ArticlePDF Available

Estimating Oxygen Uptake during Non-Steady-State Activities and Transitions Using Wearable Sensors

Authors:
  • HRV4Training

Abstract and Figures

In this paper, we present a method to estimate oxygen uptake (VO2) during daily life activities and transitions between them. First, we automatically locate transitions between activities and periods of non-steady-state VO2. Subsequently, we propose and compare activity-specific linear functions to model steady-state activities and transition-specific non-linear functions to model non-steady-state activities and transitions. We evaluate our approach in study data from 22 participants that wore a combined accelerometer and heart rate (HR) sensor while performing a wide range of activities (clustered into lying, sedentary, dynamic/household, walking, biking and running), including many transitions between intensities, thus resulting in non-steady-state VO2. Indirect calorimetry was used in parallel to obtain VO2 reference. VO2 estimation error during transitions between sedentary, household and walking activities could be reduced by 16% on average using the proposed approach, compared to state of the art methods.
Content may be subject to copyright.
1
Estimating Oxygen Uptake during Non-Steady-State
Activities and Transitions Using Wearable Sensors
Marco Altini
1
, Julien Penders
2
and Oliver Amft
3
Abstract—In this paper, we present a method to estimate
oxygen uptake (VO
2
) during daily life activities and transitions
between them. First, we automatically locate transitions between
activities and periods of non-steady-state VO
2
. Subsequently, we
propose and compare activity-specific linear functions to model
steady-state activities and transition-specific non-linear functions
to model non-steady-state activities and transitions. We evaluate
our approach in study data from 22 participants that wore a
combined accelerometer and heart rate (HR) sensor while per-
forming a wide range of activities (clustered into lying, sedentary,
dynamic/household, walking, biking and running), including many
transitions between intensities, thus resulting in non-steady-
state VO
2
. Indirect calorimetry was used in parallel to obtain
VO
2
reference. VO
2
estimation error during transitions between
sedentary, household and walking activities could be reduced by
16% on average using the proposed approach, compared to state
of the art methods.
Index Terms—Accelerometers, Energy Expenditure, Heart
Rate, Non-Steady-State, VO
2
I. INTRODUCTION
Ubiquitous sensing technologies that objectively and non-
invasively monitor human behavior, started to provide insights
into the relation between physical activity (PA) and health.
Among the parameters used to objectively quantify PA, energy
expenditure (EE) is the most commonly used single metric
[1], [2], [3]. The measurement of steady-state oxygen uptake
(VO
2
) is considered to be the gold standard for estimating
EE during light to moderate steady-state exercise [4], [5],
where aerobic pathways are predominant. In this context, VO
2
measurements are proportional to metabolic heat production
[4]. Due to the practical limitations of measuring VO
2
in free
living, different methods to estimate VO
2
using miniaturized
wearable sensors have been developed in the past.
VO
2
estimation methods are typically based on activity-
specific linear regressions developed using steady-state data
and therefore can describe the variations within each modeled
activity during steady-state. However, transitions to activities
with other VO
2
levels cannot be accurately estimated, since
VO
2
dynamics differ during steady-state and non-steady-
state periods. Proper modeling of non-steady-state transitions
including transitions of activity types (e.g. sitting to walking)
and activity intensities (e.g. walking at different speed) are
necessary in order to provide accurate VO
2
estimation in free
1
M. Altini is with Bloom Technologies, Agoralaan Building Abis 2.13,
3590 Diepenbeek, Belgium, and Eindhoven University of Technology, Eind-
hoven, The Netherlands (e-mail: marco.altini@imec-nl.nl)
2
J. Penders is with Holst Centre/imec The Netherlands, Eindhoven, Nether-
lands, (e-mail: julien.penders@imec-nl.nl)
3
O. Amft is with University of Passau, Germany, and Eindhoven University
of Technology, The Netherlands (e-mail: amft@ieee.org)
living conditions. Accurate identification of non-steady-state
periods as well as quantifications of VO
2
during transitions
could improve EE estimation because during non-steady-state
periods total EE is composed of aerobic and anaerobic compo-
nents. [4]. Earlier work on VO
2
transitions analysis focused
mainly on Post-Exercise Oxygen Consumption (EPOC) [6],
or VO
2
estimation for single activities [7], [8]. However
non-steady-state VO
2
is very frequent during varying low
intensities activities of daily living (ADLs). Studies showed
that most activities performed in free living last shorter than
the time needed to reach steady state. For example, 60% of
all walking bouts last shorter than 30 seconds [9]. Identifying
non-steady-state VO
2
can provide more insights on the aerobic
and anaerobic dynamics during the onset of exercise as well as
during ADLs characterized by short duration. While solutions
have been proposed to model non-steady-state VO
2
for spe-
cific activities [7], [8] or transitions [6], a unified approach able
to continuously estimate VO
2
is missing. Thus, we propose
a novel VO
2
estimation method, which combines activity-
specific VO
2
estimation using linear regression, with non-
steady-state detection and transition-specific VO
2
estimation
using non-linear equations.
This paper provides the following contributions:
1) We introduce a method to automatically locate periods
of non-steady-state VO
2
by analyzing the coefficient
of variation (CV) of the predicted VO
2
. Using the
CV allows for detection of transition both between and
within activities. Then, we compare linear, exponential
and logistic transfer functions to model non-steady-state
VO
2
during individual transition types.
2) We evaluate the proposed approach on a dataset ac-
quired from 22 participants performing a wide set of
physical activities, including many transitions between
activities and changes of intensity within activities. We
show that the transition-specific modeling could reduce
VO
2
estimation error by 16% during activity transitions,
compared to state of the art methods.
II. RELATED WORK
Accelerometer and HR monitors are the most commonly
used single sensor devices in epidemiological studies. Ac-
celerometers use features representative of whole body motion,
as independent variables in the linear regression model devel-
oped to predict EE. However there are limitations due to the
inability of a single linear model to fit all activities, since the
slope and intercept of the regression model change based on
the activity performed while data is collected [3]. On the other
2
hand, the high correlation between HR and EE within one
individual changes substantially between individuals [1]. The
latest EE estimation algorithms extended approaches based
on simple linear regression models performing activity recog-
nition over a predefined set of activities, and then applying
different methods - typically regression models - to predict
EE, based on the activity [2], [3]. The regression models use
accelerometer features and anthropometric characteristics as
independent variables. Some authors included HR features as
well in the activity-specific linear models, showing consistent
improvements in EE estimation accuracy compared to algo-
rithms using accelerometer only features [1]. However, none
of these models explicitly models non-steady-state VO
2
.
A. Non-Steady-State VO
2
Estimation
Non-steady-state EE is often defined as periods of time
where VO
2
and carbon dioxide production vary by more
than 5-10% [10]. Previous research on non-steady-state VO
2
focused mainly on metabolic responses to exercise (EPOC,
[6]). However, VO
2
estimation is intrinsically a temporal
problem and non-steady-state VO
2
is very common during
low intensity ADLs as well. While VO
2
increases rapidly,
normally reaching steady-state within 1-4 minutes, most ac-
tivities performed in free living last less than the time required
to reach steady state [9]. Most EE models are developed
deriving EE from VO
2
, and discarding the first 1 or 2
minutes of data [1], [3], to isolate steady-state. Thus, the
predictions of these models will be negatively affected by the
real life nature of ADL activities. Other models incorporate
non-steady-state VO
2
but without providing details on the
models accuracy during transitions [11], thus limiting our
understanding of the models performance in non-steady-state
conditions. Using physiological data to predict VO
2
involves
slower dynamics present in both physiological changes (e.g.
HR slowing increasing) and aerobic pathways (VO
2
reaching
steady-state). However, these dynamics are different, typically
with HR being much slower than VO
2
[12], [13].
