Machine Learning Model for Readmission Prediction of Patients With Heart Failure Based on Electronic Health Records: Protocol for a Quasi-Experimental Study for Impact Assessment.

BACKGROUND: Care for patients with heart failure (HF) causes a substantial load on health care systems where a prominent challenge is the elevated rate of readmissions within 30 days following initial discharge. Clinical professionals face high levels of uncertainty and subjectivity in the decision-making process on the optimal timing of discharge. Unwanted hospital stays generate costs and cause stress to patients and potentially have an impact on care outcomes. Recent studies have aimed to mitigate the uncertainty by developing and testing risk assessment tools and predictive models to identify patients at risk of readmission, often using novel methods such as machine learning (ML). OBJECTIVE: This study aims to investigate how a developed clinical decision support (CDS) tool alters the decision-making processes of health care professionals in the specific context of discharging patients with HF, and if so, in which ways. Additionally, the aim is to capture the experiences of health care practitioners as they engage with the system's outputs to analyze usability aspects and obtain insights related to future implementation. METHODS: A quasi-experimental design with randomized crossover assessment will be conducted with health care professionals on HF patients' scenarios in a region located in the South of Sweden. In total, 12 physicians and nurses will be randomized into control and test groups. The groups shall be provided with 20 scenarios of purposefully sampled patients. The clinicians will be asked to take decisions on the next action regarding a patient. The test group will be provided with the 10 scenarios containing patient data from electronic health records and an outcome from an ML-based CDS model on the risk level for readmission of the same patients. The control group will have 10 other scenarios without the CDS model output and containing only the patients' data from electronic medical records. The groups will switch roles for the next 10 scenarios. This study will collect data through interviews and observations. The key outcome measures are decision consistency, decision quality, work efficiency, perceived benefits of using the CDS model, reliability, validity, and confidence in the CDS model outcome, integrability in the routine workflow, ease of use, and intention to use. This study will be carried out in collaboration with Cambio Healthcare Systems. RESULTS: The project is part of the Center for Applied Intelligent Systems Research Health research profile, funded by the Knowledge Foundation (2021-2028). Ethical approval for this study was granted by the Swedish ethical review authority (2022-07287-02). The recruitment process of the clinicians and the patient scenario selection will start in September 2023 and last till March 2024. CONCLUSIONS: This study protocol will contribute to the development of future formative evaluation studies to test ML models with clinical professionals. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): PRR1-10.2196/52744.

Barriers and Enablers for Implementation of an Artificial Intelligence-Based Decision Support Tool to Reduce the Risk of Readmission of Patients With Heart Failure: Stakeholder Interviews.

BACKGROUND: Artificial intelligence (AI) applications in health care are expected to provide value for health care organizations, professionals, and patients. However, the implementation of such systems should be carefully planned and organized in order to ensure quality, safety, and acceptance. The gathered view of different stakeholders is a great source of information to understand the barriers and enablers for implementation in a specific context. OBJECTIVE: This study aimed to understand the context and stakeholder perspectives related to the future implementation of a clinical decision support system for predicting readmissions of patients with heart failure. The study was part of a larger project involving model development, interface design, and implementation planning of the system. METHODS: Interviews were held with 12 stakeholders from the regional and municipal health care organizations to gather their views on the potential effects implementation of such a decision support system could have as well as barriers and enablers for implementation. Data were analyzed based on the categories defined in the nonadoption, abandonment, scale-up, spread, sustainability (NASSS) framework. RESULTS: Stakeholders had in general a positive attitude and curiosity toward AI-based decision support systems, and mentioned several barriers and enablers based on the experiences of previous implementations of information technology systems. Central aspects to consider for the proposed clinical decision support system were design aspects, access to information throughout the care process, and integration into the clinical workflow. The implementation of such a system could lead to a number of effects related to both clinical outcomes as well as resource allocation, which are all important to address in the planning of implementation. Stakeholders saw, however, value in several aspects of implementing such system, emphasizing the increased quality of life for those patients who can avoid being hospitalized. CONCLUSIONS: Several ideas were put forward on how the proposed AI system would potentially affect and provide value for patients, professionals, and the organization, and implementation aspects were important parts of that. A successful system can help clinicians to prioritize the need for different types of treatments but also be used for planning purposes within the hospital. However, the system needs not only technological and clinical precision but also a carefully planned implementation process. Such a process should take into consideration the aspects related to all the categories in the NASSS framework. This study further highlighted the importance to study stakeholder needs early in the process of development, design, and implementation of decision support systems, as the data revealed new information on the potential use of the system and the placement of the application in the care process.

