Abstract
There have been increasing cases of people seeking treatment for neck and back pain. The most common cause of neck and back pain is due to long-term poor sitting posture. The most common poor sitting posture cases are humpback, and head and neck being too far forward. It is easy to cause neck and back pain and other symptoms. Therefore, the development of wearable posture monitoring system for dynamic assessment of sitting posture becomes both helpful and necessary. In addition to recording the wearer’s posture when sitting with quantitative assessment, it is needed to execute real-time action feedback for correctness of posture, in order to reduce neck and back pain due to long-term poor sitting posture. This study completed an instant recording and dynamic assessment of position measurement and feedback system. The system consists of a number of dynamic measurement units that can describe the posture trajectory, which integrates three-axis gyro meter, three-axis accelerometer, and magnetometer in order to measure the dynamic tracking. In the reliability analysis experiment, angle measurement error is less than 2%. The correlation coefficient between correlation analysis and Motion Analysis (MA) is 0.97. It is shown that the motion trajectory of this system is highly correlated with MA. In the feasibility test of sitting position detection, it is possible to detect the sitting position from the basic action of the walking, standing, sitting and lying down, and the sensitivity reaches 95.84%. In the assessment of the sitting position, the information published by the Canadian Centre for Occupational Health and Safety was used, as well as the recommendations of professional physicians as a basis for evaluating the threshold of the sitting measurement parameters and immediately feedback to the subjects. The system developed in this study can be helpful to reduce neck and back pain due to long-term poor sitting posture.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
In recent years, there have been increasing cases of people seeking treatment for neck and back pain. The most common cause of neck and back pain is due to long-term poor sitting posture. The amounts of time people sit in a day account for a large proportion of the day. Sedentary lifestyle is one of the important causes of skeletal muscle system diseases, and the risk of sedentary lifestyle is not fully understood compared to the risk of heavy labor activity [1]. Poor posture and sedentary lifestyle will increase the risk of lower back pain, arthritis, thrombosis, lumbar disc herniation, spinal deformity, cervical spur, body hypoxia, prostate problems and rectal cancer [2, 3]. Although the medical profession is not fully aware of the specific relationship between sedentary lifestyle and skeletal muscle system diseases, many studies have revealed the relationship between sedentary lifestyle and diabetic mortality [4]. The American Association for Cancer Research (AACR) surveyed 120,000 people live in 14 years. It was believed that sitting for 6 or more hours per day may increase the risk of death [5]. The angle between the thighs and the torso of the average person is almost 90° when sitting. This posture will cause the pelvis to tilt backward, and the pelvis will tilt forward when standing [6,7,8]. The pelvis, which is tilted back 30°, almost makes the bottom of the sacrum curved to a position parallel to the ground, which will make people unconsciously bend over [9,10,11,12,13]. Blood circulation is an important part of the oxygen and nutrients needed to transmit to the body. A sedentary lifestyle is a type of lifestyle with little or no physical activity, which will cause abnormal frequency of heartbeat, blood circulation and other symptoms [14]. The lymphatic system is part of the circulatory system and a vitally important part of the immune system [15]. Since the lymphatic system does not have a heart to pump it, it relies on muscle activity to promote the circulation of lymph nodes [16]. The main function of the lymphatic system is to remove waste products from tissues via the infiltration and circulation of the tissue fluid [17]. For example, our legs will be swollen after a long flight or train, because the body is in a relatively quiescent and sedentary state, and the posture is not correct, resulting in lymphatic obstruction, which leads to tissue fluid accumulation in the lower limbs, causing edema [18]. Slouching can reduce the chest and abdomen space, affect the function of the respiratory and digestive systems [19] and lead to an unbalanced body. Long-term slouching will make the cervical spine uneven, oppressing the stretching of the neck muscle. This posture develops habits, which will make the body exert excessive force, causing muscle tension and pain. The body may not be able to restore the natural posture, especially for children during their physical development period [20,21,22]. Quantitative life uses the input, state and performance parameters to express personal daily life [23]. The input refers to the external factors of body’s absorption, such as digested food, air quality etc.; the state is a current characteristic, such as mood, skin conductance, oxygen saturation, etc. [24]; the performance is the behavior of the human body, divided into psychological and physical behavior. In this paper, the posture is quantified into walking, standing, sitting and lying. International research on walking, standing, sitting and lying are integrated into this study, including the distinction of action, the rehabilitation of posture, the assessment and feedback of actions, the types of sensors and the method of measurement, as shown in Table 1.
In the study of walking and standing, mainly for gait analysis, including pace, step frequency and bending angle. Through these parameters, it is possible to assess the subject’s center of gravity, habits, rehabilitation. In the study of sitting, mainly for the assessment of poor posture caused by muscle soreness, the assessment can be accomplished by measuring the angle of the spine or the center of gravity and the posture of sitting can be corrected. In the study of lying, there are only few studies of the project, and mainly as a distinction between physical activities. Via these studies (Table 1), few researches were found that focused on walking, standing, sitting and lying at the same time.
