Using Machine Learning To Quantify Transverse Plane Lumbopelvic Rhythm

Lumbopelvic rhythm illustrates the relative motion between lumbar spine and pelvis during various activities such as lifting running could be used a biomarker for low back pain (LBP). Sagittal plane lumbopelvic has been extensively examined surrogate to measure risk factor, but trunk rotation, second component of couple is commonly missed. Since are time series not discrete variables, machine learning may viable solution in identifying clusters patterns healthy adults. PURPOSE: To categorize transverse using learning. METHODS: 80 adults with no history LBP (Young: n = 46; 26.9 ± 6.9 yr; Middle-Age: 33; 52.4 yr). 3D kinematics were calculated participants performed maximal rotation from right left. Coupling angles vector coding represented 4 coordination (in-phase, anti-phase, superior-only, inferior-only). K-means clustering (k 3) was segment coupling into clusters. Within each cluster, age groups compared. RESULTS: 3 distinct movement discovered (Figure 1). In plane, mostly moved in-phase cluster 1, start end while 2 started ended in-phase. Cluster switched in- anti- transitioning directions. Age differences seen only 1 where young middle-age by moving anti-phase. CONCLUSIONS: These represent different ways individual perform which along sagittal can potentially identify individuals LBP.Figure