Recent advances in deep learning enable estimation of a patient’s "ECG age" from standard short-term 12-lead electrocardiogram (ECG). This study introduces sequential ECG age predictions in continuous 24-hour Holter ECG recordings of chronic heart failure (CHF) patients. Using publicly available data from the MUSIC study, we analyzed data from 869 CHF patients, assessing both the differences (predicted ECG age vs actual chronological age) and dynamic -including variability and entropy-based complexity- to characterize temporal fluctuations over 24h. After automated removal of segments based on signal-to-noise ratio (SNR), 222 deceased patients were matched 1:1 with 222 surviving patients by sex, NYHA class, and age. While the predicted ECG age is similar across CHF patients, our findings indicate lower variability but increased complexity (approximate and sample entropy) in deceased compared to surviving patients, suggesting more irregular predicted ECG age dynamics among those who experienced adverse outcomes. Our results suggest that longitudinal evaluation of predictions from an end-to-end deep learning model can uncover subtle temporal dynamics potentially valuable for risk stratification.

In this work, we evaluate the inter-patient magnitude of electrocardiography (ECG) alterations over time during ischemia. The freely available STAFF III database provides 12-lead ECG recordings of patients before, during, and after elective percutaneous coronary intervention (PCI), where a coronary vessel is widened with a balloon inflation. Patients undergoing this procedure often do not show standard ECG findings and thereby cannot be assessed by the subject-independent and absolute thresholds defined in clinical guidelines. We compute alterations of ST-segment and T-wave amplitudes w.r.t. QRS amplitude over time for each patient and lead. We demonstrate that quantifying these relative amplitudes changes over time is feasible in standard and non-standard ECG recordings. To demonstrate clinical relevance, we use the features for differentiating N=54 patients with atherosclerotic plaque in either the right coronary artery (RCA) or left anterior descending (LAD). Results show significant differences in five leads for ST-segment alterations and three leads for T-wave alterations, which are also suggested by clinical guidelines for ischemia detection.

The increasing availability of wearable electrocardiography (ECG) devices enables the continuous monitoring of individual ECG alterations. This could be beneficial for patients suffering from acute ischemia but with non-standard ECG findings that do not fit to the subject-independent and absolute thresholds defined in clinical guidelines. In this work, we evaluate the inter-patient magnitude of individual ECG alterations during ischemia. The freely available STAFF III database provides 12-lead ECG recordings of patients before, during, and after elective percutaneous coronary intervention(PCI), where a coronary vessel is widened with a balloon inflation. We compute individual alterations of ST-interval and T-wave amplitudes w.r.t. QRS amplitude over time for each patient and lead. We demonstrate that determining relative ST-interval/T-wave amplitudes and deriving individual alterations over time is feasible in standard and non-standard ECG recordings. To demonstrate clinical relevance, we use the features for differentiating N=54 STAFF III patients with atherosclerotic plaque in either the right coronary artery (RCA) or left ascending artery (LAD). Results show significant differences in 5 leads for ST-interval alterations and 3 leads for T-wave alterations, which are also suggested by clinical guidelines for ischemia detection. Clinical relevance- Assessing individual alterations over time could eventually close the gap in ECG evaluation of patients presenting with pre-existing heart conditions and non-standard ECGs.

Despite their remarkable performance, deep neural networks remain unadopted in clinical practice, which is considered to be partially due to their lack of explainability. In this work, we apply explainable attribution methods to a pre-trained deep neural network for abnormality classification in 12-lead electrocardiography to open this "black box" and understand the relationship between model prediction and learned features. We classify data from two public databases (CPSC 2018, PTB-XL) and the attribution methods assign a "relevance score" to each sample of the classified signals. This allows analyzing what the network learned during training, for which we propose quantitative methods: average relevance scores over a) classes, b) leads, and c) average beats. The analyses of relevance scores for atrial fibrillation and left bundle branch block compared to healthy controls show that their mean values a) increase with higher classification probability and correspond to false classifications when around zero, and b) correspond to clinical recommendations regarding which lead to consider. Furthermore, c) visible P-waves and concordant T-waves result in clearly negative relevance scores in atrial fibrillation and left bundle branch block classification, respectively. Results are similar across both databases despite differences in study population and hardware. In summary, our analysis suggests that the DNN learned features similar to cardiology textbook knowledge.

