The evaluation of cancer patients’ recovery is still under the big subjectivity of physicians. Many different systems have been successfully implemented for physical activity evaluation, nonetheless there is still a big leap into Performance Status evaluation with ECOG and Karnofsky’s Performance Status scores. An automatic system for data recovering based on Android smartphone and wearables has been developed. A gamification implementation has been designed for increasing patients’ motivation in their recovery. Furthermore, novel and without-precedent algorithms for Performance Status (PS) and Physical Activity (PA) assessment have been developed to help oncologists in their diagnoses.
First Approach to Automatic Performance Status Evaluation and Physical Activity Recognition in Cancer Patients
1. First Approach to Automatic
Performance Status Evaluation and
Physical Activity Recognition in
Cancer Patients
Salvador Moreno · Miguel Damas · Hector Pomares
Jose A.Moral-Munoz · Oresti Banos
Email: smoreno94@correo.ugr.es
SUT4Coaching 2016 – Granada, Spain
2. Index of Contents
1. Motivation
2. State of the Art
3. System Design
4. System Implementation
5. Conclusions
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
2
3. 1. Motivation
● 14 million new cases and 8 million cancer-related
deaths in 2012
● Is projected to increase by 70%
in two decades.
● The estimated total annual economic
cost of cancer was approximately
US$ 1.16 trillion in 2010
● Between US$ 100 billion and US$
200 billion could have been saved
in 2010
3
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
[1] World Health Organization, World Cancer Report 2014. Lyon, FRA:
International Agency for Research on Cancer, 2014
4. 1. Motivation
● Cancer treatment relies in several aspects under the
subjectivity of the oncologist.
– Interviews
– Family's Feedback
● Patients are sometimes
hard to treat.
● Hard to remember by
heart and measure the activity realized.
4
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
5. 1. Motivation
Solution Proposed:
Automatic Tracking of the
Patient's Physical Activity
5
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
6. 2. State of the Art
6
Patient's
Condition
Performance
Status
Physical
Condition
ECOG
KPS
IPAQ
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
7. 2. State of the Art
KPS (Karnofsky's Performance Status) – 1949
Normal, no complaints. No evidence of disease. 100
Able to carry on normal activity. Minor signs of disease. 90
Normal activity with effort. Some signs or symptoms of disease. 80
Cares for self; unable to carry on normal activity or do active work. 70
Ocassional assistance. Able to care for most of personal needs. 60
Considerable assistance. Frequent medical care. 50
Disabled. Requires special care and assistance. 40
Severely disabled. Hospital admission indicated. Death not imminent. 30
Very sick. Hospital admission necesary. 20
Moribund. 10
Exitus. 0
7
[3] Karnofsky DA Burchenal JH. The clinical evaluation of chemotherapeutic agents in cancer.
Evaluation of chemotherapeutic agents., pages 191–205, 1949. New York Columbia.
University Press.
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
8. 2. State of the Art
8
ECOG (Eastern Cooperative Oncology Gropp) – 1960
Asymptomatic. Normal, no complains. 0
Symptomatic but completely ambulatory. Able to carry work of light or sedentary
nature like housework or office work.
1
Symptomatic and able of all self care but unable to carry out work activities.
Less than 50% of waking hours in bed or chair.
2
Symptomatic and capable of only limited self-care. Confined to bed or chair
50% or more of waking hours.
3
Bedbound. Completely disabled. Cannot carry any self-care. 4
Exitus 5
[15] C.G Zubrod et al. Appraisal of methods for the study
of chemotherapy of cancer in man: Comparative therapeutic trial of nitrogen mustard and
triethylene thiophosphoramide. Journal of Chronic Diseases, 11(1):7–33, 1960.
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
9. 2. State of the Art
9
IPAQ: International Physical Activity Questionnaire
– IPAQ 3: High Physical Activity
– IPAQ 2: Moderate Physical Activity
– IPAQ 1: Low Physical Activity
Physical activity measured in MET·min
[22] IPAQ scoring protocol - International Physical Activity Questionnaire.
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
10. 2. State of the Art
● IPAQ
●
Category 1: Low
– This is the lowest level of physical activity. Those individuals who not meet
criteria for categories 2 or 3 are considered low/inactive.
● Category 2: Moderate
– 3 or more days of vigorous activity of at least 20 minutes per day OR
– 5 or more days of moderate-intensity activity or walking of at least 30 minutes
per day OR
– 5 or more days of any combination of walking, moderate-intensity or vigorous
intensity activities achieving a minimum of at least 600 MET-min/week.
