METHODS: A convolutional auto-encoder (CAE) based nonlinear compression structure is implemented to reduce the signal size of arrhythmic beats. Long-short term memory (LSTM) classifiers are employed to automatically recognize arrhythmias using ECG features, which are deeply coded with the CAE network.
RESULTS: Based upon the coded ECG signals, both storage requirement and classification time were considerably reduced. In experimental studies conducted with the MIT-BIH arrhythmia database, ECG signals were compressed by an average 0.70% percentage root mean square difference (PRD) rate, and an accuracy of over 99.0% was observed.
CONCLUSIONS: One of the significant contributions of this study is that the proposed approach can significantly reduce time duration when using LSTM networks for data analysis. Thus, a novel and effective approach was proposed for both ECG signal compression, and their high-performance automatic recognition, with very low computational cost.
METHODS: Fifty computed tomography scans of nonarthritic knees were evaluated using three-dimensional image processing software. Four distal femoral rotational axes were determined in the axial plane: the transepicondylar axis (TEA), transcondylar axis (TCA), posterior condylar axis (PCA), and a line perpendicular to Whiteside's anterior-posterior axis. Then, angles were measured relative to the TEA. Tibial joint line obliquity was measured as the angle between the proximal tibial plane and a line perpendicular to the axis of the tibia.
RESULTS: There was a strong positive correlation between PCA-TEA and tibial joint line obliquity (r = 0.68, P < .001) as well as TCA-TEA and tibial joint line obliquity (r = 0.69, P < .001). In addition, the tibial joint line obliquity and TCA-TEA angles were similar, 3.7° ± 2.2° (mean ± standard deviation) and 3.5° ± 1.7°, respectively (mean difference, 0.2° ± 0.2°; P = .369).
CONCLUSION: Both PCA-TEA and TCA-TEA strongly correlated with proximal tibial joint line obliquity indicating a relationship between distal femoral rotational geometry and proximal tibial inclination. These findings could imply that the native knee in flexion attempts to balance the collateral ligaments toward a rectangular flexion space. A higher tibial varus inclination is matched with a more internally rotated distal femur relative to the TEA.
METHODS: Our aim was to determine normative EGJ metrics in a large international cohort of healthy volunteers undergoing HRM (Medtronic, Laborie, and Diversatek software) acquired from 16 countries in four world regions. EGJ-CI was calculated by the same two investigators using a distal contractile integral-like measurement across the EGJ for three respiratory cycles and corrected for respiration (mm Hg cm), using manufacturer-specific software tools. EGJ morphology was designated according to Chicago Classification v3.0. Median EGJ-CI values were calculated across age, genders, HRM systems, and regions.
RESULTS: Of 484 studies (28.0 years, 56.2% F, 60.7% Medtronic studies, 26.0% Laborie, and 13.2% Diversatek), EGJ morphology was type 1 in 97.1%. Median EGJ-CI was similar between Medtronic (37.0 mm Hg cm, IQR 23.6-53.7 mm Hg cm) and Diversatek (34.9 mm Hg cm, IQR 22.1-56.1 mm Hg cm, P = 0.87), but was significantly higher using Laborie equipment (56.5 mm Hg cm, IQR 35.0-75.3 mm Hg cm, P
METHODS: Children having home overnight oximetry for suspected OSA were identified over 12 months, and those with a normal result who went on to have polysomnography (PSG) were included. Oximetry, including PR-SD and PRI (rises of 8, 10 and 15 beats/min per hour), was analyzed using commercially available software. PR parameters were compared between those with OSA (obstructive apnoea-hypopnoea index (OAHI) >1 event/h) and those without OSA.
RESULTS: One hundred sixteen children had normal oximetry, of whom 93 (median age 4.5 years; 55 % M) had PSG. Fifty-seven of 93 (61 %) children had OSA (median OAHI 4.5 events/h, range 1.1-24). PR-SD was not different between the OSA and non-OSA groups (p = 0.87). PRI tended to be higher in those with OSA, but there was considerable overlap between the groups: PRI-8 (mean ± SD 58.5 ± 29.0/h in OSA group vs 48.6 ± 20.2/h in non-OSA group, p = 0.07), PRI-10 (45.1 ± 25.0 vs 36.2 ± 16.7, p = 0.06) and PRI-15 (24.4 ± 14.5 vs 18.9 ± 9.0, p = 0.04). A PRI-15 threshold of >35/h had specificity of 97 % for OSA.
CONCLUSION: The PRI-15 shows promise as an indicator of OSA in children with normal oximetry.