Few attempts to model non-steady-state VO
2
are found in
literature [14], [15], [16], [7], [8]. In [14], [15], the proposed
system relies on HR data only, suffering from all limitations
of non-activity-specific models, while [7], [8] analyzed walk-
ing data only. In [16], the authors used mono-exponential
functions to better capture the relation between movement
and EE during transitions between activities. However, the
EE prediction ignores the fact that during non-steady-state
EE cannot be derived from VO
2
alone, since total EE is
composed of both aerobic (estimated via VO
2
) and anaerobic
components. Finally, one single mono-exponential equation
might not be sufficient to model energy deficit and energy debt
situations between different activities [5]. While solutions have
been proposed to model non-steady-state VO
2
for specific
activities [7], [8] or transitions [6], a unified approach able
to continuously estimate VO
2
is missing.
III. ANA LY S IS A N D ESTIMATION APPROACH
This section describes the problem of non-steady-state VO
2
estimation and our approach to such problem.
5.0
7.5
10.0
12.5
15.0
17.5
0 20 40 60
VO2 (ml/kg/min)
State
steadystate
nonsteadystate
a)
5
10
15
0 20 40 60
time windows (4 seconds each)
VO2 (ml/kg/min)
Activity Type
sedentary
dynamic
walking
b)
Fig. 1. Example of non-steady-state VO
2
. a) Non-steady state reference
VO
2
. b) Predicted VO
2
as estimated by steady-state activity-specific models.
In this work, non-steady state modelling is used combined with classic steady-
state models to better estimate actual VO2.
Our approach combines detection of non-steady-state VO
2
,
activity-specific linear models and non-steady state transition-
specific non-linear functions. Fig. 1 shows typical non-steady
state dynamics in activities of daily living, such as transitions
between sedentary activities (e.g. sitting or standing), walking
and then sedentary again. Fig. 1.b shows color-coded activities
as detected by an activity recognition system, while Fig. 1.a
shows steady-state and non-steady-state-data. Non-steady-state
VO
2
is present when the participant starts walking and when
the participant stops walking. Fig. 1.a show measured VO
2
while Fig. 1.b shows VO
2
as predicted by state of the art
activity specific models. Prediction models (Fig. 1.b) jump
between one value to the other as soon as a new activity starts,
while transitions shown in Fig. 1.a are much slower. Thus, a
different modeling technique is required during non-steady-
state.
A. Estimation Architecture
VO
2
predictions are generated by a sequence of steady-state
and non-steady-state models (see, e.g. Fig. 2):
...W
1
,V
1
,W
2
,V
2
,W
1
,V
3
...
where W
i
are activity-specific models used when steady-
state is detected and V
i
are transition-specific non-steady-state
models used when non-steady-state is detected. Each state W
i
or V
i
comprises t predictions based on the time spent in a
specific activity or transition duration:
W
i
= {VO2
Wi
SS
1
,...,VO2
Wi
SS
t
}
V
i
= {VO2
Vi
NS
1
,...,VO2
Vi
NS
t
}
In the following sections, we describe W
i
, V
i
and model
selection.
1) Activity-Specific Steady-State Models - W
i
: W
i
are com-
posed of two parts: activity recognition and activity-specific
multiple linear regression equations. Assuming n clusters of
activities:
C = {c
1
,...,c
n
}, 8c
i
2 C, 9 W
i
3
Activity
Recognition
State
1
model
State
n
model
VO2
ss
Activity-specific VO
2
models
Transition
1
model
Transition
n
model
Transition-specific VO
2
models
Model
Selection
VO2
ns
VO2
Physiological
Features
Anthropometric
Features
activity
class
c
i
W
1
W
n
V
1
V
n
VO2
ss
Accelerometer
Features
Fig. 2. Block diagram of out approach using activity-specific linear models
and transition-specific non-linear functions to estimate VO2. Features are used
for activity recognition and activity-specific VO
2
models. Activity recognition
is used to select the proper activity-specific or transition-specific model.
where the parameters of the activity-specific model W
i
are
derived from VO
2
values for a specific cluster of activities c
i
:
VO2
Wi
SS
= X
act
i
act
i
+ (1)
is the vector of regression coefficients, and X
act
i
is the
vector of input features. Features are grouped into accelerom-
eter, physiological and anthropometric features (see Fig. 2).
2) Transition-Specific Non-Steady-State models - V
i
: We
compare linear, exponential and logistic functions as V
i
.
Assuming r types of transitions:
Tr = {tr
1
,...,tr
r
}, 8tr
i
2 Tr, 9 V
i
Linear functions: for a transition type V
i
from steady-state
W
j
to steady-state W
k
, a linear model is derived as follows:
VO2
Vi
NS
= + t + VO2
Wj
SS
+ (2)
where VO2
Vi
NS
are target VO
2
values during non-steady-state
for the transition V
i
, defined by transition-specific linear func-
tions. The predictor t is the time elapsed since the transition
started. VO2
Wj
SS
is the VO
2
of the steady-state before the
transition (W
j
). Parameters and are the slope and intercept
of the linear model.
Logistic functions: for a transition type V
i
, a non-linear
model is derived as follows:
VO2
Vi
NS
=
1
1+exp
(
Vi
2
+
Vi
3
t)
+ VO2
Wj
SS
+ (3)
where VO2
Vi
NS
are target VO
2
values during non-steady-state
for the transition V
i
, defined by transition-specific logistic
functions. The predictor t is the time elapsed since the tran-
sition started. VO2
Wj
SS
is the VO
2
of the steady-state before
the transition (W
j
). Transitions between activities of different
type and intensity will have different dynamics [5], [17], thus
require transition-specific functions. The parameter
1
is the
asymptote of the logistic function, and can be automatically
derived as the difference between the steady-state VO
2
before
(W
j
) and after (W
k
) the transition:
1
= VO2
Wk
SS
VO2
Wj
SS
. (4)
Parameters
Vi
2
and
Vi
3
control the shape of the logistic curve
(e.g the transition speed), and were determined by fitting
Recognized*ac,vity*
sedentary* walking* sedentary*
0
25
50
75
0 20 40 60
CV (%)
State
steadystate
nonsteadystate
Time*windows*(4*s)*
Fig. 3. Example waveform of the CV of the estimated VO
2
for a sequence
of sedentary, walking and sedentary activities. State is non-steady when
CV > 15%. The CV captures transitions between activities as well as within
activities, thus detecting non-steady-state VO
2
.
for each transition type. Thus,
Vi
2
and
Vi
3
are transition-
specific. The transition type (i.e. V
i
) depends on the activities
recognized by the activity recognition system during steady-
states W
j
and W
k
.
Exponential functions: a transition model for a transition
type V
i
is derived as follows:
VO2
Vi
NS
=
1
exp
+t
+ VO2
Wj
SS
+ (5)
where VO2
Vi
NS
are target VO
2
values for the transition V
i
,
defined by transition-specific exponential functions. The pre-
dictor t is the time elapsed since the transition started. VO2
Wj
SS
is the VO
2
of the steady-state before the transition (W
j
). The
parameter
1
is the same as for the logistic function, indicating
the asymptote of the exponential, and can be automatically
derived as shown in Eq. 4.
3) Model Selection: Non-steady-state is detected using the
CV of the predicted VO
2
over the last minute of data,
according to a preselected threshold:
CV
VO2
SS
=
VO2
SS
µ
VO2
SS
(6)
where
VO2
SS
and µ
VO2
SS
are the standard deviation and
mean of a 60 s sliding window of predicted VO
2
values.
Non-steady-state is detected if CV
VO2
SS
>CV
T hres
, where
CV
T hres
was set to 15%. Fig. 3 shows the CV for the predicted
VO
2
data. CV can be used to locate non-steady-states, such
as when activity intensity changes, thus even when there is no
transition between two activities.