Therefore, it is necessary to develop a wearable posture monitoring system for dynamic assessment of sitting posture. The system consists of a number of dynamic measurement units (DMU) that can describe the posture trajectory, which integrates three-axis gyro meter, three-axis accelerometer, and magnetometer in order to measure the dynamic track. The system can describe and distinguish between walking, standing, sitting and lying. These represent the basic activities of the body. In this study, the system to assessment of sitting posture was applied as the beginning of the system.
According to Table 1, the sensors used in the currently marketed products or published research are mainly divided into an inertial sensor, an accelerometer, an optical motion system, a gyroscope, a pressure sensor and an Electromyography sensor (EMG). The differences between our systems and listed products or published research are as follows:
- i.
Our system is a wearable system, not limited to the space of the venue, optical motion system and pressure sensor obviously do not apply.
- ii.
This study needs to use the three-axis rotation angle to record the motion track of the measured point. accelerometer, gyroscope and EMG obviously do not apply.
- iii.
The difference with the optical motion system is that our system is low in cost and short in calibration time.
- iv.
Our system uses inertial sensor, the listed products or the published research is mainly divided into timed up and go, respiratory effort monitoring system, sleeping position recognition system, recognition posture (lying, standing, sitting and walking). What sets us apart is the combination of recognizing four postures (lying, standing, sitting and walking) and assessing sitting posture.
Materials and methods
The purpose of this study is to develop a wearable posture monitoring system for dynamic assessment of sitting posture. The system consists of a number of DMUs (Fig. 1) that were developed by the authors. The DMU can describe the posture trajectory, which integrates three-axis gyro meter, three-axis accelerometer, and magnetometer in order to measure the dynamic track. The system needs to verify reliability and feasibility. To assure reliability, the DMU and the commercial Motion Analysis (MA) were used to measure the object’s single-point motion track, and then correlation analysis of DMU and MA to determine whether there is a high degree of correlation. In the feasibility test of sitting position detection, the sitting position can be differentiated from the basic action of the walking, standing, sitting and lying, the feasibility of this system for physical activity measurement and long-term recording can be verified.
Reliability analysis
The experimental flow chart is shown in Fig. 2. The steps are described below.
Experimental group calibration
The three axes of the stereoscopic space are usually called XYZ axes, here the X-axis is the longitudinal axis, the Y-axis is the lateral axis, and the Z-axis is the vertical axis. Roll is defined as the DMU rotation left or right along the longitudinal axis. Pitch is defined as the DMU tilts up or down along the lateral axis. Yaw is defined as the DMU going left or right along the vertical axis (Fig. 3). The output signal of The DMU is Roll, Pitch, Yaw. When performing an accuracy analysis, the DMU requires a three-axis tilt test, which uses the angle of the inclinometer and the angle measured by the DMU for error analysis. The error should be less than 5%, otherwise the DMU will need to be adjusted until the expected results are obtained.
Control group
MA is the preferred action capture system in the market. The system consists of an Eagle digital capture lens, power hub, Ethernet switch and Cortex software, which can be extremely accurate in capturing complex action. After setting up the MA system, it must be calibrated first. A long 500 mm-pole with light balls on both sides of the pole (Spot A, B) are used and the pole will be swung. The movement trajectory in 60 s of the two light balls are recorded, and the distance between the 2 points are calculated, averaged, and compared with the original length (500 mm). The calibration is completed when the error is less than 5%.
Single dynamic trajectory test
Step
-
i.
Place the DMU and the MA on the test pole at the same time. Schematic diagram shown in Fig. 4.
-
ii.
Perform a series of motion tests (90 s).
-
iii.
The output value of is converted.
-
iv.
The correlation between total rotation angle of DMU and MA is calculated.
Comparison analysis
The comparison analysis uses the total rotation angle as the basis. The output value of the MA is the spatial coordinate of the three axes, and the output value of the DMU is the rotation angle of the three axes. Because each output value is to be compared with each other, they must be converted. The first is the spatial coordinates of the MA to the rotation angle. Suppose one is given 2 points A and B on the test pole, the vector \(\overset{\lower0.5em\hbox{$\smash{\scriptscriptstyle\rightharpoonup}$}}{AB} \;{\text{is}}\;\left( {B_{X} - A_{X} ,B_{Y} - A_{Y} ,B_{Z} - A_{Z} } \right)\) at (t − 1), the vector \(\overset{\lower0.5em\hbox{$\smash{\scriptscriptstyle\rightharpoonup}$}}{A^{\prime}B^{\prime}} \;{\text{is}}\;\left( {B^{\prime}_{X} \; - \;A^{\prime}_{X} ,B^{\prime}_{Y} \; - \;A^{\prime}_{Y} ,B^{\prime}_{Z} \; - \;A^{\prime}_{Z} } \right)\) at t. The schematic diagram shown in Fig. 5.