Comparison of lightweight and traditional figure skating blades, a prototype blade with integrated damping system and a running shoe in simulated figure skating landings and vertical countermovement jumps, and evaluation of dampening properties of the prototype blade.

To date, there is no empirical evidence suggesting greater jump heights or cushioned landings when using figure skating (FS) blades of different mass and design. This study examined the effect of lightweight (Gold Seal Revolution from John Wilson) and traditional (Apex Supreme from Jackson Ultima and Volant from Riedell) blades, a new prototype blade with an integrated damping system (damping blade) in two different damping configurations, and running shoes (Runfalcon from Adidas) on kinetics and kinematics during simulated on-ice landings from 0.6 m and maximal countermovement jumps on synthetic ice, and measured dampening properties of the damping blade. Seventeen participants executed trials in the six footwear conditions blinded to the different blades and acted as their own control for statistical comparison. There were no differences between the lightweight and traditional blades on the maximal vertical ground reaction force during the landing. Image analysis showed a damping effect in the damping blade that significantly decreased the landing load for all participants (mean 4.38 ± 0.68 bodyweight) (p ≤ 0.006), on average between 10.1 and 14.3% compared to lightweight and traditional blades (4.87 ± 1.01 to 5.11 ± 0.88 bodyweight). The maximal jump height achieved was the same in all FS blades.

Pedal to the Metal: Velocity and Power in High-Level Golfers.

ABSTRACT: Parker, J, and Lundgren, LE. Pedal to the metal: Velocity and power in high-level golfers. J Strength Cond Res 35(12): 3425-3431, 2021-In most rotational power assessments, discrete variables are used for subsequent examination; however, movements are continuous, and data can be collected in time series. The purpose of this investigation was to examine the velocity- and power-time series characteristics of a standing rotation test and identify relationships with golf performance. Thirty-one golfers performed a golf-specific rotation test (GSRT) with 3 different resistances (6, 10, and 14 kg) in a robotic engine system. Time series of velocity and power was calculated from the raw data, and each repetition was then normalized to 0-100%. Principal component analyses (PCAs) were performed on velocity and power waveforms. The PCA used an eigenvalue analysis of the data covariance matrix. The relationship between clubhead speed (CHS) and all principal components (PC) was examined using linear regression. Ten velocity parameters and 6 power parameters explained 80% of the variance in the data. For velocity, the first 2 PCs identified both magnitude and phase shift features while PCs 3-5 identified difference features. For power, the first 2 PCs identified both magnitude and phase shift features, the third PC identified a phase shift feature, and the fourth PC identified a difference feature. The highest relationship with CHS was shown for GSRT with 14 kg in PC2 for power (R2 = 0.48, p < 0.001). The PCA of the GSRT power test could distinguish intraindividual differences, external loads, and sex-based differences. Athletes should focus on accelerating smoothly through the movement, particularly with heavier loads, and not pulling aggressively at the beginning of the rotational movement to achieve maximum power.

A Novel Method for Classification of Running Fatigue Using Change-Point Segmentation.

Blood lactate accumulation is a crucial fatigue indicator during sports training. Previous studies have predicted cycling fatigue using surface-electromyography (sEMG) to non-invasively estimate lactate concentration in blood. This study used sEMG to predict muscle fatigue while running and proposes a novel method for the automatic classification of running fatigue based on sEMG. Data were acquired from 12 runners during an incremental treadmill running-test using sEMG sensors placed on the vastus-lateralis, vastus-medialis, biceps-femoris, semitendinosus, and gastrocnemius muscles of the right and left legs. Blood lactate samples of each runner were collected every two minutes during the test. A change-point segmentation algorithm labeled each sample with a class of fatigue level as (1) aerobic, (2) anaerobic, or (3) recovery. Three separate random forest models were trained to classify fatigue using 36 frequency, 51 time-domain, and 36 time-event sEMG features. The models were optimized using a forward sequential feature elimination algorithm. Results showed that the random forest trained using distributive power frequency of the sEMG signal of the vastus-lateralis muscle alone could classify fatigue with high accuracy. Importantly for this feature, group-mean ranks were significantly different (p < 0.01) between fatigue classes. Findings support using this model for monitoring fatigue levels during running.

Surfing the Waves of the CMJ; Are There between-Sport Differences in the Waveform Data?