IV. EVA L UAT I O N STUDY
A. Participants and Data Acquisition
Participants were 22 (17 male, 5 female), mean age
30.4 ± 5.7 years, mean weight 71.2 ± 12. 4 kg, mean height
1.76 ± 0.09 m, mean BMI 23.0 ± 2.8 kg/m
2
. Imec’s IRB
approved the study. Each participant signed an informed con-
sent form. The sensor platform used was the ECG Necklace, a
wearable sensor acquiring one lead ECG data at 256 Hz, and
three-axial accelerometer data at 32 Hz. Activity type was
annotated manually by experimenter. Breath-by-breath data
were collected using the Cosmed K4b
2
indirect calorimeter.
We interpolated calorimeter data at 0.25 Hz to align it with
the ECG Necklace data, and applied a moving average with
window size 4 elements to reduce high frequency noise.
4
B. Experiment Design
Participants reported at the lab after refraining from drink-
ing, eating and smoking in the two hours before the experi-
ment. Activities were grouped into six clusters to be used for
activity classification. The six clusters were lying (lying down),
sedentary (sitting, standing, desk work, reading, writing, PC
work), dynamic (stacking groceries, washing dishes, cleaning,
sweeping, vacuuming), walking (treadmill flat at 3,4,5,6 km/h,
inclined 3-5%, 3-5 km/h), biking (cycle ergometer, low medium
and high resistance level), running (treadmill 7,8,9, and 10
km/h). Activities were carried out for a period of at least 4
minutes, with the exception of running (1 to 4 minutes). All
transitions were manually annotated.
C. Statistics and Performance Measure
Models were derived using leave-one-participant-out cross
validation. The same training set, consisting of data from
all participants but one, was used to perform feature selec-
tion, activity recognition, activity-specific VO
2
estimation and
transition-specific VO
2
estimation models. The data from the
remaining participant was used for validation. This procedure
was repeated n (n = number of participants) times, and re-
sults were averaged. All parameters used in transition-specific
functions were determined in the same way, no data used for
model building was used for model evaluation. Performance
of the activity recognition models was evaluated using the
class-normalized accuracy. Results for VO
2
estimates are
reported in terms of Root-mean-square error (RMSE), where
the outcome variable was VO
2
in ml/kg/min. Paired t-tests
were used to compare RMSE between models.
V. I MPLEMENTATION
A. Features Extraction and Selection
Features extracted from the sensors’ raw data were used
to derive activity recognition and VO
2
estimation models.
Activity recognition was performed to classify the six activity
clusters introduced in Section IV-B. Accelerometer data were
segmented in 4 s windows, band-pass filtered between 0.1
and 10 Hz, to isolate the dynamic component caused by
body motion, and low-pass filtered at 1 Hz, to isolate the
static component, due to gravity. Feature selection for activity
type recognition was based on mutual information. The final
feature set included: mean of the absolute signal, inter-quartile
range, median, variance, main frequency peak, low and high
frequency band signal power. Feature selection for VO
2
estimation was based on how much variation in VO
2
each
feature could explain within one cluster. The process was
automated using linear forward selection.
B. Activity Recognition
We selected a time window of 4 s, which is short enough
to detect short breaks in sedentary time, and long enough to
capture the repetitive patterns of some activities (e.g. walking
or running). Given the positive results in past research on
activity recognition, we selected Support Vector Machines
(SVMs) as classifiers. For the SVMs, we used a polynomial
kernel with degree 5 ( = 10, C = 1).
C. Activity-Specific Steady-State Models - W
i
Within one activity cluster, VO
2
can be estimated using
features representative of VO
2
changes within the activity
cluster [3], [1]. We used the mean of the absolute signal
to model changes in intensity within an activity, together
with HR. Anthropometrics features (body weight and resting
metabolic rate (RMR), estimated with the Harris-Benedict
formula) were added depending on the activity cluster.
D. Non-Steady-State Detection
The CV over one minute windows was computed to locate
non-steady-state segments of data. When the CV was higher
than CV
T hres
, a new non-steady-state transition was detected.
CV
Thres
was derived from previous literature on detection
of non-steady-state VO
2
, and empirically cross-validated on
our dataset, since no definition of non-steady-state is widely
accepted [10]. Once a non-steady-state transition V
i
was
detected between two steady-states W
j
and W
k
, the system
compared the steady-state VO
2
levels and enables the non-
steady-state models only if the difference in VO
2
was greater
than a threshold. This filter avoids activating the non-steady-
state models for transitions that are too small or short.
E. Transition-Specific Non-Steady-State models - V
i
Transition-specific non-steady-state models V
i
were devel-
oped for the most common transitions. More specifically,
logistic functions were derived by fitting the parameters
2
and
3
while linear and exponential functions were derived by
fitting the parameters and of the respective models, for
the following transitions: sedentary to walking (SW), walking
to sedentary (WS), sedentary to dynamic/household (SD),
dynamic/household to sedentary (DS), walking fast to walking
slow (WWDOWN), walking slow to walking fast (WWUP).
Data used for model development was not used for validation.
VI. RESULTS
We report results for activity recognition, activity-specific
steady-state models, non-steady-state detection and transition-
specific non-steady-state models, together with summary
statistics on the specific transitions considered in this work.
An example of the proposed method applied to a transition
between sedentary behavior and walking is shown in Fig. 4.
5
10
15
0 20 40 60 80
Samples
VO2 ml/kg/min
type
ActSpec ACCHR
ActSpec ACCHR + Logistic
referenceVO2
ac#vity(
sedentary(
walking(
sedentary(
Fig. 4. Example of the results obtained when combining activity-specific
steady state models and transition-specific non-stead-state models (Act-Spec
ACC-HR + Logistic), as proposed by our method, for a transition between
sedentary to walking and walking to sedentary. Act-Spec ACC-HR show the
inability of steady-state models to predict VO
2
accurately during transitions.
5
Fig. 5. RMSE for VO
2
estimation using different sensing modalities and
methods for steady-state data. Activity-specific (Act-Spec) models outperform
all others, combined accelerometer and heart rate (ACC-HR) data improves
performance compared to accelerometer (ACC) or heart rate (HR) only.
A. Activity Recognition
Activity recognition accuracy was 94% on average across
all participants using the validation procedure detailed in Sec.
IV-C. More specifically, accuracy was 100% for lying down,
96% for sedentary, 81% for dynamic/household, 99% for
walking, 91% for biking and 98% for running.
B. Activity-Specific Steady-State Models
RMSE for activity-specific steady-state VO
2
estimation
models using combined accelerometer and HR data was 3.5
ml/kg/min. More specifically, RMSE was 1.16 ml/kg/min for
lying down, 1.75 ml/kg/min for sedentary, 3.93 ml/kg/min
for walking, 4.55 ml/kg/min for dynamic/household, 4.24
ml/kg/min for biking and 5.81 ml/kg/min for running. Sim-
ilarly to what was reported in literature for EE estimation
models, activity-specific VO
2
estimation models combining
accelerometer and HR data outperformed single regression
models relying on accelerometer only (46% RMSE reduc-
tion, p =4e
11
<), HR only (23% RMSE reduction,
p =0.0002 <), combined accelerometer and HR data
(17% RMSE reduction, p =0.001 <) and activity-specific
estimation models relying on accelerometer data only (10%
RMSE reduction, p =0.002 <). Fig. 5 provides an
overview.
C. Non-Steady-State Detection
221 transitions were analyzed in total (45 SW, 28 WS, 42
SD, 53 DS, 39 WWUP, 14 WWDOWN). 85% of all transitions
were correctly identified by the transition detection system.
Transition detection was more accurate for transitions between
sedentary to walking activities (98%) and for transitions be-
tween walking to sedentary activities (100%). Transition detec-
tion accuracy dropped to 78% and 84% for transitions between
sedentary to dynamic/household and dynamic/household to
sedentary respectively. Transitions in VO
2
within activities
(e.g. following changes in walking speed or inclination), that
were smaller than 1 ml/kg/min were not considered. All other
within activities transitions were correctly identified (100% ).