The output value of the DMU is the rotation angle of the three axes (Roll, Pitch, Yaw), where the rotation angle at (t − 1) is \(\left( {A_{t - 1} ,B_{t - 1} ,C_{t - 1} } \right)\), and the rotation angle at t is \(\left( {A_{t} ,B_{t} ,C_{t} } \right)\).
Correlation
From Eqs. (1) and (2) we get \(V_{\theta } \cdot V_{\emptyset }\). The correlation coefficient of A and B is obtained by Eq. (3). If the correlation coefficient is close to +1.0, then there is a strong positive linear relationship between MA and DMU.
Zx and Zy are the standardized z-values for MA and DMU. If correlation coefficient r is 1, the MA and DMU have a complete positive correlation. In other words, the data points from MA and DMU lie on a perfectly straight, positively-sloped line. If correlation coefficient r is − 1, MA and DMU have a complete negative correlation. In other words, the data points from MA and DMU lie on a perfectly straight, negatively-sloped line. If correlation coefficient r is 0, MA and DMU have no correlation.
Feasibility analysis of sitting posture
Selection of DMU position and judgment of action
In this experiment, two DMUs were placed in the subject’s chest (nipple center, DMU C) and the lateral thigh (femoral center, DMU T) and the rotation angle (Roll, Pitch, Yaw) signal were analyzed to determine the current state of the subject (Fig. 6). When Rollc= − 90° ± 30° and Rollt= 0° ± 30°, the current state of action was determined as “Sitting”. When Rollc= − 90° ± 30° and Rollt= − 90° ± 30°, the current state of action was determined as “Standing”. For lying, the subject has three types of lying positions, when Rollc= 0° ± 30° and Rollt= 0° ± 30°, the current state of lying was determined as type 1 or type 3. When Rollc= (− 150° to − 180°∪)∪(150° to 180°) and Rollt= (− 150° to − 180°)∪(150° to 180°), the current state of action was determined as type 2. No matter what kind of position, the current state of action was determined as “Lying”. When Rollc = − 90° ± Rollc(t) and Rollt = − 90° ± Rollt(t), the current state of action was determined as “Walking”. If the system cannot determine whether the current state of action is one of walking, standing, sitting or lying, the current state of action was determined as “Unidentified”.
The subject participation
This experiment is mainly used to detect whether the system can recognize four basic physical activities (walking, standing, sitting, and lying) and assess sitting posture. Therefore, we choose a healthy male (age: 20–30 years old) as a system tester.
Measurement action and sampling frequency
The experiment took 20 min and the sampling rate of the DMU was set to 50 Hz. Four actions were taken, which are shown in Fig. 6, as a test, and each action was executed for about 300 s.
Signal filtering
In this study, the signal is a multi-to-one transmission, and multiple DMUs transmit the signal to the receiver. Noise was filtered using moving average. Moving average is a simple and smooth prediction method. Its basic idea is based on time series data, item by item, the average number of fixed sequence are calculated one by one. Moving average can smooth short-term fluctuations, which eliminates the noise caused by body shaking.
Data analysis and comparison
For DMU, the first task was to calibrate Rollc and Rollt Moving average was used to filter the noise, and the left and right ranks were set to five. The processed signals are defined as Rollcf and Rolltf. We used 1 s as the basis for judging the interval of action state and calculated the median of each interval. The calculated value: Rollcmed is the median of Rollcf and Rolltmed is the median of Rolltf. The judgment method is based in Sect. 2.2.1, to analyze the Rollcmed and Rolltmed, to distinguish between lying, sitting, standing, walking and unidentified (Fig. 7).
Sensitivity was used as the basis for the assessment of A value, which is the current measurement action (lying, sitting, standing, walking and unidentified).
Assessment of sitting posture
In the assessment of the sitting posture, we use the information published by the Canadian Centre for Occupational Health and Safety (CCOHS) organization and the recommendations of the professional physician as the basis for the study. After we have integrated, the proposed good “sitting posture” has the following:
- i.
To keep the joint angle: femur and back 90°–120°, Knee 90°–130°, Ankle 100°–120°.
- ii.
To keep the knee below the hip.
- iii.
The ankle is held in front of the knee.
- iv.
To keep the feet on the floor or foot pad.
- v.
Keep the upper body within 30° of the upright position.
- vi.
Do not be sedentary, it had better not to exceed 50 min.
We are based on the above 6 points as a basis, in accordance with the recommendations of professional physicians to assess the threshold of sitting parameters, and placement and number of DMU.
