Measuring movement and its contributors is done using state of the art equipment in specialized laboratory sites. While resulting in highly accurate measures, the technology-person-interference is high and the environment restricts the movement radius and also movement specific situations. Therefore, the need for methods that allow for the reduction of equipment and data collection in more realistic scenarios has arisen. With the advancement in the use of artificial intelligence for biomechanical modelling, ranging from utilizing regression algorithms to replace or support conventional laboratory measurements, to use classification approaches to distinguish movement conditions, on-field data collection has become realistic. Several papers have reported promising results by applying machine learning techniques with biomechanical data which generally show the ability of these algorithms to reduce the negative impact of laboratory high-tech approaches.
The majority of the proposed applications are prototypes that have shown considerable outcomes in comparison to the golden-standard approach. Still, the level of accuracy is below the accepted standards for conventional methods, which could be supported by more advanced neural networks which will then make them applicable in biomechanical research. However, there is a clear lack of proof-of-concept studies that investigate the laboratory-based results in the target environment but also the target population. Also, studies that realistically estimate the technical, financial and level-of expert cost when applying novel techniques are needed. This would uncover potential over-engineering and at the same time help to improve these novel techniques. The applicability of any novel method needs to be investigated to be accepted in the field, especially when it comes to clinical settings. In addition to this, there also needs to be clarity on how to evaluate and validate novel approaches.
The overall aim of this Research Topic is therefore to raise awareness about the importance of such proof-of-concept and validation studies in order to support the applicability and accuracy of proposed algorithms and to improve them further.
Topics of interest include (but are not limited to):
• Proof of concept of on-field testing
• Reduction of technology to allow on-field testing
• Support of technology to allow applicability by technology reduction
• Real-time feedback applications
• Health-monitoring
Measuring movement and its contributors is done using state of the art equipment in specialized laboratory sites. While resulting in highly accurate measures, the technology-person-interference is high and the environment restricts the movement radius and also movement specific situations. Therefore, the need for methods that allow for the reduction of equipment and data collection in more realistic scenarios has arisen. With the advancement in the use of artificial intelligence for biomechanical modelling, ranging from utilizing regression algorithms to replace or support conventional laboratory measurements, to use classification approaches to distinguish movement conditions, on-field data collection has become realistic. Several papers have reported promising results by applying machine learning techniques with biomechanical data which generally show the ability of these algorithms to reduce the negative impact of laboratory high-tech approaches.
The majority of the proposed applications are prototypes that have shown considerable outcomes in comparison to the golden-standard approach. Still, the level of accuracy is below the accepted standards for conventional methods, which could be supported by more advanced neural networks which will then make them applicable in biomechanical research. However, there is a clear lack of proof-of-concept studies that investigate the laboratory-based results in the target environment but also the target population. Also, studies that realistically estimate the technical, financial and level-of expert cost when applying novel techniques are needed. This would uncover potential over-engineering and at the same time help to improve these novel techniques. The applicability of any novel method needs to be investigated to be accepted in the field, especially when it comes to clinical settings. In addition to this, there also needs to be clarity on how to evaluate and validate novel approaches.
The overall aim of this Research Topic is therefore to raise awareness about the importance of such proof-of-concept and validation studies in order to support the applicability and accuracy of proposed algorithms and to improve them further.
Topics of interest include (but are not limited to):
• Proof of concept of on-field testing
• Reduction of technology to allow on-field testing
• Support of technology to allow applicability by technology reduction
• Real-time feedback applications
• Health-monitoring