Fig. 6. RMSE for VO
2
estimation using different methods for non-steady-
state data. While standard activity-specific models that combine acc and
HR (Act-Spec ACC+HR) perform better than other models during steady-
state (see Fig .5), they offer poor performance during transitions. Linear,
exponential and logistic functions progressively reduce RMSE, with logistic
functions providing the lowest errors. Logistic Generic refers to logistic
functions were parameters did not vary depending on the transition type.
D. Transition Types
There was a significant relation between transition time
and VO
2
change between the starting and target activities
during a transition (correlation coefficient r =0.68, p =
0.0015 <). VO
2
difference between the starting and target
activities during a transition was 6.40 ml/kg/min for SW, 9.25
ml/kg/min for WS, 7.47 ml/kg/min for SD, 7.46 ml/kg/min
for DS , 4.38 ml/kg/min for WWUP and 5.80 ml/kg/min
for WWDOWN. Average transition duration was 97 s for
SW , 108 s for WS, 88 s for SD, 98 s for DS, 87 s for
WWUP and 125 s for WWDOWN. VO
2
estimation error
for steady-state models during transitions involving sedentary
and household activities was 4.62 ml/kg/min (SD) and 4.50
ml/kg/min (DS ), even though error during steady-state was
only 1.75 ml/kg/min for sedentary and 4.55 ml/kg/min for
household. VO
2
estimation error for steady-state models
during transitions involving sedentary and walking activities
was 5.63 ml/kg/min (SW) and 4.23 ml/kg/min (WS), even
though error during steady-state was only 1.75 ml/kg/min for
sedentary and 3.93 ml/kg/min for walking.
E. Transition-Specific Non-Steady-State models
Fig. 6 shows VO
2
RMSE during transitions for the best
performing steady-state models (i.e. activity-specific models
combining accelerometer and HR data) and non-steady-state
models. VO
2
RMSE during transitions for steady-state models
was 4.38 ml/kg/min. Linear models reduced RMSE to 4.27
ml/kg/min (3% RMSE reduction, p =0.5 >), expo-
nential functions to 4.07 ml/kg/min (7% RMSE reduction,
p =0.2 >), and logistic functions to 3.68 ml/kg/min (16%
RMSE reduction, p =0.0007 <). Thus, logistic functions
were the best performing non-steady-state models for activities
transitions. For comparison, we evaluated logistic functions
were parameters were not fitted to specific transitions, but were
held constant for all transitions, denoted as Logistic Generic in
Fig. 6 and Table I. Logistic generic results were VO
2
RMSE
of 3.77 ml/kg/min (14% RMSE reduction, p =0.002 <).
6
TABLE I
VO
2
RMSE FOR TRANSITIONS MODELS CONSIDERING PERFECT TRANSITION DETECTION (ML/KG/MIN)
Transition Activity-Specific Linear Exponential Logistic Logistic Generic
Sedentary to walking (SW) 5.42 ± 1.46 5.79 ± 1.86 4.67 ± 3.18 4.27 ± 2.20 4.94 ± 2.40
Walking to sedentary (WS) 3.84 ± 1.12 4.10 ± 1.62 3.26 ± 2.60 2.43 ± 1.59 2.48 ± 1.33
Sedentary to dynamic (SD) 4.28 ± 1.25 3.81 ± 1.54 4.07 ± 1.99 3.30 ± 1.29 3.29 ± 1.36
Dynamic to sedentary (DS) 4.20 ± 1.48 3.17 ± 1.43 3.32 ± 1.28 3.57 ± 1.31 3.42 ± 1.23
Walking slow to fast (WWUP) 4.10 ± 2.06 4.97 ± 2.43 3.15 ± 1.79 4.05 ± 2.10 4.06 ± 2.10
Walking fast to slow (WWDOWN) 4.03 ± 2.17 3.69 ± 2.32 4.87 ± 3.01 4.4 ± 2.28 3.95 ± 2.14
Mean 4.38 ± 0.80 4.27 ± 1.05 4.07 ± 1.28 3.68 ± 1.10 3.77 ± 1.03
VII. DISCUSSION
In this paper we introduced a method to estimate VO
2
dur-
ing ADLs. Our method combines non-steady-state detection
and transition-specific VO
2
estimation non-linear equations.
It is misleading to assume that EE can be derived from
VO
2
alone. Proper modeling of non-steady-state transitions
is important since both anaerobic and aerobic EE result in
heat production, which is the true EE. However, only aerobic
EE results in VO
2
consumption. To tackle non-steady-state
VO
2
estimation, we proposed a method to locate periods of
non-steady-state VO
2
using the CV over the predicted VO
2
level. Our results showed that the CV can be used for detecting
transitions between and within activities. Then, we showed that
modeling non-steady-state transitions using non-linear logistic
functions for individual transition types could reduce VO
2
estimation error by 16%.
A. Activity Recognition and Transition Detection
In previous research both unsupervised and supervised
methods for activity recognition were proposed in the context
of VO
2
estimation, with supervised methods providing better
results [18]. For this reason, we opted for supervised methods
typically able to provide good accuracy on big clusters of
activities as confirmed by our results, where 94% average
accuracy was obtained in leave-one-participant-out validation
design. Transition detection was highly accurate for all tran-
sitions including the walking activity (98 100%). However,
transition detection was less accurate for transitions involving
dynamic/household activities (78 84%), since these activities
are both more difficult to recognize and highly variable in
VO
2
. In literature, several approaches have been proposed
to detect transitions between activities [19], [20]. Detecting
transitions based on the CV of the predicted VO
2
provides
the advantage to detect not only transitions between activities,
but also changes in intensity within the same activity (e.g.
walking at a lower speed). Various definitions of steady-state
have been reported [10]. Some researchers defined steady state
as 5 consecutive 1-minutes intervals where O
2
consumption
changes by less than 10%. However this definition resulted
from assessments of basal metabolic rate and is not directly
applicable to free living conditions, where we are interested in
detecting much shorter transitions. Our analysis shows that one
minute windows and a slightly less strict CV (i.e. 15%), are a
good trade-off for non-steady-state and transitions detection.
B. Transition-Specific Non-Steady-State Models
Our analysis focused on sedentary, dynamic/household and
walking activities, since our objective is to improve VO
2
estimation in common transitions during ADLs, where current
methods lack proper modeling.
Thus, we have not considered transitions during intense
activities (e.g. biking or running), which are less frequent in
daily life and well studied in literature. The ADLs considered
are characterized by similar levels of VO
2
for both the starting
and target activity of a transition. Thus, the VO
2
change
during a transition and the time required to reach steady
state are similar. While parameters differ between transitions,
and transition-specific logistic functions provided the best
results, generic logistic functions which do not vary parameters
based on the transition were sufficient to significantly reduce
RMSE compared to steady-state models (14% RMSE reduc-
tion, p =0.002 <). Logistic functions were proposed in
literature to model HR recovery after intense exercise [21], due
to the ability to model different dose-response relationships
between two relatively stable levels. While we applied this
technique to model transitory oxygen uptake during ADLs,
the methodology could be extended to other physiological
parameters undergoing similar changes. For comparison, we
also fitted linear and exponential functions, as previously
reported in literature [16]. However we found suboptimal
results compared to logistic functions.
Another advantage of the proposed technique is that at
runtime VO
2
estimations during transitions are only dependent
on the previously fitted parameters and on the parameter
1
,
representing the difference in VO
2
between the starting and
the target activity. Thus, these models could be applied to
systems based on accelerometers only, and do not depend on
other predictors or sensing modalities.
Performance of non-steady-state VO
2
models should be
assessed outside of the lab [22]. However, there is currently no
unobtrusive reference system that is able to measure breath-by-
breath VO
2
in unconstrained settings. Only by using indirect
calorimetry and supervised settings we can record data, which
allows us to analyze how non-steady-state models can improve
VO
2
estimations during transitions. In our evaluation protocol,
we could not elicitate abrupt VO
2
changes within one activity,
e.g. when changing the treadmill speed and inclination level,
there was a gradual change, which could last up to 30
seconds. Hence, it was difficult to even collect representative
data that could benefit from non-steady-state modeling within
an activity (e.g. walking at different speeds) and the VO
2
7
estimation did not provide improvements over the steady-state.