Results and discussion
Reliability analysis
Step A: DMU accuracy calibration
The inclinometer was used to do the calibration of the action. The calibration angle was selected in 0°, 45°, 90°, − 90°, − 45°. At each angle, five separate measurements were made. The error between the mean and the value of the inclinometer was within 2% (as shown in Table 2). This shows that the accuracy of the DMU had reached more than 98%.
Step B: MA calibration
The test pole was used to calibrate, and the MA captured the distance between the two spots on the test pole. The error as compared to the actual length was calculated (500 mm) (Fig. 8). The error of 603 samples was within 1.5%. Table 3 shows some of the data results.
Step C
In step C, 15 experiments were done with 90 s for each experiment, where the sampling rate of the MA was 70 Hz and the sampling rate of the DMU was 50 Hz.
Steps D and E: comparison and relevance
\(V_{\theta }\) and \(V_{\emptyset }\) was a one-dimensional array, and the sampling rate of MA was 70 Hz, with each experiment was 90 s, so the \(V_{\theta }\) array contained 6300 data. On the other hand, the sampling rate of DMU was 50 Hz, so the \(V_{\emptyset }\) array contained 4500 data. In order to compare the data between the two, the \(V_{\theta }\) array of MA was resampled to reduce the sampling rate to 50 Hz. Then low-pass filter was used to eliminate high-frequency noise and the cut-off frequency was set to 10 Hz. Lastly, moving average filter was used to smooth the data. Here, other objects or body touch usually causes the exception value to the DMU. The correlation between DMU and MA is shown in Table 4, and the average correlation is 0.97. Figure 9 is a comparison chart of the total average rotation angle of DMU and MA. The solid line represents the DMU and the dotted line represents the MA. The horizontal axis is the number of samples, with a total of 4500 data (90 s, 50 Hz). In the reliability experiment, the accuracy of the DMU in the angular measurement was verified. With the comparison of synchronization (Table 4 and Fig. 9), there is a high correlation between DMU and MA. It is shown that the DMU is synchronized with the MA on the trajectory of the measurement action. So multiple DMUs can be used to place a specific point on the body, to describe and construct the dynamic track of the body.
Feasibility analysis of sitting posture
The system is designed to discriminate the sitting posture from other activities of daily life, such as walking, standing, sitting and lying. Tables 5, 6, 7 and 8 are the results of judgment of walking, standing, sitting and lying, Figs. 10, 11, 12 and 13 are the waveforms of lying, sitting, standing and walking, where the unit of the vertical axis is the angle in each figure.
For lying, \(Roll_{c} \left( {\text{Lying DMU C}} \right)\; \subset \;0^{ \circ } \; \pm \;30^{ \circ }\) and \(Roll_{t} \left( {\text{Lying DMU T}} \right)\; \subset \;0^{ \circ } \; \pm \;30^{ \circ }\) (Fig. 10). The action judgment of each interval is the same as the actual posture (Table 5), then the sensitivity of lying is calculated \(S_{{ {\text{lying}}}} \; = \;\frac{274}{274} \times 100\% = 100\%\).
In the part of sitting, most \(Roll_{c} \left( {\text{Sitting DMU C}} \right)\; \subset \; - 90^{ \circ } \; \pm \;30^{ \circ }\) and \(Roll_{t}\)\(\left( {\text{Sitting DMU T}} \right) \subset - 90^{^\circ } \pm 30^{^\circ }\)(Fig. 11). From Table 6, it is clear that the result of the judgment of 28 1-s intervals is different from actual action. The rest of which, the posture judgment of each interval is the same as actual posture, then the sensitivity of sitting is calculated \(S_{{ {\text{sitting}}}} = \frac{274}{302} \times 100\% = 90.72\%\).
For standing, \(Roll_{c} \left( {\text{Standing DMU C}} \right)\; \subset \; - 90^{ \circ } \pm 30^{ \circ }\) and \(Roll_{t} \left( {\text{Standing DMU T}} \right) \subset \; - 90^{ \circ } \pm 30^{ \circ }\) (Fig. 11). The action judgment of each interval is the same as the actual posture (Table 7), then the sensitivity of standing is calculated \(S_{\text{standing}} = \frac{295}{295} \times 100\% = 100\%\).
In the part of walking, most \(Roll_{c} \left( {\text{Walking DMU C}} \right)\; \subset \; - 90^{ \circ } \pm Roll_{c} (t)\) and \(Roll_{t}\)\(Roll_{t} \left( {\text{Walking DMU T}} \right) \subset \; - 90^{ \circ } Roll_{t} (t)\) (Fig. 13). From Table 8, it is clear that the result of the judgment of 19 1-s intervals is different from actual action. The rest of which, the posture judgment of each interval is the same as actual posture, then the sensitivity of walking is \(S_{{ {\text{walking}}}} = \frac{239}{258} \times 100\% = 92.64\%\).
