Visualizing changes in lower body coordination with different types of foot orthoses using self-organizing maps (SOM)

Heterogeneity of Treatment Effects for Intensive Blood Pressure Therapy by Individual Components of FRS: An Unsupervised Data-Driven Subgroup Analysis in SPRINT and ACCORD

Background

Few studies have answered the guiding significance of individual components of the Framingham risk score (FRS) to the risk of cardiovascular disease (CVD) after antihypertensive treatment. This study on the systolic blood pressure intervention trial (SPRINT) and the Action to Control Cardiovascular Risk in Diabetes blood pressure trial (ACCORD-BP) aimed to reveal previously undetected association patterns between individual components of the FRS and heterogeneity of treatment effects (HTEs) of intensive blood pressure control.

Methods

A self-organizing map (SOM) methodology was applied to identify CVD-risk-specific subgroups in the SPRINT (n = 8,773), and the trained SOM was utilized directly in 4,495 patients from the ACCORD. The primary endpoints were myocardial infarction (MI), non-myocardial infarction acute coronary syndrome (non-MI ACS), stroke, heart failure (HF), death from CVD causes, and a primary composite cardiovascular outcome. Cox proportional hazards models were then used to explore the potential heterogeneous response to intensive SBP control.

Results

We identified four SOM-based subgroups with distinct individual components of FRS profiles and the CVD risk. For individuals with type 2 diabetes mellitus (T2DM) in the ACCORD or without diabetes in the SPRINT, subgroup I characterized by male with the lowest concentrations for total cholesterol (TC) and high-density lipoprotein (HDL) cholesterol measures, experienced the highest risk for major CVD. Conversely, subgroup III characterized by a female with the highest values for these measures represented as the lowest CVD risk. Furthermore, subgroup II, with the highest systolic blood pressure (SBP) and no antihypertensive agent use at baseline, had a significantly greater frequency of non-MI ACS under intensive BP control, the number needed to harm (NNH) was 84.24 to cause 1 non-MI ACS [absolute risk reduction (ARR) = −1.19%; 95% CI: −2.08, −0.29%] in the SPRINT [hazard ratio (HR) = 3.62; 95% CI: 1.33, 9.81; P = 0.012], and the NNH of was 43.19 to cause 1 non-MI ACS (ARR = −2.32%; 95% CI: −4.63, 0.00%) in the ACCORD (HR = 1.81; 95% CI: 1.01–3.25; P = 0.046). Finally, subgroup IV characterized by mostly younger patients with antihypertensive medication use and smoking history represented the lowest risk for stroke, HF, and relatively low risk for death from CVD causes and primary composite CVD outcome in SPRINT, however, except stroke, a low risk for others were not observed in ACCORD.

Conclusion

Similar findings in patients with hypertensive with T2DM or without diabetes by multivariate subgrouping suggested that the individual components of the FRS could enrich or improve CVD risk assessment. Further research was required to clarify the potential mechanism.

A global view on how local muscular fatigue affects human performance

There is a growing interest in scientific literature on identifying how and to what extent interventions applied to a specific body region influence the responses and functions of other seemingly unrelated body regions. To investigate such a construct, it is necessary to have a global multivariate model that considers the interaction among several variables that are involved in a specific task and how a local and acute impairment affects the behavior of the output of such a model. We developed an artificial neural network (ANN)-based multivariate model by using parameters of motor skills obtained from kinematic, postural control, joint torque, and proprioception variables to assess the local fatigue effects of the abductor hip muscles on the functional profile during a single-leg drop landing and a squatting task. Findings suggest that hip abductor muscles' local fatigue produces a significant effect on a general functional profile, built on different control systems. We propose that expanded and global approaches, such as the one used in this study, have great applicability and have the potential to serve as a tool that guarantees ecological validity of future investigations.

Use of self-organizing maps to study sex- and speed-dependent changes in running biomechanics

BACKGROUND

Up to 79% of runners get injured every year, with higher rates of injuries occurring in females than males. A self-organizing map (SOM) is a type of artificial neural network that can be used to inspect large datasets and study coordination patterns. The purpose of this study was to use an SOM to study the effects of sex and speed on biomechanical coordination patterns.