A set of thirty activities of varying intensities was used in
this study and further work could extend the set, for example
including activities involving carrying loads. However, we
believe that the lifestyle activities used here are representative
and relevant to assess VO
2
estimation in non-stationary daily
life conditions. The generalized method proposed in this work
is able to model abrupt within-activity changes in VO
2
without
modifications to the model, since it detects transitions relying
on the CV, instead of the activity type. While it is not feasible
for us to robustly detect activities such as carrying a weight
using a wearable sensor on the chest, the use of HR partially
solves the problem. Carrying a load would create higher
strain, thus raise HR, which is used in the estimation models,
and therefore result in higher predicted VO
2
. In this work,
CV changes were shown to even indicate intensity changes
within the same activity confirming the method’s sensitivity.
By properly modeling VO
2
during transitions and locating
periods of non-steady-state, we quantified VO
2
for aerobic
dynamics during the onset of exercise, as well as during ADLs
characterized by short duration, with 16% error reduction
compared to current methods.
REFERENCES
[1] M. Altini, J. Penders, and O. Amft, “Personalizing energy expenditure
estimation using a cardiorespiratory fitness predicate, in Pervasive
Computing Technologies for Healthcare (PervasiveHealth), 2013 7th
International Conference on. IEEE, 2013, pp. 65–72.
[2] A. G. Bonomi, “Improving assessment of daily energy expenditure
by identifying types of physical activity with a single accelerometer.
Journal of Applied Physiology, vol. 107, no. 3, pp. 655–661, 2009.
[Online]. Available: http://www.ncbi.nlm.nih.gov/pubmed/19556460
[3] E. Tapia, “Using machine learning for real-time activity recognition and
estimation of energy expenditure, in PhD thesis, MIT, 2008.
[4] C. B. Scott, “Contribution of anaerobic energy expenditure to whole
body thermogenesis, Nutr Metab, vol. 2, pp. 14–23, 2005.
[5] K. Wasserman, A. L. Van Kessel, and G. G. Burton, “Interaction of
physiological mechanisms during exercise, J Appl Physiol, vol. 22,
no. 1, pp. 71–85, 1967.
[6] K. R. Short and D. A. Sedlock, “Excess postexercise oxygen consump-
tion and recovery rate in trained and untrained subjects, Journal of
Applied Physiology, vol. 83, no. 1, pp. 153–159, 1997.
[7] S. W. Su, L. Wang, B. G. Celler, and A. V. Savkin, “Oxygen uptake
estimation in humans during exercise using a hammerstein model,
Annals of biomedical engineering, vol. 35, no. 11, pp. 1898–1906, 2007.
[8] S. W. Su, B. G. Celler, A. V. Savkin, H. T. Nguyen, T. M. Cheng, Y. Guo,
and L. Wang, “Transient and steady state estimation of human oxygen
uptake based on noninvasive portable sensor measurements, Medical
& biological engineering & computing, vol. 47, no. 10, pp. 1111–1117,
2009.
[9] M. S. Orendurff, J. A. Schoen, G. C. Bernatz, A. D. Segal, and G. K.
Klute, “How humans walk: bout duration, steps per bout, and rest
duration, J Rehabil Res Dev, vol. 45, no. 7, pp. 1077–89, 2008.
[10] M. M. Reeves, P. S. Davies, J. Bauer, and D. Battistutta, “Reducing
the time period of steady state does not affect the accuracy of energy
expenditure measurements by indirect calorimetry, Journal of Applied
Physiology, vol. 97, no. 1, pp. 130–134, 2004.
[11] J. Wang, S. J. Redmond, M. Voleno, M. R. Narayanan, N. Wang,
S. Cerutti, and N. H. Lovell, “Energy expenditure estimation during nor-
mal ambulation using triaxial accelerometry and barometric pressure,
Physiological measurement, vol. 33, no. 11, pp. 1811–1830, 2012.
[12] T. Yoshida and B. J. Whipp, “Dynamic asymmetries of cardiac output
transients in response to muscular exercise in man. The Journal of
Physiology, vol. 480, no. Pt 2, pp. 355–359, 1994.
[13] J. Smolander, T. Juuti, M.-L. Kinnunen, K. Laine, V. Louhevaara,
K. M
¨
annikk
¨
o, and H. Rusko, A new heart rate variability-based
method for the estimation of oxygen consumption without individual
laboratory calibration: application example on postal workers, Applied
Ergonomics, vol. 39, no. 3, pp. 325–331, 2008.
[14] A. Pulkkinen, J. Kettunen, K. Martinm
¨
aki, S. Saalasti, and H. Rusko,
“On–and off dynamics and respiration rate enhance the accuracy of
heart rate based vo2 estimation, in Proceedings of the ACSM Congress,
Indianapolis, Abstract, vol. 36, 2004.
[15] L. R. Dugas, L. van der Merwe, H. Odendaal, T. D. Noakes, and
E. V. Lambert, A novel energy expenditure prediction equation for
intermittent physical activity. Medicine and Science in Sports and
Exercise, vol. 37, no. 12, pp. 2154–2161, 2005.
[16] F.-C. Chuang, Y.-T. C. Yang, and J.-S. Wang, Accelerometer-based
energy expenditure estimation methods and performance comparison,
in 2nd International Conference on Advances in Computer Science and
Engineering (CSE 2013). Atlantis Press, 2013, pp. 99–103.
[17] E. Børsheim and R. Bahr, “Effect of exercise intensity, duration and
mode on post-exercise oxygen consumption, Sports Medicine, vol. 33,
no. 14, pp. 1037–1060, 2003.
[18] S. Chen, J. Lach, O. Amft, M. Altini, and J. Penders, “Unsupervised
activity clustering to estimate energy expenditure with a single body
sensor, in Body Sensor Networks (BSN), 2013 IEEE International
Conference on. IEEE, 2013, pp. 1–6.
[19] R. Ali, L. Atallah, B. Lo, and G.-Z. Yang, “Detection and analysis
of transitional activity in manifold space, Information Technology in
Biomedicine, IEEE Transactions on, vol. 16, no. 1, pp. 119–128, 2012.
[20] C. Zhu and W. Sheng, “Recognizing human daily activity using a single
inertial sensor, in Intelligent Control and Automation (WCICA), 2010
8th World Congress on. IEEE, 2010, pp. 282–287.
[21] N. Christenfeld, L. M. Glynn, and W. Gerin, “On the reliable assessment
of cardiovascular recovery: An application of curve-fitting techniques,
Psychophysiology, vol. 37, no. 4, pp. 543–550, 2000.
[22] L. Bouarfa, L. Atallah, R. M. Kwasnicki, C. Pettitt, G. Frost, and G.-Z.
Yang, “Predicting free-living energy expenditure using a miniaturized
ear-worn sensor: An evaluation against doubly labelled water, IEEE
Transactions on Biomedical Engineering, vol. 61, no. 2, pp. 566–575,
2014.
Marco Altini received the M.Sc. degree in engi-
neering and computer science from the University of
Bologna in 2010. He is currently pursuing the Ph.D.
degree with the Technical University of Eindhoven,
Eindhoven, The Netherlands. His current research
interests include machine learning techniques for
the development and implementation of biomedical
applications using wearable sensors data.
Julien Penders received the M.Sc. degree in systems
engineering from the University of Liege, Belgium,
and the M.Sc. degree in biomedical engineering
from Boston University, in 2004 and 2006, respec-
tively. He is currently the Program Manager with the
Holst Centre/IMEC, Eindhoven, The Netherlands.
He has authored or co-authored over 30 papers in
journals and conference proceedings on body area
networks and autonomous wireless sensor networks.
