Article
Biochemistry & Molecular Biology
Fabian Horst, Djordje Slijepcevic, Marvin Simak, Brian Horsak, Wolfgang Immanuel Schoellhorn, Matthias Zeppelzauer
Summary: This work utilizes machine learning to model individual gait signatures and identify factors contributing to inter-individual variability in gait patterns. The study demonstrates the uniqueness of gait signatures and highlights the distinctive gait characteristics of individuals. The proposed approach provides valuable insights into understanding biological individuality and has potential applications in healthcare and clinical diagnosis.
COMPUTATIONAL AND STRUCTURAL BIOTECHNOLOGY JOURNAL
(2023)
Article
Computer Science, Artificial Intelligence
Valerio La Gatta, Vincenzo Moscato, Marco Postiglione, Giancarlo Sperli
Summary: In this paper, a novel model-agnostic Explainable AI technique named CASTLE is proposed to provide rule-based explanations based on both the local and global model's workings. The framework has been evaluated on six datasets in terms of temporal efficiency, cluster quality and model significance, showing a 6% increase in interpretability compared to another state-of-the-art technique, Anchors.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Carlo Combi, Beatrice Amico, Riccardo Bellazzi, Andreas Holzinger, Jason H. Moore, Marinka Zitnik, John H. Holmes
Summary: This paper focuses on the importance of explainable artificial intelligence (XAI) in the field of biomedicine. By bringing together researchers with different roles and perspectives, it explores XAI in depth and presents a series of requirements for achieving explainability in AI.
ARTIFICIAL INTELLIGENCE IN MEDICINE
(2022)
Article
Computer Science, Artificial Intelligence
Leander Weber, Sebastian Lapuschkin, Alexander Binder, Wojciech Samek
Summary: Explainable Artificial Intelligence (XAI) is a research field that aims to bring transparency to complex and opaque machine learning models. This paper provides an overview of techniques that practically apply XAI to improve ML models, categorizing and comparing their strengths and weaknesses. Theoretical perspectives and empirical experiments demonstrate how explanations can enhance properties such as model generalization and reasoning. The potential caveats and drawbacks of these methods are also discussed.
INFORMATION FUSION
(2023)
Article
Chemistry, Analytical
Marco Iosa, Maria Grazia Benedetti, Gabriella Antonucci, Stefano Paolucci, Giovanni Morone
Summary: Many recent studies have shown that the harmony of physiological walking depends on the proportions between the phases of the gait cycle. The gait cycle assumes a fractal structure when the proportion is close to the golden ratio. In stroke patients, this harmony is disrupted, and it is unclear which factor is associated with the ratios between gait phases. Using an artificial neural network, researchers found that the gait ratio is associated with walking speed, stride length, and hip muscle forces.
Article
Engineering, Biomedical
Isaly Tappan, Erica M. Lindbeck, Jennifer A. Nichols, Joel B. Harley
Summary: This paper discusses the use of explainable AI (XAI) to enhance the interpretability of biomechanics data, and demonstrates its effectiveness in a simulation-based classification task.
ANNALS OF BIOMEDICAL ENGINEERING
(2023)
Article
Automation & Control Systems
Alberto Barbado, Oscar Corcho
Summary: This study combines unsupervised anomaly detection techniques, domain knowledge, and interpretable machine learning models to explain abnormal fuel consumption in vehicle fleets. Results evaluated on real-world data show that this approach provides recommendations for fuel optimization adjusted to different user profiles.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2022)
Article
Engineering, Industrial
B. Hu, S. Li, Y. Chen, R. Kavi, S. Coppola
Summary: This study demonstrates that deep learning models utilizing wearable sensors can accurately identify irregular walking surfaces, potentially preventing falling injuries.
APPLIED ERGONOMICS
(2021)
Article
Computer Science, Information Systems
Angel Delgado-Panadero, Beatriz Hernandez-Lorca, Maria Teresa Garcia-Ordas, Jose Alberto Benitez-Andrades
Summary: This paper proposes a feature contribution method for GBDT, which can calculate the contribution of each feature to predictions. The method not only serves as a local explainability model for GBDT, but also reflects its internal behavior. It is significant for ethical analysis of AI and compliance with relevant regulations.
