Baixiang Zhao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Baixiang Zhao has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Andreas Fontalis
Department of Trauma and Orthopaedic Surgery, University College Hospital, London, UK
1 shared publication - Evangelos B. Mazomenos
University College London
1 shared publication - Fabio Mancino
Sydney Orthopaedic Research Institute (SORI) The Knee Institute, Landmark Orthopaedics St. Leonards New South Wales Australia
1 shared publication - Fabrice Glod
Hôpitaux Robert Schuman, Luxembourg City, Luxembourg
1 shared publication - Fares Sami Haddad
Department of Trauma and Orthopaedic Surgery, University College Hospital, London, UK
1 shared publication Jörn Mehnen
University of Strathclyde, Department of Design, Manufacturing and Engineering Management, G1 1XQ, United Kingdom
1 shared publication
Latest Publications
- Open Access
- 27 August 2024
- Published by EDP Sciences in MATEC Web of Conferences
- Vol. 401, 12002
A novel non-intrusive mental workload evaluation concept in human-robot collaboration
The integration of Human-Robot Collaboration (HRC) in industrial robotics introduces challenges, particularly in adapting manufacturing environments to work seamlessly with collaborative robots. A key objective in HRC system optimization is enhancing human acceptance of these robots and improving productivity. Traditionally, the assessment of human mental workload in these settings relies on methods like EEG, fNIRS, and heart rate monitoring, which require direct physical contact and can be impractical in manufacturing environments. To address these issues, we propose an innovative and non-intrusive method that employs cameras to measure mental workload. This technique involves capturing video footage of human operators on the shop floor, focusing specifically on facial expressions. Advanced AI algorithms analyse these videos to predict heart rate ranges, which are then used to estimate mental workload levels in real time. This approach not only circumvents the need for direct contact with measurement devices but also enhances privacy and data security through privacy computing measures. Our proposed method was tested in an HRC experiment to provide preliminary validation. This pioneering use of non-intrusive AI-based vision techniques for real-time mental workload assessment represents a significant advancement in managing human factors in industrial HRC settings.
- 27 August 2024
- Published by EDP Sciences in MATEC Web of Conferences
- Vol. 401, 12002
Open AccessA novel non-intrusive mental workload evaluation concept in human-robot collaboration
The integration of Human-Robot Collaboration (HRC) in industrial robotics introduces challenges, particularly in adapting manufacturing environments to work seamlessly with collaborative robots. A key objective in HRC system optimization is enhancing human acceptance of these robots and improving productivity. Traditionally, the assessment of human mental workload in these settings relies on methods like EEG, fNIRS, and heart rate monitoring, which require direct physical contact and can be impractical in manufacturing environments. To address these issues, we propose an innovative and non-intrusive method that employs cameras to measure mental workload. This technique involves capturing video footage of human operators on the shop floor, focusing specifically on facial expressions. Advanced AI algorithms analyse these videos to predict heart rate ranges, which are then used to estimate mental workload levels in real time. This approach not only circumvents the need for direct contact with measurement devices but also enhances privacy and data security through privacy computing measures. Our proposed method was tested in an HRC experiment to provide preliminary validation. This pioneering use of non-intrusive AI-based vision techniques for real-time mental workload assessment represents a significant advancement in managing human factors in industrial HRC settings.
- Open Access
- 14 August 2024
- Published by British Editorial Society of Bone & Joint Surgery in Bone & Joint Open
- Vol. 5 (8) , 671-680
Is it feasible to develop a supervised learning algorithm incorporating spinopelvic mobility to predict impingement in patients undergoing total hip arthroplasty?