Electrocardiography analysis is widely used in various clinical applications and Deep Learning models for classification tasks are currently in the focus of research. Due to their data-driven character, they bear the potential to handle signal noise efficiently, but its influence on the accuracy of these methods is still unclear. Therefore, we benchmark the influence of four types of noise on the accuracy of a Deep Learning-based method for atrial fibrillation detection in 12-lead electrocardiograms. We use a subset of a publicly available dataset (PTB-XL) and use the metadata provided by human experts regarding noise for assigning a signal quality to each electrocardiogram. Furthermore, we compute a quantitative signal-to-noise ratio for each electrocardiogram. We analyze the accuracy of the Deep Learning model with respect to both metrics and observe that the method can robustly identify atrial fibrillation, even in cases signals are labelled by human experts as being noisy on multiple leads. False positive and false negative rates are slightly worse for data being labelled as noisy. Interestingly, data annotated as showing baseline drift noise results in an accuracy very similar to data without. We conclude that the issue of processing noisy electrocardiography data can be addressed successfully by Deep Learning methods that might not need preprocessing as many conventional methods do.

Objective. We present a concept for processing 6-lead electrocardiography (ECG) signals which can be applied to various use cases in quantitative electrocardiography. Approach. Our work builds upon the mathematics of the well-known Cabrera sequence which is a re-sorting of the six limb leads (I, II, III, aV R, aV L, aV F) into a clockwise and physiologically-interpretable order. By deriving correction factors for harmonizing lead strengths and choosing an appropriate basis for the leads, we extend this concept towards what we call the 'Cabrera Circle' based on a mathematically sound foundation. Main results. To demonstrate the practical effectiveness and relevance of this concept, we analyze its suitability for deriving interpolated leads between the six limb leads and a 'radial' lead which both can be useful for specific use cases. We focus on the use cases of i) determination of the electrical heart axis by proposing a novel interactive tool for reconstructing the heart's vector loop and ii) improving accuracy in time of automatic R-wave detection and T-wave delineation in 6-lead ECG. For the first use case, we derive an equation which allows projections of the 2-dimensional vector loops to arbitrary angles of the Cabrera Circle. For the second use case, we apply several state-of-the-art algorithms to a freely-available 12-lead dataset (Lobachevsky University Database). Out-of-the-box results show that the derived radial lead outperforms the other limb leads (I, II, III, aV R, aV L, aV F) by improving F1 scores of R-peak and T-peak detection by 0.61 and 2.12, respectively. Results of on- and offset computations are also improved but on a smaller scale. Significance. In summary, the Cabrera Circle offers a methodology that might be useful for quantitative electrocardiography of the 6-lead subsystem-especially in the digital age.

Risk prediction in patients with heart failure (HF) is essential to improve the tailoring of preventive, diagnostic, and therapeutic strategies for the individual patient, and effectively use health care resources. Risk scores derived from controlled clinical studies can be used to calculate the risk of mortality and HF hospitalizations. However, these scores are poorly implemented into routine care, predominantly because their calculation requires considerable efforts in practice and necessary data often are not available in an interoperable format. In this work, we demonstrate the feasibility of a multi-site solution to derive and calculate two exemplary HF scores from clinical routine data (MAGGIC score with six continuous and eight categorical variables; Barcelona Bio-HF score with five continuous and six categorical variables). Within HiGHmed, a German Medical Informatics Initiative consortium, we implemented an interoperable solution, collecting a harmonized HF-phenotypic core data set (CDS) within the openEHR framework. Our approach minimizes the need for manual data entry by automatically retrieving data from primary systems. We show, across five participating medical centers, that the implemented structures to execute dedicated data queries, followed by harmonized data processing and score calculation, work well in practice. In summary, we demonstrated the feasibility of clinical routine data usage across multiple partner sites to compute HF risk scores. This solution can be extended to a large spectrum of applications in clinical care.

Abstract Continuous non-invasive blood pressure measurements bear a high potential. Particular in Somnology they allow to derive comfortably the systolic and diastolic blood pressure from an electrocardiogram and a synchronous photoplethysmogram without sleep disruption. In this short article some possible problems of this method are discussed along overnight recordings with a SOMNOtouch NIBP device.