METHOD

Thirty-two healthy runners ran on an instrumented treadmill at their long slow distance speed (LSD) and at speed 30% faster (LSD + 30%). Vertical ground reaction force (vGRF), vertical tibial acceleration, step parameters, electromyograms (EMG) of six lower limb muscles, and joint angles were collected across speeds. Rate of loading (ROL), tibial impact shock (TIS), coupling angle variability (CAV) and movement pattern proportions for hip/knee sagittal and hip frontal / knee sagittal plane couplings, peak EMG, step length, step rate, and knee and ankle joint angle at initial contact were used as an input for the SOM (37 variables).

RESULTS

The analysis identified four clusters (i.e., running patterns). While males and females showed similar distribution across clusters at LSD (p = .36) and at LSD + 30% (p = .51), females did exhibit a significant (p = .03) shift between clusters as the speed increased from LSD to LSD + 30% whereas males did not (p = .17). The shift was associated with an increase in TIS, ROL, step length, step rate, vastus lateralis EMG, hip flexion/knee extension movement pattern proportion, and a decrease in ST EMG and CAV_IC for hip sagittal/knee sagittal coupling.

CONCLUSION

As running speed increased there was a significant change in the coordination pattern in females, which was characterized by increases in several variables that are purported risk factors for running related injuries.

Changes in balance coordination and transfer to an unlearned balance task after slackline training: a self-organizing map analysis

How humans maintain balance and change postural control due to age, injury, immobility or training is one of the basic questions in motor control. One of the problems in understanding postural control is the large set of degrees of freedom in the human motor system. Therefore, a self-organizing map (SOM), a type of artificial neural network, was used in the present study to extract and visualize information about high-dimensional balance strategies before and after a 6-week slackline training intervention. Thirteen subjects performed a flamingo and slackline balance task before and after the training while full body kinematics were measured. Range of motion, velocity and frequency of the center of mass and joint angles from the pelvis, trunk and lower leg (45 variables) were calculated and subsequently analyzed with an SOM. Subjects increased their standing time significantly on the flamingo (average +2.93 s, Cohen's d = 1.04) and slackline (+9.55 s, d = 3.28) tasks, but the effect size was more than three times larger in the slackline. The SOM analysis, followed by a k-means clustering and marginal homogeneity test, showed that the balance coordination pattern was significantly different between pre- and post-test for the slackline task only (χ ² = 82.247; p < 0.001). The shift in balance coordination on the slackline could be characterized by an increase in range of motion and a decrease in velocity and frequency in nearly all degrees of freedom simultaneously. The observation of low transfer of coordination strategies to the flamingo task adds further evidence for the task-specificity principle of balance training, meaning that slackline training alone will be insufficient to increase postural control in other challenging situations.

Analyses of a cirrhotic patient's evolution using self organizing mapping and Child-Pugh scoring

Due to the importance of cirrhosis evolution, this study examined cirrhotic patients using Self Organizing Mapping (SOM) based on the Child-Pugh scoring method. Because Colored Doppler Ultrasound (CDU) has too many parameters, scoring can be a very difficult task. Classifying cirrhotic patients via SOM and investigating weights of the cirrhotic CDU parameters are aimed in this study. SOM was used to map high dimensional cirrhotic data onto two dimensional clustered data. These clusters provided a feature map of cirrhotic patients. In this study, 103 cirrhotic patients and a control group of 44 healthy individuals were examined in the hospital, and parameters were analyzed using SOM. These data were obtained using CDU, and age and sex parameters were analyzed in this study. Cirrhotic patients were histopathologically separated into subgroups using the Child-Pugh scoring method, and the presence of ascites was determined using SOM. In this study, differences between the control group and cirrhotic patients with their subgroups were investigated using SOM, and the results were discussed. Renal artery indices, hepatic artery indices, portal vein parameters, age and the degree of ascites were analyzed using SOM for a total of 147 individuals. The combination of SOM and Child-Pugh scoring method can be useful for the interpretation of cirrhotic patient's evolution. Computer-based SOM algorithm and negative effectiveness of a large scale dataset could be minimized by adjusting the weight of the parameters. This study will faciliate doctors to make better decisions for their patients.