Aims: Precise implant positioning, tailored to individual spinopelvic biomechanics and phenotype, is paramount for stability in total hip arthroplasty (THA). Despite a few studies on instability prediction, there is a notable gap in research utilizing artificial intelligence (AI). The objective of our pilot study was to evaluate the feasibility of developing an AI algorithm tailored to individual spinopelvic mechanics and patient phenotype for predicting impingement.Methods: This international, multicentre prospective cohort study across two centres encompassed 157 adults undergoing primary robotic arm-assisted THA. Impingement during specific flexion and extension stances was identified using the virtual range of motion (ROM) tool of the robotic software. The primary AI model, the Light Gradient-Boosting Machine (LGBM), used tabular data to predict impingement presence, direction (flexion or extension), and type. A secondary model integrating tabular data with plain anteroposterior pelvis radiographs was evaluated to assess for any potential enhancement in prediction accuracy.Results: We identified nine predictors from an analysis of baseline spinopelvic characteristics and surgical planning parameters. Using fivefold cross-validation, the LGBM achieved 70.2% impingement prediction accuracy. With impingement data, the LGBM estimated direction with 85% accuracy, while the support vector machine (SVM) determined impingement type with 72.9% accuracy. After integrating imaging data with a multilayer perceptron (tabular) and a convolutional neural network (radiograph), the LGBM’s prediction was 68.1%. Both combined and LGBM-only had similar impingement direction prediction rates (around 84.5%).Conclusion: This study is a pioneering effort in leveraging AI for impingement prediction in THA, utilizing a comprehensive, real-world clinical dataset. Our machine-learning algorithm demonstrated promising accuracy in predicting impingement, its type, and direction. While the addition of imaging data to our deep-learning algorithm did not boost accuracy, the potential for refined annotations, such as landmark markings, offers avenues for future enhancement. Prior to clinical integration, external validation and larger-scale testing of this algorithm are essential.Cite this article: Bone Jt Open 2024;5(8):671–680.
- 14 August 2024
- Published by British Editorial Society of Bone & Joint Surgery in Bone & Joint Open
- Vol. 5 (8) , 671-680
Open AccessIs it feasible to develop a supervised learning algorithm incorporating spinopelvic mobility to predict impingement in patients undergoing total hip arthroplasty?
Aims: Precise implant positioning, tailored to individual spinopelvic biomechanics and phenotype, is paramount for stability in total hip arthroplasty (THA). Despite a few studies on instability prediction, there is a notable gap in research utilizing artificial intelligence (AI). The objective of our pilot study was to evaluate the feasibility of developing an AI algorithm tailored to individual spinopelvic mechanics and patient phenotype for predicting impingement.Methods: This international, multicentre prospective cohort study across two centres encompassed 157 adults undergoing primary robotic arm-assisted THA. Impingement during specific flexion and extension stances was identified using the virtual range of motion (ROM) tool of the robotic software. The primary AI model, the Light Gradient-Boosting Machine (LGBM), used tabular data to predict impingement presence, direction (flexion or extension), and type. A secondary model integrating tabular data with plain anteroposterior pelvis radiographs was evaluated to assess for any potential enhancement in prediction accuracy.Results: We identified nine predictors from an analysis of baseline spinopelvic characteristics and surgical planning parameters. Using fivefold cross-validation, the LGBM achieved 70.2% impingement prediction accuracy. With impingement data, the LGBM estimated direction with 85% accuracy, while the support vector machine (SVM) determined impingement type with 72.9% accuracy. After integrating imaging data with a multilayer perceptron (tabular) and a convolutional neural network (radiograph), the LGBM’s prediction was 68.1%. Both combined and LGBM-only had similar impingement direction prediction rates (around 84.5%).Conclusion: This study is a pioneering effort in leveraging AI for impingement prediction in THA, utilizing a comprehensive, real-world clinical dataset. Our machine-learning algorithm demonstrated promising accuracy in predicting impingement, its type, and direction. While the addition of imaging data to our deep-learning algorithm did not boost accuracy, the potential for refined annotations, such as landmark markings, offers avenues for future enhancement. Prior to clinical integration, external validation and larger-scale testing of this algorithm are essential.Cite this article: Bone Jt Open 2024;5(8):671–680.