AIM:The traditional hinge axis theory of temporomandibular joint (TMJ) dynamics is increasingly being replaced by the theory of instantaneous centers of rotation (ICR). Typically, ICR determinations are based on theoretical calculations or three-dimensional approximations of finite element models. MATERIALS AND METHODS:With the advent of real-time magnetic resonance imaging (MRI), natural physiologic movements of the TMJ may be visualized with 15 frames per second. The present study employs real-time MRI to analyze the TMJ biomechanics of healthy volunteers during mandibular movements, with a special emphasis on horizontal condylar inclination (HCI) and ICR pathways. The Wilcoxon rank sum test was used to comparatively analyze ICR pathways of mandibular opening and closure. RESULTS:Mean HCI was 34.8 degrees (± 11.3 degrees) and mean mandibular rotation was 26.6 degrees (± 7.2 degrees). Within a mandibular motion of 10 to 30 degrees, the resulting x- and y-translation during opening and closure of the mandible differed significantly (10 to 20 degrees, x: P = 0.02 and y: P < 0.01; 20 to 30 degrees, x: P < 0.001 and y: P = 0.01). Rotation of both 0 to 10 degrees and > 30 degrees showed no significant differences in x- and y-translation. Near occlusion movements differed only for y-translation (P < 0.01). CONCLUSION:Real-time MRI facilitates the direct recording of TMJ structures during physiologic mandibular movements. The present findings support the theory of ICR. Statistics confirmed that opening and closure of the mandible follow different ICR pathways, which might be due to muscular activity discrepancies during different movement directions. ICR pathways were similar within maximum interincisal distance (MID) and near occlusion (NO), which might be explained by limited extensibility of tissue fibers (MID) and tooth contact (NO), respectively.

Funding agencies and field experts promote reuse of scientific data and biomaterial beyond the scope of the original project. The availability of research data, however, is limited by the interest of original authors as well as the privacy rights of the study participants, especially in the biomedical sciences. On the other hand, for an available data set to be a useful contribution to the scientific community, it has to be findable and accessible with reasonable effort. Therefore, using the R Shiny library, we designed and implemented a software for data discovery and feasibility analyses with compliance to regulatory and contractual regulations. Due to its genericity, it was successfully tested with heterogeneous data sets and ultimately applied to the data and biomaterial of the German Center for Cardiovascular Research (DZHK). The resulting tool - named the Feasibility Explorer - is publicly available and can be used by researchers to get an overview of data and biomaterial available in the DZHK and to select collectives in the process of submitting a usage application. To implement the rights of participants and original authors, data is integrated by querying the informed consent and not persistently stored. All calculations on the data are performed server-sided and only aggregated information is send to a client, whereas the extent of information was strictly limited to a necessary minimum that allows an applicant to assess whether an application is worthwhile.

AIMS:Health data captured by commercially available smart devices may represent meaningful patient-reported outcome measures (PROMs) in heart failure (HF) patients. The purpose of this study was to test this hypothesis by evaluating the feasibility of a new telemonitoring concept for patients following initial HF hospitalization.METHODS AND RESULTS:We designed a cardio patient monitoring platform (CPMP) that comprised mobile iOS-based applications for patients' smartphone/smartwatch and the equivalent application on a physicians' tablet. It allowed for safe and continuous data transmission of self-measured physiological parameters, activity data, and patient-reported symptoms. In a prospective feasibility trial with 692 patient days from 10 patients hospitalized for newly diagnosed HF with reduced ejection fraction (mean left ventricular ejection fraction (LVEF) 26.5 ± 9.8%), we examined the CPMP during the first 2 months following discharge (69 ± 15 observation days per patient). The mean daily step count recorded by the mobile devices emerged as a promising new PROM. Its 14 day average increased over the study period (3612 ± 3311 steps/day at study inclusion and 7069 ± 5006 steps/day at end of study; P < 0.0001). It is unique for continuously reflecting real-life activity and correlated significantly with traditional surrogate parameters of cardiac performance including LVEF (r = 0.44; 95% CI 0.07-0.71; P = 0.0232), 6 min walk test (r = 0.67; 95% CI 0.38-0.84; P = 0.0002), and scores in health-related quality of life questionnaires.CONCLUSIONS:We provide the first patient monitoring platform for HF patients that relies on commercially available iOS/watchOS-based devices. Our study suggests it is ready for implementation as a tool for recording meaningful PROMs in future HF trials and telemonitoring.

The so-called Nebra Sky Disc' is arguably the most important archaeological find in central Europe of the last two decades. It is a richly-decorated bronze plate with gold symbols, some of them being interpreted as the Sun, the crescent-shaped Moon and 32 stars. Being dated to the end of the Early Bronze Age around 1600 bc, the Sky Disc is considered the earliest known transportable astronomical representation of the night sky.Earlier works identified at least four phases in the history of the Disc. In this work we analyse the Disc's geometric pattern of the earliest phase, considering possible archaeoastronomical relations. Using mathematical techniques, we identify several potential geometric structures: We find that subsets of three or more Disc stars may represent straight alignments, and we argue that symmetries shown by parts of the geometric pattern may be related to an eight- or 16-month calendar - possibly used as a calendrical device prior-ranking to a symbolic or decorative character.