Oliver Amft received the M.Sc. from TU Chem-
nitz in 1999 and the Ph.D. from ETH Zurich in
2008, both in Electrical Engineering and Information
Technology. Oliver is a Full Professor (W3) and
Chair of Sensor Technology at University of Passau.
He is also affiliated with the Wearable Computing
Lab, ETH Zurich (CH) and the Signal Processing
Systems section at TU Eindhoven (NL). Oliver is
interested in multi-modal activity recognition and
human behaviour inference algorithms and has co-
authored more than 90 publications in this field.
... It is used to provide general medical thresholds and guidelines for people with chronic diseases such as obesity and Type 2 diabetes (Hupin et al., 2015). In addition, steady state VO 2 measurements are considered to be the gold standard for estimating energy expenditure (EE) in light to moderate steady motion (Scott, 2005;Altini et al., 2015). The peak VO 2 reached during incremental motion is called the maximum VO 2 (as VO 2 max). ...
... The transition process between states is not considered in the experiment. Altini et al. (2015) mentioned that transition states have an impact on the estimation results. While in our experiment, the model does not need to distinguish the type of states and it shows good estimation results throughout the experiment. ...
... In the previous paper, Altini et al. (2015) proposed activityspecific linear functions to model steady-state activities and transition-specific non-linear functions to model non-steadystate activities and transitions. The result showed that the MAE between the predicted and true results of activity-specific models based on distinguishing activity states is lower than other linear or nonlinear models. ...
Article
Full-text available
Oxygen uptake (VO 2 ) is an important parameter in sports medicine, health assessment and clinical treatment. At present, more and more wearable devices are used in daily life, clinical treatment and health care. The parameters obtained by wearables have great research potential and application prospect. In this paper, an instantaneous VO 2 estimation model based on XGBoost was proposed and verified by using data obtained from a medical-grade wearable device (Beijing SensEcho) at different posture and activity levels. Furthermore, physiological characteristics extracted from single-lead electrocardiogram, thoracic and abdominal respiration signal and tri-axial acceleration signal were studied to optimize the model. There were 29 healthy volunteers recruited for the study to collect data while stationary (lying, sitting, standing), walking, Bruce treadmill test and recuperating with SensEcho and the gas analyzer (Metalyzer 3B). The results show that the VO 2 values estimated by the proposed model are in good agreement with the true values measured by the gas analyzer (R ² = 0.94 ± 0.03, n = 72,235), and the mean absolute error (MAE) is 1.83 ± 0.59 ml/kg/min. Compared with the estimation method using a separate heart rate as input, our method reduced MAE by 54.70%. At the same time, other factors affecting the performance of the model were studied, including the influence of different input signals, gender and movement intensity, which provided more enlightenment for the estimation of VO 2 . The results show that the proposed model based on cardio-pulmonary physiological signals as inputs can effectively improve the accuracy of instantaneous VO 2 estimation in various scenarios of activities and was robust between different motion modes and state. The VO 2 estimation method proposed in this paper has the potential to be used in daily life covering the scenario of stationary, walking and maximal exercise.
... Both Su et al. (2007) and Savkin et al. (2012) proposed to use a nonlinear Hammerstein model to express the transfer betweenV O 2 and the running speed, andV O 2 and the HR and respiratory frequency respectively. Nonlinear models based on machine learning and sensor fusion were proposed in Altini et al. (2015), Beltrame et al. (2017), Shandhi et al. (2020). Even though the proposed models allow the user to predict the physiological response of the exercising individual their applicability is often restricted toV O 2 prediction and do not takeV CO 2 into account. ...
... Since the intensity of the stairs exercise can be adjusted by the exercise phase (ascending or descending) without causing any discomfort to the participants, the regular stairs exercise is a proper method to investigate the cardiopulmonary response to the interval training exercise. Due to the difficulties of practically measuringV O 2 in free living, different methods to estimateV O 2 using wearable sensors have been developed 16 . The ascending or descending phase can be considered as a square signal if we describe ascending and descending as two numerical indicators. ...
Preprint
Full-text available
Oxygen uptake plays a vital role in evaluating endurance performance during exercise and is widely used for metabolic assessment. In this study, the oxygen uptake during the exercise phase (i.e. ascending or descending) of the stairs exercise are recorded and the experimental dataset features ten participants and a range of different exercise periods. Based on the designed experiment protocol, a non-parametric modeling method with kernel-based regularization is generally applied to estimate the oxygen uptake changes during the switching stairs exercise that is close to the daily life status. Compared with the fixed-order models on accuracy, stability and compatibility, the modeling results demonstrate the effectiveness of the non-parametric modeling approach. The influence of exercise duration on estimated fitness reveals that the model of the phase-oxygen uptake system is not time-invariant related to respiratory metabolism regulation and muscle fatigue. Accordingly, it allows us to study the humans' conversion mechanism at different metabolic rates and facilitates the standardization and development of exercise prescriptions.
... Since the intensity of the stairs exercise can be adjusted by the exercise phase (ascending or descending) without causing any discomfort to the participants, the regular stairs exercise is a proper method to investigate the cardiopulmonary response to the interval training exercise. Due to the difficulties of practically measuringV O 2 in free living, different methods to estimateV O 2 using wearable sensors have been developed 16 . The ascending or descending phase can be considered as a square signal if we describe ascending and descending as two numerical indicators. ...
Preprint
Full-text available
Background: Oxygen uptake plays a vital role in evaluating endurance performance during exercise and is widely used for metabolic assessment. In order to guarantee the accuracy of oxygen uptake estimation, suitable modeling method applied to estimating cardiorespiratory response during interval training exercise should be investigated. Methods: In this study, the oxygen uptake during the exercise phase (i.e. ascending or descending) of the stairs exercise is recorded and the experimental dataset features ten participants and a range of different exercise periods. Based on the designed experiment protocol, a non-parametric modeling method with kernel-based regularization is generally applied to estimate the oxygen uptake changes during the interval stairs exercise that is close to the daily life status. Results: The non-parametric model exhibits the smallest variance and higher accuracy than fixed-order models in both one period and two periods during exercise. Moreover, the fitness of two periods is significantly lower than those of half or one period for the non-parametric model and the IR model is quite different in different periods, which indicates that the model of the system is changing with exercise periods. Conclusions: Compared with the fixed-order models on accuracy, stability and compatibility, the modeling results demonstrate the effectiveness of the non-parametric modeling approach. The influence of exercise duration on estimated fitness reveals that the model of the phase-oxygen uptake system is not time-invariant related to respiratory metabolism regulation and muscle fatigue. Accordingly, it allows us to study the humans’ conversion mechanism at different metabolic rates and facilitates the standardization and development of exercise prescriptions.
... Studies in recent times employ accelerometerbased measurements for recognition and classification of various physical activities that can provide vital information about individual's health and functional ability (Casamassima et al., 2014;Kwapisz et al., 2011;Chung et al., 2008). Some researchers have also worked on the use of wearable sensors and heart rate monitors for activity recognition and EE estimation during non-steady states and transitions (Altini et al., 2015;Park et al.,2017). In our study, linear motion extracted from the accelerometer measurement is used to determine the patient's level of exercise or physical activity. ...
Preprint
Full-text available
While physical activity has been described as a primary prevention against chronic diseases, strenuous physical exertion under adverse ambient conditions has also been reported as a major contributor to exacerbation of chronic respiratory conditions. Maintaining a balance by monitoring the type and the level of physical activities of affected individuals, could help in reducing the cost and burden of managing respiratory ailments. This paper explores the potentiality of motion sensors in Smartphones to estimate physical activity thresholds that could trigger symptoms of exercise induced respiratory conditions (EiRCs). The focus is on the extraction of measurements from the embedded motion sensors to determine the activity level and the type of activity that is tolerable to individuals respiratory health. The calculations are based on the correlation between Signal Magnitude Area (SMA) and Energy Expenditure (EE). We also consider the effect of changes in the ambient conditions like temperature and humidity, as contributing factors to respiratory distress during physical exercise. Real time data collected from healthy individuals were used to demonstrate the potentiality of a mobile phone as tool to regulate the level of physical activities of individuals with EiRCs. We describe a practical situation where the experimental outcomes can be applied to promote good respiratory health.