The ability to analyse countermovement jump (CMJ) waveform data using statistical methods, like principal component analysis, can provide additional information regarding the different phases of the CMJ, compared to jump height or peak power alone. The aim of this study was to investigate the between-sport force-time curve differences in the CMJ. Eighteen high level golfers (male = 10, female = 8) and eighteen high level surfers (male = 10, female = 8) performed three separate countermovement jumps on a force platform. Time series of data from the force platform was normalized to body weight and each repetition was then normalized to 0⁻100 percent. Principal component analyses (PCA) were performed on force waveforms and the first six PCs explained 35% of the variance in force parameters. The main features of the movement cycles were characterized by magnitude (PC1 and PC5), waveform (PC2 and PC4), and phase shift features (PC3). Surf athletes differ in their CMJ technique and show a greater negative centre of mass displacement when compared to golfers (PC1), although these differences are not necessarily associated with greater jump height. Principal component 5 demonstrated the largest correlation with jump height (R² = 0.52). Further studies are recommended in this area, to reveal which features of the CMJ that relate to jumping performance, and sport specific adaptations.

Workloads of Competitive Surfing: Work-to-Relief Ratios, Surf-Break Demands, and Updated Analysis.

Farley, ORL, Secomb, JL, Raymond, ER, Lundgren, LE, Ferrier, BK, Abbiss, CR, and Sheppard, JM. Workloads of competitive surfing: work-to-relief ratios, surf-break demands, and updated analysis. J Strength Cond Res 32(10): 2939-2948, 2018-The study provides an in-depth descriptive and quantitative time-motion analysis of competitive surfing, using Global Positioning System (GPS) units and video synchronization, which serves to extend upon the results of Farley, Harris, and Kilding (Journal of Strength and Conditioning Research, 26, 7 [2012]). In addition, comparisons between locations and surfers competing in the same heats were performed. Global Positioning System and video data were collected from 41 male competitive surfers (23.2 ± 6.1 years, 71 ± 10.3 kg, 177.2 ± 6.4 cm) participating in 3 professional domestic surfing events, with competitive heats of 20-minute duration. Fifty data sets were analyzed across the 3 competitions, with velocities and distances covered, proportion of time spent performing various surfing activities, and total work-to-relief ratio determined. Results revealed surfers paddled 44% of the total time, followed by stationary periods (42%). Surfers performed at a significantly (p ≤ 0.05) higher work-to-relief ratio (1.7:1) at the beach-break (an exposed beach) compared with point-break 1 and 2 (waves breaking around a rocky point). Point-breaks 1 and 2 had longer continuous durations of paddling, with significantly longer rides at point-break 1 over the beach-break (p ≤ 0.01) and point-break 2 (p ≤ 0.01). The average maximal speed (24.8 km·h) from point-break 2 was significantly faster than point-break 1 (p ≤ 0.01) and beach-break (p ≤ 0.05). This information should influence surfing drills and conditioning methods to prepare these athletes for the disparate demands, such as training for a point-break competition involving longer durations of continuous paddling and short, high-intensity workloads for a beach-break.

Maximal Strength Training Improves Surfboard Sprint and Endurance Paddling Performance in Competitive and Recreational Surfers.

Coyne, JOC, Tran, TT, Secomb, JL, Lundgren, LE, Farley, ORL, Newton, RU, and Sheppard, JM. Maximal strength training improves surfboard sprint and endurance paddling performance in competitive and recreational surfers. J Strength Cond Res 31(1): 244-253, 2017-Upper-body (UB) strength has very high correlations with faster surfboard paddling speeds. However, there is no research examining the effects of improving UB strength has on surfboard paddling ability. This study aimed to determine the influence that improvements in UB closed-kinetic chain maximal strength have on surfboard paddling in both competitive and recreational surfers. Seventeen competitive and recreational male surfers (29.7 ± 7.7 years, 177.4 ± 7.4 cm, 76.7 ± 9.9 kg) participated in a repeated-measures, parallel control study design. Anthropometry; 5-, 10-, and 15-m sprint; and 400-m endurance surfboard paddling tests along with pull-up and dip 1 repetition maximum strength tests were assessed pre- and postintervention. Subjects in the training group performed 5 weeks of maximal strength training in the pull-up and dip. Differences between the training and control groups were examined postintervention. The training group increased their speed over the 5-, 10-, and 15-m sprint, whereas the control group became slower (d = 0.71, 0.51, and 0.4, respectively). The training group also displayed faster endurance paddling performance compared with the control group (d = 0.72). Short-term exposure to maximal strength training elicits improvements in paddling performance measures. However, the magnitude of performance increases seems to be dependent on initial strength levels with differential responses between strong and weaker athletes. Although a longer maximal strength training period may have produced more significant paddling improvements in stronger subjects, practitioners are unlikely to have any more than 5 weeks in an uninterrupted block with competitive surfing athletes. This study reveals that a "threshold" level of maximal strength that if possessed, short-term maximal strength training may only provide little improvement in paddling performance.