Purpose: The purpose is to present a mathematical model of the function of the thumb carpometacarpal joint (TCMCJ) based on measurements of human joints. In the TCMCJ both articulating surfaces are saddle-shaped. The aim was to geometrically survey the shapes of the articulating surfaces using precise replicas of 28 TCMCJs. Methods: None of these 56 articulating surfaces did mathematically extend the differential geometrical neighbourhood around the main saddle point so that each surface could be characterised by three main parameters: the two extreme radii of curvature in the main saddle point and the angle between the saddles' asymptotics (straight lines). Results: The articulating surfaces, when contacting at the respective main saddle points, are incongruent. Hence, the TCMCJ has functionally five kinematical degrees of freedom (DOF); two DOF belong to flexion/extension, two to ab-/adduction. These four DOF are controlled by the muscular apparatus. The fifth DOF, axial rotation, cannot be adjusted but stabilized by the muscular apparatus so that physiologically under compressive load axial rotation does not exceed an angle of approximately +/- 3 degrees. Conclusions: The TCMCJ can be stimulated by the muscular apparatus to circumduct. The mechanisms are traced back to the curvature incongruity of the saddle surfaces. Hence we mathematically proved that none of the individual saddle surfaces can be described by a quadratic saddle surface as is often assumed in literature. We derived an algebraic formula with which the articulating surfaces in the TCMCJ can be quantitatively described. This formula can be used to shape the articulating surfaces in physiologically equivalent TCMCJ-prostheses.

Objectives To retrospectively compare two compensatory approaches taken in skeletal Class III patients during the main treatment stage, including a study group of multiband treatment plus isolated extraction of mandibular teeth and a control group of multiband treatment without extraction of teeth. Patients and methods The extraction group included 22 (12 female, 10 male) patients receiving compensatory multiband treatment for a mean of 3.47 ± 1.14 years and 16.22 ± 1.92 years old at debonding. The nonextraction group included 24 (14 female, 10 male) patients undergoing multiband treatment for 2.76 ± 1.28 years and 15.38 ± 1.46 years old at debonding. Lateral cephalograms obtained at baseline and upon completion of active treatment were traced for skeletal, dentoalveolar, and soft tissue parameters. Welch and Wilcoxon tests were used to analyze intergroup differences (initial values, final values, initial-to-final changes) and within-group differences ( p < 0.05). Results Upon completion of active treatment, the only significant intergroup differences were noted for U1NA and L1ML. Significant within-group changes over the courses of treatment were seen for SNB, MLNL, U1NA, U1NL, L1NB, L1ML, H-angle, ULipEL, and LLipEL (extraction group) or for SNB, ANB, individual ANB, Wits appraisal, U1NA, U1NL, H-angle, Naslab-a, ULipEL, and LLipEL (nonextraction group). Parameters that changed by significantly different amounts in both groups included Wits appraisal, L1NB, L1ML, and LLipEL. Conclusion The added value of isolated extraction therapy basically lies in increasing the potential for retruding the lower incisor inclinations, so that compensatory treatment becomes an option even in selected patients presenting with adverse occlusal situations that would otherwise require orthognathic surgery. Given the successful outcomes in both groups, which had been established by Peer Assessment Rating (PAR) scores, it was possible to define the skeletal, dentoalveolar, and soft tissue characteristics of successful treatment more precisely than before.

Purpose: We investigate the dependence of the kinematics of the human knee on its anatomy. The idea of describing the kinematics of the knee in the sagittal plane using four-bar linkage is almost as old as kinematics as an independent discipline. We start with a comparison of known four-bar linkage constructions. We then focus on the model by H. Nagerl which is applicable under form closure. Methods: We use geometry and analysis as the mathematical methods. The relevant geometrical parameters of the knee will be determined on the basis of the dimensions of the four-bar linkage. This leads to a system of nonlinear equations. Results: The four-bar linkage will be calculated from the limits of the constructively accessible parameters by means of a quadratic approximation. Conclusions: By adapting these requirements to the dimensions of the human knee, it will be possible to obtain valuable indications for the design of an endoprosthesis which imitates the kinematics of the natural knee.