The average sensitivity of all postures is \(S_{{ {\text{average}}}} \; = \;\frac{274 + 274 + 295 + 239}{274 + 302 + 295 + 258} \times 100\% = 95.84\%\). This shows that the system can differentiate the sitting posture from other activities of the daily life, such as walking, standing, sitting and lying.
Assessment of sitting posture
According to the information released by the CCOHS organization and the advice of professional physicians, we have developed a standard for assessing sitting posture. According to the actual test of the system by the subject, we can evaluate the number and placement of sensors. It requires four DMUs, the placement points of which is the back (seventh thoracic, t7), thigh lateral (femoral center), calf lateral (tibia center) and head occipital bone. There are three rules of judgment for assessment of sitting posture, “to keep the joint angle”, “to keep the upper body within 30° of the upright position” and “do not be sedentary exceeding 50 min”. Each rule has its own threshold setting. When the threshold is exceeded, after the system has performed analysis, the program will give feedback to the subject.
The evaluation threshold and corresponding reminder feedback are as follows:
- i.
To keep the joint angle: femur and back (90°–120°), knee (90°–130°), ankle (100°–120°): The angle of the femur and the back can be measured by two DMUs (back and lateral thigh). The angle of the knee can be measured by two DMUs (thigh lateral and calf lateral). The angle of the ankle can be measured by the DMU placed on the calf lateral. When the angle is not within the threshold setting, the system will feedback the subject to correct the posture. The feedback part uses the horn sound to make a reminder. In this evaluation project, three short beeps will be issued when the evaluation angle is not within the threshold range.
- ii.
To keep the upper body within 30° of the upright position: It can be measured by two DMUs (back and head occipital bone). When the angle is not within the threshold setting, the system will feedback the subject to correct. The feedback part uses the horn sound to make a reminder. In this evaluation project, one long and two short beeps will be issued when the evaluation angle is not within the threshold range.
- iii.
Do not be sedentary, it had better not to exceed 50 min: If 50 min remain in the sitting posture, the system will immediately remind and advise the subject to move around. The feedback part uses the horn sound to make a reminder. In this evaluation project, three long beeps will be issued when the evaluation angle is not within the threshold range.
At present, the results of this part have completed the preliminary evaluation of the threshold setting of sitting posture and feedback reminder. However, people of different heights and weights may have different threshold settings. Therefore, it is necessary to measure more subjects in the future, analyze the differences between the subjects, and find the optimal threshold to improve the system.
Conclusions
At this stage of the research results, a real-time recording and dynamic assessment of position measurement and feedback system is completed. The system consists of a number of dynamic measurement units (DMU) that can describe the posture trajectory, which integrates three-axis gyro meter, three-axis accelerometer, and magnetometer in order to measure the dynamic track. In the reliability analysis experiment, angle measurement error is less than 2%, and the correlation coefficient between correlation analysis and MA is 0.97. It is proved that the motion trajectory of this system is highly correlated with MA. In the feasibility test of sitting position detection, the sitting position can be differentiated from the basic action of the walking, standing, sitting and lying, and the sensitivity is 95.84%. In the assessment of the sitting position, the information published by CCOHS and the recommendations of professional physicians was used as a basis to evaluate the threshold of the sitting measurement parameters to provide immediate feedback to the subjects, in order to reduce neck and back pain due to long-term poor sitting posture. For the future development, in order to allow the subjects to more easily understand the real information and interaction, human–machine interface (HMI) will need to be improved.
Future work
In the future, we have three main work items: the improvement of threshold setting, accuracy test for sitting posture feedback and human–machine interaction.
- i.
People of different heights and weights may have different threshold settings. Therefore, it is necessary to measure more subjects in the future, analyze the differences between the subjects, and find the optimal threshold to improve the system.
- ii.
After the improvement of the threshold setting, the accuracy test of sitting posture feedback is performed.
- iii.
At present, the human–machine interface of our system is mainly the beep of the speaker, although it is intuitive but lacks interest. For the current society, mobile phones are an indispensable 3C product for people. People are used to carrying them. Therefore, in order to make users more convenient and used to use the system, we will combine mobile phone software to enhance the human–machine interface.
