
19
|
|
|
2
|
22
|
36
|
80
|
69
|
90
|
33
|
22
|
6
|
|
|
3
|
74
|
26
|
32
|
26
|
38
|
52
|
63
|
12
|
|
|
4
|
66
|
71
|
71
|
52
|
42
|
88
|
89
|
70
|
|
|
5
|
55
|
72
|
61
|
41
|
91
|
39
|
50
|
96
|
|
|
6
|
34
|
58
|
22
|
84
|
84
|
61
|
95
|
57
|
|
|
7
|
23
|
70
|
39
|
65
|
16
|
71
|
96
|
78
|
|
|
8
|
9
|
19
|
67
|
43
|
2
|
20
|
92
|
3
|
|
|
9
|
6
|
71
|
20
|
6
|
27
|
58
|
6
|
22
|
|
|
10
|
68
|
40
|
86
|
82
|
82
|
44
|
61
|
48
|
|
|
11
|
84
|
14
|
88
|
68
|
62
|
92
|
52
|
98
|
|
|
12
|
46
|
78
|
62
|
81
|
23
|
55
|
92
|
20
|
|
|
13
|
38
|
44
|
63
|
33
|
42
|
87
|
34
|
92
|
|
|
14
|
65
|
56
|
30
|
7
|
68
|
50
|
51
|
10
|
|
|
15
|
43
|
98
|
81
|
38
|
87
|
46
|
40
|
74
|
|
|
16
|
44
|
24
|
73
|
93
|
52
|
23
|
79
|
19
|
|
|
17
|
88
|
70
|
29
|
14
|
71
|
22
|
9
|
15
|
|
|
18
|
15
|
98
|
13
|
80
|
38
|
86
|
95
|
92
|
|
|
19
|
15
|
5
|
92
|
22
|
100
|
13
|
16
|
63
|
|
|
20
|
25
|
49
|
6
|
89
|
72
|
36
|
32
|
84
|
|
Question 2:
Problem - Patients Receiving Misdiagnoses Leading To Hospital Readmissions Nicholas Sazy, RN-BC, MS
Patient misdiagnoses have been an ongoing problem leading to hospital readmissions.
A misdiagnosis is defined as a diagnostic error occurring during an initial assessment detected by either a subsequent assessment or lab test.
Study by Johns Hopkins Researchers found number of patients each year given a misdiagnosis is approximately 80,000 - 160,000.
Diagnostic errors were found to be responsible for the largest degree of claims and payouts ($38.8 billion between 1986 and 2010) in addition to the highest amount of harm to patients.
According to the medical malpractice payment data from the National Practitioner Data Bank, of the 350, 706 paid claims, misdiagnoses accounted for the highest amount over all other medical errors at 28.6 percent.
(Tehrani, Lee, Shore, Makary & Pronovost, 2013)
Due to the ongoing issue with patient misdiagnoses, clinical decision support systems with embedded artificial intelligence (AI) must be utilized within electronic medical records (EMR's) in clinical environments across the country.
Use of such clinical decision support systems has become mandatory per the guidelines instituted for Meaningful Use by the American
Recovery and Reinvestment Act (ARRA) of 2009.
AI based systems must be integrated with clinical data and information to produce clinically relevant knowledge to providers.
Such knowledge can be presented to the provider containing diagnostic suggestions for the provider to choose from based on the clinical data entered into the EMR system.
Provider can choose an more focused diagnosis allowing for improved medical treatments, lower lengths of stays and reduced costs resulting in improved patient outcomes.
References
Castaneda, C. Nalley, K., Mannion, C., Bhattacharyya, P., Blake, P., Pecora, A., Goy, A. & Suh, K. S. (2015). Clinical decision support systems for improving diagnostic accuracy and achieving precision medicine. Journal of Clinical Bioinformatics, 5(4), pp. 1-16.
Tehrani, A. S. S., Lee, H., Mathews, S. C., Shore, A., Makary, M. A. & Pronovost, P. J. (2013). Diagnostic errors more common, costly and harmful than treatment mistakes. Retrieved from http://www.hopkinsmedicine.org/news/media/releases/diagnostic_errors_more_common_costly_and_harmful_than_treatment_mistakes
Attachment:- Project Problem.rar