Purpose: The curvature morphology of the articulating surfaces determines the physiological movement pattern. We quantitatively examined the curvature morphology of the tibiotalar articulating surfaces and specified their geometric contact patterns. Methods: Geometrically equivalent cartographic nets were marked on the talar and tibial articulating surfaces of true-to-scale moldings of 20 human ankle joints (intervals of 5 mm) to relate corresponding articulating units of the surfaces. The corresponding contours of the net lines were compared, and the incongruity of articulating surfaces could thus be quantified locally. Results: All tibial sagittal net lines represented circular arcs. Along the sagittal talar net lines, the curvature radii increased medially from anterior to posterior but decreased laterally. Each net line could be approximated by three circular arcs. Examining these three parts of the talar net lines, the anterior sagittal curvature radii increased from medial to lateral, whereas the posterior radii decreased. The tibial and talar transversal net lines were congruent. The articulation surfaces showed a transversal contact line in every dorsal/plantar joint position. The degree of local congruity was solely ascertained by the incongruity of the corresponding sagittal net lines. The maximal degrees of congruity were found laterally for dorsal flexion, laterally/centrally for neutral joint position, and centrally/medially for plantar flexion. Conclusions: By the transversal line contact, the contact area is broadened over the articulating surfaces from lateral to medial. In dorsal flexion, compressive loads are mainly transferred by lateral/anterior zones and in plantar flexion by medial/posterior zones of the articulating surfaces. Reconstruction of the transversal contact line is essential.

Zusammenfassung Ziel: Bei primärem Engstand im Wechselgebiss kann ausreichend Platz für die permanenten Zähne geschaffen werden, durch deren Durchbruchssteuerung mittels frühzeitiger Extraktionen von Milchzähnen und anschließender Extraktion von bleibenden Zähnen. Dieser Artikel stellt eine Orientierungsstudie zur Überprüfung der Modifikation zur klassischen Serienextraktion nach 10-jähriger klinischer Anwendung vor. Patienten und Methodik: 3 Patientengruppen mit jeweils 50 Probanden wurden gebildet und untersucht: Bei den Patienten der Gruppe 1 (G1) kamen initial die Milcheckzähne im Unterkiefer zur Extraktion. Bei der Gruppe 2 (G2) wurden initial die ersten Milchmolaren im unteren Zahnbogen extrahiert. Die dritte Gruppe (G3) fungierte als Kontrollgruppe ohne Extraktionen. Zur Kontrolle dienten OPAN-Röntgenbilder, einmal zum Zeitpunkt t1 (jeweils vor Beginn der durchgeführten Extraktionen) und zum Zeitpunkt t2 (durchschnittlichen 2,5 Jahre post extractionem). Für die Kontrollgruppe ohne Extraktionen wurden zeitlich äquivalente OPAN-Aufnahmen zur Beurteilung verwendet. Ausgewertet wurden die relativen Lageveränderungen der permanenten Eckzahn- und Prämolarenkeime. Für die sich anschließende Orientierungsstudie zur Prüfung einer Behandlungsoptimierung und deren statistische Eruierung der Anwendungshäufigkeit bezüglich der vorgenommenen Extraktionsmodifikation, kamen 63 Probanden mit der Indikation zur Extraktion von ersten Prämolaren bei primärem Engstand zur Auswertung. Ergebnisse: In über 90% der Extraktionsfälle wurde die modifizierte Serienextraktion angewandt. Die Vorteile, wie die Umkehrung der natürlichen Durchbruchssequenz von Eckzahn und Prämolaren im Unterkiefer als Folge des geänderten Extraktionsvorgehens, wirkten sich positiv auf die Eruption des ersten Prämolaren aus. Schlussfolgerungen: Die Modifikation der Serienextraktion im Unterkiefer erster Milchmolaren statt Milcheckzähnen führt zu einer Änderung des Eruptionsverhaltens des bleibenden Eckzahnes und des ersten Prämolaren. Das Lückenmanagement im Extraktionsfall ist klinisch besser zu handhaben und die Dauer der Multibandbehandlungszeit verkürzt sich signifikant. Die vorgestellte Modifikation der Serienextraktion kann als Alternative der klassischen Serienextraktion vorgezogen werden.

Purpose: We present a novel method to process kinematical data typically coming from measurements of joints. This method will be illustrated through two examples. Methods: We adopt theoretical kinematics together with the principle of least action. We use motion and inverse motion for describing the whole experimental situation theoretically. Results: By using the principle of least action, the data contain information about inherent reference points, which we call characteristic points. These points are unique for direct and inverse motion. They may be viewed as centers of the fixed and moving reference systems. The respective actions of these characteristic points are analytically calculated. The sum of these actions defines the kinematical action. This sum is by design independent of the choice of reference system. The minimality of the kinematical action can be used again to select numerically one representative cycle in empirically given, approximately periodic motions. Finally, we illustrate the theoretical approach making use of two examples worked out, hinge movement and the sagittal component of the movement of a human leg during gait. Conclusions: This approach enables automatic cycle choices for evaluating large databases in order to compare and to distinguish empirically given movements. The procedure can be extended to three dimensional movements.