References
Schneider E, Irastorza X (2010) Work-related musculoskeletal disorders in the EU: facts and figures. European Agency for Safety and Health at Work, Bilbao
American Cancer Society (2010) Epidemiology research program. Am J Epidemiol 172(4):419–429
Pereira S, Power C et al (2013) Sedentary behaviours in mid-adulthood and subsequent body mass index. PLoS ONE. https://doi.org/10.1371/journal.pone.0065791
Uffelen J, Wong J et al (2010) Occupational sitting and health risks—a systematic review. Am J Prev Med 39(4):379–388
Katzmarzyk P, Lee IM et al (2012) Sedentary behaviour and life expectancy in the USA: a cause—deleted life table analysis. BMJ open. https://doi.org/10.1136/bmjopen-2012-000828
Aissaoui R, Lacoste M, Dansereau J et al (2001) Analysis of sliding and pressure distribution During a repositioning of persons in a simulator chair. IEEE Trans Neural Sys Rehabil Eng 9(2):215–224
Schneider R (1998) Pantograph for tilting trains. In: IEEE current collections for high speed trains seminar (Ref. No. 1998/509) 2/1-2/6
Neville L (2005) The fundamental principles of seating and positioning in children and young people with physical disabilities. University of Ulster, Coleraine
Mandal AC (1976) Work chair with tilting seat. Ergonomics 19(2):157–164
Hobson D (1992) Comparative effects of posture on pressure and shear at the body-seat interface. J Rehabil Res Dev 29(4):21–31
Koo T, Mak A, Lee YL et al (1996) Posture effect on seating interface biomechanics: comparison between two seating cushions. Arch Phys Med Rehabil 77(1):40–47
Mandal AC (1991) Investigation of the lumbar flexion of the seated man. Int J Ind Ergon 8(1):75–87
Rosemeyer B (1972) Die aufrechte körperhaltung des menschen. Eine vergleichende Untersuchung 112:115–163
Redfern R (2014) The miracle enzyme is serrapeptase. serrapeptase the 2nd gift from silkworms giving the answer to pain chronic inflammation and clogged arteries. Naturally Healthy Publications, Davenham
Martini F, Timmons M, Tallitsch R (2014) Human anatomy, 8th edn. Pearson Education, London
Gottlieb B (2002) Alternative cures: the most effective natural home remedies for 160 health problems. Rodale, Emmaus
Owen N, Healy G, Matthews C, Dunstan D (2010) Too much sitting: the population-health science of sedentary behavior. Exerc Sport Sci Rev 38(3):105–113
Grøntved A, Hu F (2011) Television viewing and risk of type 2 diabetes, cardiovascular disease, and all-cause mortality: a meta-analysis. JAMA 305(23):2448–2455
Seperson S (2008) Elder care and service learning: a handbook. Information Age Publishing, Charlotte
Mandal AC (1996) Balanced sitting posture on forward sloping seat. http://www.acmandal.com/. Accessed 22 June 2015
Prevention Magazine Editors (ed) (1996) Symptoms: their causes & cures: how to understand and treat 265 health concerns. Bantam, New York
Tremblay M, LeBlanc A, Kho M et al (2011) Systematic review of sedentary behaviour and health indicators in school-aged children and youth. Int J Behav Nutr Phys Act 8(1):98
Bloch D (2012) The Coolest Startups in America. Building Bloch Books, New York, pp 215–217
Millikan GA (1942) The oximeter: an instrument for measuring continuously oxygen-saturation of arterial blood in man. Rev Sci Instrum 13(10):434–444
Andersson BJG, Ortengren R et al (1974) Lumbar disc pressure and myoelectric back muscle activity during sitting. I. Studies on an experimental chair. Scandinavian. J Rehabil Med 6(3):104–114
Andersson BJG, Ortengren R (1974) Lumbar disc pressure and myoelectric back muscle activity during sitting. II. Studies on an office chair. Scandinavian. J Rehabil Med 6(3):115–121
Watson DH, Trott PH (1993) Cervical headache: an investigation of natural head posture and upper cervical flexor muscle performance. Cephalagia 13(4):272–284
Azrin N, Rubin H, O’Brien F, Ayllon T, Roll D (1968) Behavioral engineering: postural control by a portable operant apparatus. J Appl Behav Anal 1(2):99–108
Wong MS, Mak AF, Luk KD, Evans JH, Brown B (2001) Effectiveness of audio-biofeedback in postural training for adolescent idiopathic scoliosis patients. Prosthet Orthot Int 25(1):60–70
Okada S, Sakaki T, Hirata R, Okajima Y, Uchida S, Tomita Y (2001) TEM: a therapeutic exercise machine for the lower extremities of spastic patients. Adv Robot 14(7):597–606
Hirata R, Sakaki T, Okada S, Nakamoto Z, Hiraki N, Okajima Y et al (2002) BRMS bio responsive motion system. In: IEEE/RSJ international conference on intelligent robots and systems pp 1344–1348