INFORMATION SCIENCES
(2022)
Article
Business, Finance
Ying Zhou, Haoran Li, Zhi Xiao, Jing Qiu
Summary: This paper aims to produce user-centered explanations for financial fraud detection models using Explainable Artificial Intelligence (XAI) methods. By combining an ensemble predictive model with a explainable framework based on Shapley values, an accurate and explainable financial fraud detection approach is developed. The results show that the explainable framework can meet the requirements of different external stakeholders by providing local and global explanations.
FINANCE RESEARCH LETTERS
(2023)
Article
Orthopedics
Pouria Rouzrokh, Taghi Ramazanian, Cody C. Wyles, Kenneth A. Philbrick, Jason C. Cai, Michael J. Taunton, Hilal Maradit Kremers, David G. Lewallen, Bradley J. Erickson
Summary: This study demonstrates the potential of a convolutional neural network model in assessing the risk of hip dislocation based on postoperative radiographs. The model achieved high sensitivity and negative predictive value, suggesting it can be utilized in combination with clinical risk factors for rapid risk assessment in THA patients.
JOURNAL OF ARTHROPLASTY
(2021)
Article
Biotechnology & Applied Microbiology
Wei-Chun Lee, Tsan-Yang Chen, Li-Wei Hung, Ting-Ming Wang, Chia-Hsieh Chang, Tung-Wu Lu
Summary: Long-term follow-up studies have shown that surgically treated DDH children may develop premature osteoarthritis on both affected and unaffected hips. Gait analysis revealed higher loading rates on both sides of the hip, which were correlated to acetabular index and peak unloading rates. Regular assessment of hip morphology and loading rates is crucial for early identification of hip dysplasia and cartilage degeneration risk.
FRONTIERS IN BIOENGINEERING AND BIOTECHNOLOGY
(2021)
Article
Multidisciplinary Sciences
Y. Palmowski, S. Popovic, D. Kosack, P. Damm
Summary: Footwear choice post joint replacement surgery plays a crucial role in affecting hip joint loads during physiotherapy. This study examined the impact of different types of shoes on in vivo hip joint loads, finding that shoes, especially those with stiff soles or elaborate cushioning elements, tend to increase hip joint loading during walking. Preferably, low-profile shoes with a flexible sole may be more suitable for patients aiming for reduced hip joint loads, especially during postoperative physiotherapy or to alleviate osteoarthritis symptoms.
SCIENTIFIC REPORTS
(2021)
Article
Chemistry, Multidisciplinary
Giordano Valente, Fulvia Taddei, Alberto Leardini, Maria Grazia Benedetti
Summary: This study investigated differences in gait function and musculoskeletal loads during walking between hip dysplasia patients who received hip abductor strengthening after total hip replacement and those who only performed standard rehabilitation. The results showed that patients who underwent muscle strengthening exhibited a more physiological force pattern and greater force in the operated limb, although this was statistically significant only in certain portions of the gait cycle and potentially related to higher gait speed. Further studies on hip dysplasia patients would be beneficial in understanding the effects of muscle strengthening on loads.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Artificial Intelligence
Marco Virgolin, Saverio Fracaros
Summary: Counterfactual explanations (CEs) provide insights into changing algorithmic decisions. This paper examines the relationship between robustness and sparsity of CEs, and introduces definitions of robustness for sparse CEs. Experimental results show that robust CEs are more cost-effective and preferable for users.
ARTIFICIAL INTELLIGENCE
(2023)
Article
Computer Science, Interdisciplinary Applications
Carlo Dindorf, Jurgen Konradi, Claudia Wolf, Bertram Taetz, Gabriele Bleser, Janine Huthwelker, Friederike Werthmann, Philipp Drees, Michael Froehlich, Ulrich Betz
Summary: This study examines the potential of identifying individuals based on dynamic spinal data, suggesting the presence of a personal spinal 'fingerprint'. High accuracies were achieved, indicating the feasibility of subject recognition and potential clinical applications.