... These approaches are generally modeled as regression problems, where a machine learning model learns a transformation function between physiological inputs from wearable sensors and _ VO 2 measured using a gas analyzer system. Earlier work primarily used a combination of HR and activity-related inputs to predict _ VO 2 during various forms of exercise 19,20 . Sensors embedded in textile fabrics have enabled _ VO 2 prediction in low-to-moderate-intensity exercise during activities of daily living 21,22 . ...
Article
Full-text available
Oxygen consumption (V̇O2) provides established clinical and physiological indicators of cardiorespiratory function and exercise capacity. However, V̇O2 monitoring is largely limited to specialized laboratory settings, making its widespread monitoring elusive. Here we investigate temporal prediction of V̇O2 from wearable sensors during cycle ergometer exercise using a temporal convolutional network (TCN). Cardiorespiratory signals were acquired from a smart shirt with integrated textile sensors alongside ground-truth V̇O2 from a metabolic system on 22 young healthy adults. Participants performed one ramp-incremental and three pseudorandom binary sequence exercise protocols to assess a range of V̇O2 dynamics. A TCN model was developed using causal convolutions across an effective history length to model the time-dependent nature of V̇O2. Optimal history length was determined through minimum validation loss across hyperparameter values. The best performing model encoded 218 s history length (TCN-VO2 A), with 187, 97, and 76 s yielding <3% deviation from the optimal validation loss. TCN-VO2 A showed strong prediction accuracy (mean, 95% CI) across all exercise intensities (−22 ml min−1, [−262, 218]), spanning transitions from low–moderate (−23 ml min−1, [−250, 204]), low–high (14 ml min−1, [−252, 280]), ventilatory threshold–high (−49 ml min−1, [−274, 176]), and maximal (−32 ml min−1, [−261, 197]) exercise. Second-by-second classification of physical activity across 16,090 s of predicted V̇O2 was able to discern between vigorous, moderate, and light activity with high accuracy (94.1%). This system enables quantitative aerobic activity monitoring in non-laboratory settings, when combined with tidal volume and heart rate reserve calibration, across a range of exercise intensities using wearable sensors for monitoring exercise prescription adherence and personal fitness.
... These approaches are generally modeled as regression problems, where a machine learning model learns a transformation function between physiological inputs from wearable sensors andVO 2 measured using a gas analyzer system. Earlier work primarily used a combination of HR and activity-related inputs to predictVO 2 during various forms of exercise [19], [20]. Sensors embedded in textile fabrics have enabledVO 2 prediction in low-to moderateintensity exercise during activities of daily living [21], [22]. ...
Preprint
Full-text available
Oxygen consumption (VO$_2$) provides established clinical and physiological indicators of cardiorespiratory function and exercise capacity. However, VO$_2$ monitoring is largely limited to specialized laboratory settings, making its widespread monitoring elusive. Here, we investigate temporal prediction of VO$_2$ from wearable sensors during cycle ergometer exercise using a temporal convolutional network (TCN). Cardiorespiratory signals were acquired from a smart shirt with integrated textile sensors alongside ground-truth VO$_2$ from a metabolic system on twenty-two young healthy adults. Participants performed one ramp-incremental and three pseudorandom binary sequence exercise protocols to assess a range of VO$_2$ dynamics. A TCN model was developed using causal convolutions across an effective history length to model the time-dependent nature of VO$_2$. Optimal history length was determined through minimum validation loss across hyperparameter values. The best performing model encoded 218 s history length (TCN-VO$_2$ A), with 187 s, 97 s, and 76 s yielding less than 3% deviation from the optimal validation loss. TCN-VO$_2$ A showed strong prediction accuracy (mean, 95% CI) across all exercise intensities (-22 ml.min$^{-1}$, [-262, 218]), spanning transitions from low-moderate (-23 ml.min$^{-1}$, [-250, 204]), low-heavy (14 ml.min$^{-1}$, [-252, 280]), ventilatory threshold-heavy (-49 ml.min$^{-1}$, [-274, 176]), and maximal (-32 ml.min$^{-1}$, [-261, 197]) exercise. Second-by-second classification of physical activity across 16090 s of predicted VO$_2$ was able to discern between vigorous, moderate, and light activity with high accuracy (94.1%). This system enables quantitative aerobic activity monitoring in non-laboratory settings across a range of exercise intensities using wearable sensors for monitoring exercise prescription adherence and personal fitness.
... The emerging field of digital biomarkers may help in the future to provide solutions like this [44]. In the past, home-based VO2max estimates have been technically challenging to develop [45,46] though technology continues to improve [47] with projects ongoing (http://sagebionetworks.org/researchprojects/cardiorespiratory-fitness-module/). We also found that though we implemented a standardized home-based interval exercise testing program in order to try to control dose intensity of the intervention, the rigidity of the protocol was challenging from an adherence standpoint and may not have been optimally suited for participants of varying levels of baseline motivation and fitness. ...
Article
Full-text available
Allogeneic hematopoietic cell transplantation (alloHCT) is a life-saving technology that can cure otherwise incurable diseases, but imposes significant physiologic stress upon recipients. This stress leads to short-term toxicity and mid- to long-term physical function impairment in some recipients. Exercise interventions have demonstrated preliminary efficacy in preserving physical function in HCT recipients, but the role of these interventions prior to HCT (prehabilitative) is less known. We tested a 5- to 12-week, prehabilitative higher intensity home-based aerobic exercise intervention in a randomized study of alloHCT candidates. Of 113 patients screened, 34 were randomized to control or intervention groups, 16 underwent pre- and post-intervention peak oxygen consumption (VO2peak) testing, and 12 underwent pre- and post-intervention 6-min walk distance (6MWD) testing. No significant differences in VO2peak or 6MWD were seen pre- to post-intervention between intervention and control groups, but final numbers of evaluable participants in each group were too small to draw inferences regarding the efficacy of the intervention. We conclude that the design of our prehabilitative intervention was not feasible in this pilot randomized study, and make recommendations regarding the design of future exercise intervention studies in alloHCT.
Article
Full-text available
Recreational exercisers continue to take a greater interest in monitoring their personal fitness levels. One of the more notable measurements that are monitored and estimated by wrist-worn tracking devices is maximum aerobic capacity (VO2max), which is currently the accepted measure of cardiorespiratory fitness. Traditional methods of obtaining VO2max present expensive barriers, whereas new wearable technology, such as of the Garmin Fenix 5x (GF5) provides a more cost-effective alternative. PURPOSE: To determine the validity of the GF5 VO2max estimation capabilities against the ParvoMedics TrueOne 2400 (PMT) metabolic measurement system in recreational runners. METHODS: Twenty-five recreational runners (17 male and 8 female) ages 18-55 participated in this study. Participants underwent two testing sessions: one consisting of the Bruce Protocol utilizing the PMT, while the other test incorporated the GF5 using the Garmin outdoor protocol. Both testing sessions were conducted within a few days of each other, with a minimum of 24 hours rest between sessions. RESULTS: The mean VO2max values for the PMT trial (49.1 ± 8.4 mL/kg/min) and estimation for the GF5 trial (47 ± 6.0 mL/kg/min) were found to be significantly different (t = 2.21, p = 0.037). CONCLUSION: The average difference between the GF5 estimation and the PMT was 2.16 ml/kg/min. Therefore, the watch is not as accurate compared to a PMT for obtaining VO2max. However, although not statically significant, the proximity of scores to the PMT shows that the GF5 can be an option for a person seeking an affordable and easily available method of determining VO2max.