Bernhardt M, Frey M, Colombo G, Riener R (2005) Hybrid force-position control yields cooperative behaviour of the rehabilitation robot LOKOMAT. In: Proceedings of the 2005 IEEE 9th international conference on rehabilitation robotics pp 536–539
Lunenburger L, Colombo G, Riener R, Dietz V (2005) Clinical assessments performed during robotic rehabilitation by the gait training robot Lokomat. In: Proceedings of the 2005 IEEE 9th international conference on rehabilitation robotics pp 345–348
Mayagoitia R, Nene A, Veltink P (2002) Accelerometer and rate gyroscope measurement of kinematics: an inexpensive alternative to optical motion analysis systems. J Biomech 35(4):537–542
Roetenberg D, Slycke P, Veltink P (2007) Ambulatory position and orientation tracking fusing magnetic and inertial sensing. IEEE Trans Biomed Eng 54(5):883–890
Yang S, Li Q (2012) Inertial sensor-based methods in walking speed estimation: a systematic review. Sensors 12(5):6102–6116
Rueterbories J, Spaich E, Larsen B, Andersen O (2010) Methods for gait event detection and analysis in ambulatory systems. Med Eng Phys 32(6):545–552
Watanabe T, Saito H, Koike E, Nitta K (2011) A preliminary test of measurement of joint angles and stride length with wireless inertial sensors for wearable gait evaluation system. Comput Intel Neurosci. https://doi.org/10.1155/2011/975193
Aminian K, Robert P, Buchser EE, Rutschmann B, Hayoz D, Depairon M (1999) Physical activity monitoring based on accelerometry: validation and comparison with video observation. Med Biol Eng CompuT 37(3):304–308
Mathie M, Coster A, Lovell N, Celler B (2004) Accelerometry: providing an integrated, practical method for long-term, ambulatory monitoring of human movement. Physiol Meas 25(2):R1–R20
Grant P, Ryan C, Tigbe W, Granat M (2006) The validation of a novel activity monitor in the measurement of posture and motion during everyday activities. Br J Sports Med 40(12):992–997
Mansfield A, Lyons G (2003) The use of accelerometry to detect heel contact events for use as a sensor in FES based assisted walking. Med Eng Phys 25(10):879–885
Liu K, Liu T, Shibata K, Inoue Y, Zheng T (2009) Novel approach to ambulatory assessment of human segmental orientation a wearable sensor system. J Biomech 42(16):2747–2752
Willemsen A, van Alsté JA, Boom HBK (1990) Real-time gait assessment utilizing a new way of accelerometry. J Biomech 23(8):859–863
Duric M, Dosen S, Popovic M, Popovic D (2008) Sensors for assistive system for restoring of the walking. In: Proceedings 52nd ETRAN conference, society for electronics, telecommunications, computers, automatic control and nuclear engineering
Shimada Y, Ando S, Matsunaga T, Misawa A, Aizawa T, Shirahata T, Itoi E (2005) Clinical application of acceleration sensor to detect the swing phase of stroke gait in functional electrical stimulation. Tohoku J Exp Med 207(3):197–202
Willemsen A, Friego C, Boom HBK (1991) Lower extremity angle measurement with accelerometers-errors and sensitivity analysis. IEEE Trans Biomed Eng 38(12):1186–1193
Sabatini A, Martelloni C, Scapellato S, Cavallo F (2005) Assessment of walking features from foot inertial sensing. IEEE Trans Biomed Eng 52(3):486–494
Wu G, Ladin Z (1996) The study of kinematic transients in locomotion using the integrated kinematic sensor. IEEE Trans Rehabil Eng 4(3):193–200
Williamson R, Andrews B (2000) Gait event detection for FES using acceleromenters and supervised machine learning. IEEE Trans Rehabil Eng 8(3):312–319
Lee SW, Mase K, Kogure K (2005) Detection of spatio-temporal gait parameters by using wearable motion sensors. In: IEEE engineering in medicine and biology 27th annual conference vol 7, pp 6836–6839
M. Duric (2008) Automatic recognition of gait phases from accelerations of leg segments. In: Neural network applications in electrical engineering
Williamson R, Andrews B (2000) Sensor system for lower limb functional electrical stimulation (FES) control. Med Eng Phys 22(5):313–325
Morris J (1973) Accelerometry-a technique for the measurement of human body movements. J Biomech 6(6):729–736
O’Donovan K, Kamnik R, O’Keeffe D, Lyons G (2007) An inertial and magnetic sensor based technique for joint angle measurement. J Biomech 40(12):2604–2611
Coley B, Najafi B, Paraschiv-Ionescu A, Aminian K (2005) Stair climbing detection during daily physical activity using a miniature gyroscope. Gait Posture 22(4):287–294
Aminian K, Najafi B, Bla C, Lezvraz P, Robert P (2002) Spatio-temporal parameters of gait measured by an ambulatory system using miniature gyroscopes. J Biomech 35(5):689–699