COMPUTER METHODS IN BIOMECHANICS AND BIOMEDICAL ENGINEERING
(2022)
Article
Automation & Control Systems
Min Wu, Bertram Taetz, Yanhao He, Gabriele Bleser, Steven Liu
Summary: This article proposes a framework based on Dynamic Movement Primitives (DMP) that enables robots to learn from human demonstrations and perform handovers with humans. The framework extends the conventional DMP formalism by incorporating uncertainty-aware learning, weighting functions, orientation-based scaling, and online parameter adaption. Experimental validation and evaluation show enhanced success rate, fluency, and human comfort.
ROBOTICS AND AUTONOMOUS SYSTEMS
(2022)
Article
Sport Sciences
Carlo Dindorf, Eva Bartaguiz, Elena Janowicz, Michael Froehlich, Oliver Ludwig
Summary: This study aimed to assess the effects of asymmetric muscle fatigue on the skin surface temperature of abdominal and back muscles. Skin surface temperatures were recorded using infrared thermography, and muscle soreness and fatigue were evaluated using questionnaires. The results showed that skin temperature was significantly lower in the post-tests compared to the pre- and follow-up tests, and there were asymmetric differences in the upper and lower areas of the back and front.
Article
Environmental Sciences
Joshua Berger, Michael Froehlich, Wolfgang Kemmler
Summary: Whole-body electromyostimulation (WB-EMS) is a time-efficient and effective training method, but misuse and lack of supervision pose risks. Expert guidelines and legal restrictions ensure safe and effective use, but the growing private market undermines these standards, potentially endangering trainees and the overall market.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2022)
Review
Environmental Sciences
Carlo Dindorf, Eva Bartaguiz, Freya Gassmann, Michael Froehlich
Summary: Artificial intelligence and its subcategories of machine learning and deep learning are increasingly important in sports research. However, the large number of corresponding publications in this field is difficult to manage. This study analyzed 1215 documents and found that research interest in AI in sports is growing exponentially, with the top 20 most cited works accounting for 32.52% of total citations. The top 10 journals are responsible for 28.64% of all published documents, and China, the USA, and Germany are the most productive countries.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2023)
Article
Sport Sciences
Carlo Dindorf, Eva Bartaguiz, Jonas Dully, Max Sprenger, Stephan Becker, Michael Froehlich, Oliver Ludwig
Summary: This study explores the objectification of acute and cumulative fatigue in sport climbing. The results suggest that consecutive maximal isometric holding induces cumulative fatigue. However, the reduction in climbing-specific holding time did not correlate with a reduction in handgrip strength. Monitoring athletes' termination SmO2 saturation seems promising for measuring acute fatigue. On the other hand, measuring handgrip strength and muscle oxygen metabolism is not suitable for monitoring cumulative fatigue during intermittent isometric climbing-specific muscle contractions.
Article
Biotechnology & Applied Microbiology
Carlo Dindorf, Oliver Ludwig, Steven Simon, Stephan Becker, Michael Froehlich
Summary: The study proposes a data-driven ML system for medical decision support using machine learning and explainable artificial intelligence tools. The system provides objective posture assessment and interpretation. Results show that the system performs well in diagnosing and correcting postural problems and could be applied in personalized medicine and the development of posture assessment apps for prevention.
BIOENGINEERING-BASEL
(2023)
Article
Sport Sciences
Stephan Becker, Steven Simon, Jan Muehlen, Carlo Dindorf, Michael Froehlich
Summary: This pilot study suggests that sensorimotor insoles may be helpful in reducing subjective pain. However, caution should be exercised in interpreting these results due to the lack of a control group and confounding variables. Further randomized controlled trials and systematic reviews are needed to validate these findings.
JOURNAL OF FUNCTIONAL MORPHOLOGY AND KINESIOLOGY
(2023)
Article
Sport Sciences
Stephan Becker, Joshua Berger, Oliver Ludwig, Daniel Gunther, Jens Kelm, Michael Frohlich
Summary: This study investigated the cumulative effect of purposeful heading and the relationship between head-neck-torso alignment and head acceleration. While no relationship was found for head acceleration among standing, jumping, and running headers, a significant relationship was identified between head acceleration and maximum ball speed specifically for standing headers. Additional research is needed to further explore these relationships and confirm the findings.
JOURNAL OF HUMAN KINETICS
(2021)