● Category 3: High
– Vigorous-intensity activity on at least 3 days and accumulating at least 1500
MET-minutes/week OR
– 7 or more days of any combination of walking, moderate-intensity or vigorous
intensity activities achieving a minimum of at least 3000 MET-minutes/week
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
[22] IPAQ scoring protocol - International Physical Activity Questionnaire.
SUT4Coaching – Granada, Spain.
10
11. 2. State of the Art
11
Physical
Activity
Measurement
Heart Rate
Intensity
(Karvonen)
Daily
Step Count
(Tudor-Locke)
Class Steps/day
Sedentary <5000
Low Act. 5000 – 7499
Medium Act. 7500 – 9999
High Act. 10000 – 12499
Very High Act. >12500
[35] Dr Juha Karvonen and Timo Vuorimaa. Heart Rate
and Exercise Intensity During Sports Activities. Sports
Medicine, 5(5):303–311, November 2012.
[39] Catrine Tudor-Locke and David R. Bassett Jr. How
many steps/day are enough? Sports medicine, 34(1):1–
8, 2004.
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
12. 3. System Design
Objectives:
● Automatic tracking of the
activity performed.
● Develop algorithm for IPAQ
estimation
● Develop algorithm for ECOG
and KPS estimation.
● Focus on tendencies rather than the absolute
measures.
12
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
13. 3. System Design
13
Google FIT:
– Sensors API
– Recording API
– History API
– Sessions API
– BLE (Bluetooth
Low Energy) API
– Config API
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
Data Recorded
– Activity Recognition
● Still, walking,
running...
● Sleep tracking
– Calories Expended
(cal)
– Heart Rate (BPM)
– Step Count (count)
14. 3. System Design
14
Wearable selected:
Sony Smartband 2 SWR12
Cost <100€
Measures:
– Heart Rate (HR)
– Step Count
– Activity Recognition
– Calories Expended
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
15. 3. System Design
15
● Gamification
[31] Kevin Werbach and Dan Hunter. For the Win: How Game Thinking Can
Revolutionize Your Business. Wharton Digital Press, October 2012.
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
16. 3. System Design
16
Gamification Design:
– Interactive interface for
the patient.
– Highlight achievements
done in the recovery.
– Engagement and
progression loops.
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
17. 17
Trabajo de Fin de Grado
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
18. 3. System Design
18
Planteamiento del sistema.
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
20. 4. System Implementation
20
IPAQ Estimation*
1. SQLite query
2. Sampling Frequency adaptation between data.
3. Daily detection and data bucketting.
4. IPAQ rules application.
5. Weekly approach of the estimation.
– *Intensity of the Activity Detected by using
Heart Rate and Calories Count.
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
21. 4. System Implementation
21
IPAQ Algorithm Validation
● Comparison between
IPAQ MET estimation
and Google Fit's
Calorie Count converted
to MET
● Pearson Correlation
– 0.8505 with all data
– 0.9665 without
outliers
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
22. 4. System Implementation
ECOG and KPS estimation*
1. SQLite Query
2. Sampling Frequency adaptation between data.
3. Daily detection and data bucketting.
4. ECOG and KPS Table Rules application.
5. Weekly approach of the estimation.
– *Activity detected, Intensity of the Act. Detected,
Steps Count, Proportion of Active Waking Hours.
22
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
23. Daily proportion of active Waking Hours
23
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
4. System Implementation
26. 5. Conclusions
● Extensive review of the State of the Art.
● Needs assessment performed with the help of
Oncologists and psychologists from the Hostpital
Virgen de las Nieves in Granada.
● Android, SQLite and MATLAB have been used for
the system implementation.
● It is the first approach to automatic performance
status evaluation in cancer patients.
26
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
27. First Approach to Automatic
Performance Status Evaluation and
Physical Activity Recognition in
Cancer Patients
Salvador Moreno · Miguel Damas · Hector Pomares
Jose A.Moral-Munoz · Oresti Banos
Email: smoreno94@correo.ugr.es
SUT4Coaching 2016 – Granada, Spain
28. 3. System Design
● Wearable market grows 171.6% between 2014 y 2015
and smartphones keep their sales (Source: IDC)
28
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
29. 4. System Implementation
Physical Activity time detected
29
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
30. 4. System Implementation
30
● Execution time of a nineteen-days-query.
Execution Time:
ET = (3,4 ± 0,4) s
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
SUT4Coaching – Granada, Spain.
31. 2. State of the Art
31
● Gamification:
SUT4Coaching – Granada, Spain.
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016
32. 2. State of the Art
32
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
● IoT & mHealth
SUT4Coaching – Granada, Spain.
Automatic Performance Status Evaluation.
S. Moreno Gutiérrez
15th
December 2016