Jasiewicz J, Allum J, Middleton J, Barriskill A, Condie P, Purcell B, Li RC (2006) Gait event detection using linear accelerometers or angular velocity transducers in able-bodied and spinal-cord injured individuals. Gait Posture 24(4):502–509
Tong K, Granat M (1999) A practical gait analysis system using gyroscopes. Med Eng Phys 21(2):87–94
Pappas IP, Popovic MR, Keller T, Dietz V, Morari M (2001) A reliable gait phase detection system. IEEE Trans Neural Syst Rehabil Eng 9(2):113–125
Veltink P, Slycke P, Hemssems J, Buschman R, Bultstra G, Hermens H (2003) Three dimensional inertial sensing of foot movements for automatic tuning of a two-channel implantable drop-foot stimulator. Med Eng Phys 25(1):21–28
Dai R, Stein R, Andrews B, James K, Wieler M (1996) Application of tilt sensors in functional electrical stimulation. IEEE Trans Rehabil Eng 4(2):63–72
Weber D, Stein B, Chan K, Loeb G, Richmond F, Rolf R, James K, Chong S (2005) BIONic WalkAide for correcting foot drop. IEEE Trans Neural Syst Rehabil Eng 13(2):242–246
Skelly M, Chizeck H (2001) Real-time gait event detection for paraplegic FES walking. IEEE Trans Neural Syst Rehabil Eng 9(1):59–68
Nilsson J, Stokes V, Thorstensson A (1985) A new method to measure foot contact. J Biomech 18(8):625–627
Pappas IP, Keller T, Mangold S, Popovic M, Dietz V, Morari M (2004) A reliable gyroscope-based gait-phase detection sensor embedded in a shoe insole. IEEE J Sensors 4(2):268–274
Claus A, Hides J, Moseley G, Hodges P (2016) Thoracic and lumbar posture behaviour in sitting tasks and standing: progressing the biomechanics from observations to measurements. Appl Ergon 53:161–168
Zemp R, Taylor W, Lorenzetti S (2016) Seat pan and backrest pressure distribution while sitting in office chairs. Appl Ergon 53:1–9
Claeys K, Brumagne S, Deklerck J, Vanderhaeghen J, Dankaerts W (2016) Sagittal evaluation of usual standing and sitting spinal posture. J Bodyw Mov Ther 20(2):326–333
Wong WY, Wong MS (2008) Detecting spinal posture change in sitting positions with tri-axial accelerometers. Gait Posture 27(1):168–171
Araújo F, Lucas R, Alegrete N, Azevedo A, Barros H (2014) Individual and contextual characteristics as determinants of sagittal standing posture: a population-based study of adults. Spine J 14(10):2373–2383
O’Sullivan K, O’Keeffe M, O’Sullivan L, O’Sullivan P, Dankaerts W (2013) Perceptions of sitting posture among members of the community, both with and without non-specific chronic low back pain. Manual Ther 18(6):551–556
Nguyen H, Lebel K, Boissy P, Bogard S, Goubault E, Duval C (2017) Auto detection and segmentation of daily living activities during a Timed Up and Go task in people with Parkinson’s disease using multiple inertial sensors. J NeuroEng Rehabil 14(1):26
Lee CH, Lee S, Shin G (2017) Reliability of forward head posture evaluation while sitting, standing, walking and running. Hum Mov Sci 55:81–86
Hernandez JE, Cretu E (2017) Respiratory effort monitoring system for sleep apnea screening for both supine and lateral recumbent positions. In: 8th IEEE annual information technology, electronics and mobile communication conference (IEMCON)
Eyobu OS, Kim YW, Cha D, Han DS (2018) A real-time sleeping position recognition system using IMU sensor motion data. In: IEEE international conference on consumer electronics (ICCE)
Hellmers S, Izadpanah B, Dasenbrock L, Diekmann R, Bauer JM, Hein A, Fudickar S (2018) Towards an automated unsupervised mobility assessment for older people based on inertial TUG measurements. Sensors 18(10):3310
Martinez-Hernandez U, Dehghani-Sanij AA (2019) Probabilistic identification of sit-to-stand and stand-to-sit with a wearable sensor. Pattern Recogn Lett 118:32–41
Acknowledgements
The authors would like to thank the Ministry of Science and Technology, Taiwan, ROC, for supporting this research under Contracts MOST 105-2221-E-035-064-.
Funding
This study was funded by the Ministry of Science and Technology, Taiwan, ROC (Grant Number: MOST 105-2221-E-035-064-).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. This article does not contain any studies with animals performed by any of the authors.
Informed consent
Informed consent was obtained from all individual participants included in the study.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Wu, CC., Chiu, CC. & Yeh, CY. Development of wearable posture monitoring system for dynamic assessment of sitting posture. Phys Eng Sci Med 43, 187–203 (2020). https://doi.org/10.1007/s13246-019-00836-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13246-019-00836-4