Conference Paper
Full-text available
Body sensor networks (BSNs) have provided the opportunity to monitor energy expenditure (EE) in daily life and with that information help reduce sedentary behavior and ultimately improve human health. Current approaches for EE estimation using BSNs require tedious annotation of activity types and multiple body sensor nodes during data collection and high accuracy activity classifiers during post processing. These drawbacks impede deploying this technology in daily life — the primary motivation of using BSNs to monitor EE. With the goal of achieving the highest EE estimation accuracy with the least invasiveness and data collection effort, this paper presents an unsupervised, single-node solution for data collection and activity clustering. Motivated by a previous finding that clusters of similar activities tend to have similar regression models for estimating EE, we apply unsupervised clustering to implicitly group activities with homogeneous features and generate specific regression models for each activity cluster without requiring manual annotation. The framework therefore does not require specific activity classification, hence eliminating activity type labels. With leave-one-subject-out cross-validation across 10 subjects, an RMSE of 0.96 kcal/min was achieved, which is comparable to the activity-specific model and improves upon a single regression model.
Conference Paper
Full-text available
Accurate Energy Expenditure (EE) estimation is key in understanding how behavior and daily physical activity (PA) patterns affect health, especially in today's sedentary society. Wearable accelerometers (ACC) and heart rate (HR) sensors have been widely used to monitor physical activity and estimate EE. However, current EE estimation algorithms have not taken into account a person's cardiorespiratory fitness (CRF), even though CRF is the main cause of inter-individual variation in HR during exercise. In this paper we propose a new algorithm, which is able to significantly reduce EE estimate error and inter-individual variability, by automatically modeling CRF, without requiring users to perform specific fitness tests. Results show a decrease in Root Mean Square Error (RMSE) between 28 and 33% for walking, running and biking activities, compared to state of the art activity-specific EE algorithms combining ACC and HR.
Conference Paper
Full-text available
As robot assisted living is becoming increasingly important for elderly people, human daily activity recognition is necessary for human-robot interaction. In this paper, we proposed an approach to daily activity recognition for elderly people. This approach uses a single wearable inertial sensor worn on the right thigh of a human subject to collect motion data. This setup can reduce the obtrusiveness to the minimum. Human daily activities can be recognized in two steps. First, two neural networks are used to classify the basic activities. Second, the activity sequence is modeled by an HMM to consider the sequential constraints exhibited in human daily life and the modified short-time Viterbi algorithm is used for realtime daily activity recognition as the fine-grained classification. We conducted experiments in a mock apartment environment and the obtained results proved the effectiveness and accuracy of our approach.
Article
Obesity is now considered a global epidemic and is predicted to become the number one preventive health threat in the industrialized world. Presently, over 60% of the U.S. adult population is overweight and 30% is obese. This is of concern because obesity is linked to leading causes of death, such as heart and pulmonary diseases, stroke, and type 2 diabetes. The dramatic rise in obesity rates is attributed to an environment that provides easy access to high caloric food and drink and promotes low levels of physical activity. Unfortunately, many people have a poor understanding of their own daily energy (im)balance: the number of calories they consume from food compared with what they expend through physical activity. Accelerometers offer promise as an objective measure of physical activity. In prior work they have been used to estimate energy expenditure and activity type. This work further demonstrates how wireless accelerometers can be used for real-time automatic recognition of physical activity type, intensity, and duration and estimation of energy expenditure. The parameters of the algorithms such as type of classifier/regressor, feature set, window length, signal preprocessing, sensor set utilized and their placement on the human body are selected by performing a set of incremental experiments designed to identify sets of parameters that may balance system usability with robust, real-time performance in low processing power devices such as mobile phones. The algorithms implemented are evaluated using a dataset of examples of 52 activities collected from 20 participants at a gymnasium and a residential home. The algorithms presented here may ultimately allow for the development of mobile phone-based just-in-time interventions to increase self-awareness of physical activity patterns and increases in physical activity levels in real-time during free-living that scale to large populations.
Conference Paper
This paper presents two accelerometer-based metabolic equivalent (MET) estimation methods: 1) MET regression models estimation; and 2) a mono-exponential MET estimation equation. The MET estimation methods can predict MET for the following seven categories of activities: sitting, standing, walking upstairs, walking downstairs, level walking, running, and cycling. The intensity of these activities ranged from sedentary ( 1.5 METs) to vigorous (> 6 METs), where 1 MET is equivalent to the metabolic rate associated with sitting. First, MET regression model estimations were developed, including a single MET regression model and multiple MET regression models. Then, the second estimation method, a mono-exponential MET estimation equation, was developed to enhance estimation accuracy in non-steady and steady states during exercise. The experimental results successfully validate the effectiveness of the proposed method, the mono-exponential MET estimation equation, which substantially ameliorates estimation errors in non-steady states and achieves satisfactory accuracy for both non-steady and steady states.
Article
Accurate estimation of daily total Energy Expenditure (EE) is a prerequisite for assisted weight management and assessing certain health conditions. The use of wearable sensors for predicting free-living EE is challenged by consistent sensor placement, user compliance and estimation methods used. This paper examines whether a single ear-worn accelerometer can be used for EE estimation under free-living conditions. An EE prediction model was first derived and validated in a controlled setting using healthy subjects involving different physical activities. Ten different activities were assessed showing a ten-fold cross validation error of 0.24. Furthermore, the EE prediction model shows a Mean Absolute Deviation (MAD) below 1.2 Metabolic Equivalent of Tasks [METs]. The same model was applied to a free-living setting with a different population for further validation. The results were compared against those derived from Doubly Labelled Water (DLW). In free-living settings, the predicted daily EE has a correlation of 0:74; p = 0:008 and a MAD of 272kcal=day. These results demonstrate that laboratory-derived prediction models can be used to predict EE under free-living conditions.
Article
Energy expenditure (EE) is an important parameter in the assessment of physical activity. Most reliable techniques for EE estimation are too impractical for deployment in unsupervised free-living environments; those which do prove practical for unsupervised use often poorly estimate EE when the subject is working to change their altitude by walking up or down stairs or inclines. This study evaluates the augmentation of a standard triaxial accelerometry waist-worn wearable sensor with a barometric pressure sensor (as a surrogate measure for altitude) to improve EE estimates, particularly when the subject is ascending or descending stairs. Using a number of features extracted from the accelerometry and barometric pressure signals, a state space model is trained for EE estimation. An activity classification algorithm is also presented, and this activity classification output is also investigated as a model input parameter when estimating EE. This EE estimation model is compared against a similar model which solely utilizes accelerometry-derived features. A protocol (comprising lying, sitting, standing, walking, walking up stairs, walking down stairs and transitioning between activities) was performed by 13 healthy volunteers (8 males and 5 females; age: 23.8 ± 3.7 years; weight: 70.5 ± 14.9 kg), whose instantaneous oxygen uptake was measured by means of an indirect calorimetry system (K4b(2), COSMED, Italy). Activity classification improves from 81.65% to 90.91% when including barometric pressure information; when analyzing walking activities alone the accuracy increases from 70.23% to 98.54%. Using features derived from both accelerometry and barometry signals, combined with features relating to the activity classification in a state space model, resulted in a [Formula: see text] estimation bias of -0.00 095 and precision (1.96SD) of 3.54 ml min(-1) kg(-1). Using only accelerometry features gives a relatively worse performance, with a bias of -0.09 and precision (1.96SD) of 5.99 ml min(-1) kg(-1), with the largest errors due to an underestimation of [Formula: see text] when walking up stairs.
Article
Activity monitoring is important for assessing daily living conditions for elderly patients and those with chronic diseases. Transitions between activities can present characteristic patterns that may be indicative of quality of movement. To detect and analyze transitional activities, a manifold-based approach is proposed in this paper. The proposed method uses a recursive spectral graph-partitioning algorithm to segment transitions in activity. These segments are subsequently mapped to a reference manifold space. Categorization of transitions is performed with the corresponding features in the manifold space. The practical value of the work is demonstrated through data collected under laboratory conditions, as well as patients recovering from total knee replacement operations, demonstrating specific transitions and motion impairment compared to normal subjects.