Five Weeks of Sprint and High-Intensity Interval Training Improves Paddling Performance in Adolescent Surfers.

Farley, ORL, Secomb, JL, Parsonage, JR, Lundgren, LE, Abbiss, CR, and Sheppard, JM. Five weeks of sprint and high-intensity interval training improves paddling performance in adolescent surfers. J Strength Cond Res 30(9): 2446-2452, 2016-The purpose of our study was to examine the effects of sprint interval training (SIT; 10 seconds) and high-intensity interval training (HIT; 30 seconds) on surfing athletes paddling performance (400-m time trial and repeat-sprint paddle performance). Twenty-four competitive adolescent surfers (19 male, 5 female; age = 14.4 ± 1.3 years, mass: 50.1 ± 10.7 kg, and stature: 159.9 ± 10.3 cm) were assigned to perform either 5 weeks of SIT and HIT. Participants completed a repeated-sprint paddle ability test (RSPT, 15-m surfboard sprint paddle initiated every 40 seconds × 10 bouts) and 400-m endurance surfboard paddle time trial before and after training. High-intensity interval training decreased the total time to complete the 400 m by 15.8 ± 16.1 seconds (p = 0.03), and SIT decreased the total time to complete the RSPT by 6.5 ± 4.3 seconds (p = 0.02). Fatigue index during the RSPT (first-slowest effort) was lower after HIT and SIT (p ≤ 0.001 and p = 0.02, respectively). There were no significant differences in performance changes in the 400 m (total time) and RSPT (total time, fastest 15 m time, and peak velocity) between HIT and SIT. Our study indicates that HIT and SIT may be implemented to the training program of surfers to improve aerobic and repeat-sprint paddle ability, both of which are identified as key aspects of the sport. In addition, these findings indicate that 400-m paddle and RSPT can discriminate between aerobic and anaerobic training adaptations, with aerobic gains likely from HIT and anaerobic gains from SIT.

Comparison of impact forces, accelerations and ankle range of motion in surfing-related landing tasks.

This study aimed to describe the impact forces, accelerations and ankle range of motion in five different landing tasks that are used in training and testing for competitive surfing athletes, to assist coaches in the prescription of landing task progression and monitoring training load. Eleven competitive surfing athletes aged 24 ± 7 years participated, and inertial motion sensors were fixed to the anterior aspect of the feet, mid-tibial shafts, sacrum and eighth thoracic vertebrae on these athletes. Three tasks were performed landing on force plates and two tasks in a modified gymnastics set-up used for land-based aerial training. Peak landing force, resultant peak acceleration and front and rear side ankle dorsiflexion ranges of motion during landing were determined. The peak acceleration was approximately 50% higher when performing aerial training using a mini-trampoline and landing on a soft-density foam board, compared to a similar landing off a 50 cm box. Furthermore, the ankle ranges of motion during the gymnastic type landings were significantly lower than the other landing types (P ≤ 0.05 and P ≤ 0.001), for front and rear sides, respectively. Conclusively, increased task complexity and specificity of the sport increased the tibial peak acceleration, indicating greater training load.

Relationships Between Lower-Body Muscle Structure and, Lower-Body Strength, Explosiveness and Eccentric Leg Stiffness in Adolescent Athletes.