Prolonged cycling alters stride time variability and kinematics of a post-cycle transition run in triathletes

Previous studies have employed relatively short cycling protocols to investigate the effect of cycling on muscle activation and kinematics in running. The aim of this study was to investigate the effect of 3h of cycling on stride time variability (STV), stride length, tibialis anterior (TA) activation, and lower limb range of motion (ROM) in a transition run. Eight triathletes completed a run-cycle-run protocol. Data were collected from a pre-cycle run and a transition run after 3h of cycling. STV, stride length and ROM were assessed using three-dimensional motion analysis, and TA activation was recorded using surface electromyography. Results showed that compared with the pre-cycle run triathletes exhibited increased STV (Cohen's d=0.95) and shorter strides (d=0.15) in the transition run (p<0.05). TA activation and ROM did not change. After 10min of transition running, ankle and hip ROM significantly increased (d=0.40 and 0.41 respectively) compared to the beginning of the transition run (p<0.05) but no other changes were observed. The results suggest that locomotor control and kinematics in a transition run are affected by prolonged cycling and stride time variability is potentially a novel method of evaluating the immediate effect of prolonged cycling on the locomotor control of running.

Multi-dimensional coordination in cross-country skiing analyzed using self-organizing maps

This study sought to ascertain how multi-dimensional coordination patterns changed with five poling speeds for 12 National Standard cross-country skiers during roller skiing on a treadmill. Self-organizing maps (SOMs), a type of artificial neural network, were used to map the multi-dimensional time series data on to a two-dimensional output grid. The trajectories of the best-matching nodes of the output were then used as a collective variable to train a second SOM to produce attractor diagrams and attractor surfaces to study coordination stability. Although four skiers had uni-modal basins of attraction that evolved gradually with changing speed, the other eight had two or three basins of attraction as poling speed changed. Two skiers showed bi-modal basins of attraction at some speeds, an example of degeneracy. What was most clearly evident was that different skiers showed different coordination dynamics for this skill as poling speed changed: inter-skier variability was the rule rather than an exception. The SOM analysis showed that coordination was much more variable in response to changing speeds compared to outcome variables such as poling frequency and cycle length.

The impact of altered task mechanics on timing and duration of eccentric bi-articular muscle contractions during cycling

In order to understand muscle adaptations to altered task mechanics during cycling, this study investigated the impact of altered seat height and cadence on timing and duration of gastrocnemius (GAST), biceps femoris (BF) and vastus lateralis (VL) eccentric contractions and muscle activation patterns, and cycling economy. Ten male cyclists completed 9 × 5 min of cycling at 3 seat heights and 3 cadences. Three-dimensional leg kinematics and muscle activation patterns were recorded to estimate timing of eccentric muscle contractions. Onset, offset and duration of eccentric contractions and, onset, offset and duration of muscle activation were calculated, along with cycling economy. Duration of GAST and VL eccentric contractions decreased with increasing seat height due to earlier offset of eccentric muscle contractions. Duration of BF eccentric contractions significantly increased with seat height due to a later eccentric contraction offset. Offset of GAST and BF muscle activation occurred earlier with increasing cadence. Cycling economy was significantly affected by cadence but not seat height. The results suggest that as a consequence of altered seat height, proprioceptive feedback is used to fine-tune the timing of bi-articular eccentric muscle contractions. These results may have implications for seat height self-selection.

Visualization of trunk muscle synergies during sitting perturbations using self-organizing maps (SOM)

The purpose of this study was to demonstrate the use of the self-organizing map (SOM) method for visualization, modeling, and comparison of trunk neuromuscular synergies during perturbed sitting. Thirteen participants were perturbed at the level of the sternum, in eight directions during sitting. Electromyographic (EMG) responses of ten trunk muscles involved in postural control were recorded. The SOM was used to encode the EMG responses on a 2-D projection (i.e., visualization). The result contains similar patterns mapped close together on the plot therefore forming clusters of data. Such visualization of ten EMG responses, following eight directional perturbations, allows for comparisons of direction-dependent postural synergies. Direction-dependent neuromuscular response models for each muscle were then constructed from the SOM visualization. The results demonstrated that the SOM was able to encode neuromuscular responses, and the SOM visualization showed direction-dependent differences in the postural synergies. Moreover, each muscle was modeled using the SOM-based method, and derived models showed that all muscles, except for one, produced a Gaussian fit for direction-dependent responses. Overall, SOM analysis offers a reverse engineering method for exploration and comparison of complex neuromuscular systems, which can describe postural synergies at a glance.