The purpose of the present study was to determine whether any relationships were present between lower-body muscle structure and, lower-body strength, variables measured during a countermovement jump (CMJ) and squat jump (SJ), and eccentric leg stiffness, in adolescent athletes. Thirty junior male (n = 23) and female (n = 7) surfing athletes (14.8 ± 1.7 y; 1.63 ± 0.09 m; 54.8 ± 12.1 kg) undertook lower-body muscle structure assessment with ultrasonography and performed a; CMJ, SJ and an isometric mid-thigh pull (IMTP). In addition, eccentric leg stiffness was calculated from variables of the CMJ and IMTP. Moderate to very large relationships (r = 0.46-0.73) were identified between the thickness of the vastus lateralis (VL) and lateral gastrocnemius (LG) muscles, and VL pennation angle and; peak force (PF) in the CMJ, SJ and IMTP. Additionally, moderate to large relationships (r = 0.37-0.59) were found between eccentric leg stiffness and; VL and LG thickness, VL pennation angle, and LG fascicle length, with a large relationship (r = 0.59) also present with IMTP PF. These results suggest that greater thickness of the VL and LG were related to improved maximal dynamic and isometric strength, likely due to increased hypertrophy of the extensor muscles. Furthermore, this increased thickness was related to greater eccentric leg stiffness, as the associated enhanced lower-body strength likely allowed for greater neuromuscular activation, and hence less compliance, during a stretch-shortening cycle. Key pointsGreater thickness of the VL and LG muscles were significantly related to an enhanced ability to express higher levels of isometric and dynamic strength, and explosiveness in adolescent athletes.Isometric strength underpinned performance in the CMJ and SJ in these athletes.Greater lower-body isometric strength was significantly related to eccentric leg stiffness, which is potentially the result of greater neuromuscular activation in the muscle-tendon unit.

Development and Evaluation of a Simple, Multifactorial Model Based on Landing Performance to Indicate Injury Risk in Surfing Athletes.

PURPOSE: To develop and evaluate a multifactorial model based on landing performance to estimate injury risk for surfing athletes. METHODS: Five measures were collected from 78 competitive surfing athletes and used to create a model to serve as a screening tool for landing tasks and potential injury risk. In the second part of the study, the model was evaluated using junior surfing athletes (n = 32) with a longitudinal follow-up of their injuries over 26 wk. Two models were compared based on the collected data, and magnitude-based inferences were applied to determine the likelihood of differences between injured and noninjured groups. RESULTS: The study resulted in a model based on 5 measures--ankle-dorsiflexion range of motion, isometric midthigh-pull lower-body strength, time to stabilization during a drop-and-stick (DS) landing, relative peak force during a DS landing, and frontal-plane DS-landing video analysis--for male and female professional surfers and male and female junior surfers. Evaluation of the model showed that a scaled probability score was more likely to detect injuries in junior surfing athletes and reported a correlation of r = .66, P = .001, with a model of equal variable importance. The injured (n = 7) surfers had a lower probability score (0.18 ± 0.16) than the noninjured group (n = 25, 0.36 ± 0.15), with 98% likelihood, Cohen d = 1.04. CONCLUSIONS: The proposed model seems sensitive and easy to implement and interpret. Further research is recommended to show full validity for potential adaptations for other sports.

Relationships Between Lower-Body Muscle Structure and Lower-Body Strength, Power, and Muscle-Tendon Complex Stiffness.

The purpose of this study was to determine whether any relationships were present between lower-body muscle structure and strength and power qualities. Fifteen elite male surfing athletes performed a battery of lower-body strength and power tests, including countermovement jump (CMJ), squat jump (SJ), isometric midthigh pull (IMTP), and had their lower-body muscle structure assessed with ultrasonography. In addition, lower-body muscle-tendon complex (MTC) stiffness and dynamic strength deficit (DSD) ratio were calculated from the CMJ and IMTP. Significant relationships of large to very large strength were observed between the vastus lateralis (VL) thickness of the left (LVL) and right (RVL) leg and peak force (PF) (r = 0.54-0.77, p < 0.01-0.04), peak velocity (PV) (r = 0.66-0.83, p < 0.01), and peak jump height (r = 0.62-0.80, p < 0.01) in the CMJ and SJ, as well as IMTP PF (r = 0.53-0.60, p = 0.02-0.04). Furthermore, large relationships were found between left lateral gastrocnemius (LG) pennation angle and SJ and IMTP PF (r = 0.53, p = 0.04, and r = 0.70, p < 0.01, respectively) and between LG and IMTP relative PF (r = 0.63, p = 0.01). Additionally, large relationships were identified between lower-body MTC stiffness and DSD ratio (r = 0.68, p < 0.01), right (LG) pennation angle (r = 0.51, p = 0.05), CMJ PF (r = 0.60, p = 0.02), and jump height (r = 0.53, p = 0.04). These results indicate that greater VL thickness and increased LG pennation angle are related to improved performance in the CMJ, SJ, and IMTP. Furthermore, these results suggest that lower-body MTC stiffness explains a large amount of variance in determining an athlete's ability to rapidly apply force during a dynamic movement.