Machine learning-based prediction of joint moments based on kinematics in patients with cerebral palsy

Özates, Mustafa Erkam; Karabulut, Derya; Salami, Firooz; Wolf, Sebastian I.; Arslan, Yunus Ziya

doi:10.1016/j.jbiomech.2023.111668

Cited by 4 publications

(7 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This study utilized the same input and output data as Ozates et al 10 . The dataset comprises anonymized retrospective gait kinematics and kinetics data from 622 CP patients with spastic diplegia, walking at self-selected speeds.…”

Section: Datamentioning

confidence: 99%

“…Only data from the stance phase were utilized throughout the process. For further details, we refer you to Ozates et al 10 .…”

Section: Datamentioning

confidence: 99%

“…Given that the AI model, employed in this study, is thoroughly detailed in reference [10], a concise overview of the model is provided in this section. The implemented machine learning architecture is based on a one-dimensional convolutional neural network, which had ve convolutional layers with an increasing number of lters [128, 128, 512, 1024, 2048] and decreasing one-dimensional kernel sizes [30, 15,10,5,3]. This architecture aimed to extract features for varying time intervals.…”

Section: Ai Model (The Convolutional Neural Network)mentioning

confidence: 99%

“…Ozates et al addressed this gap by training a convolutional neural network model to predict lower extremity joint moments during walking. They utilized retrospective gait data from 132 typically developed individuals and 622 patients with bilateral cerebral palsy (CP), drawing from a sizable database for their study 10 .…”

Section: Introductionmentioning

confidence: 99%

“…In this follow-up investigation, our aim was to comprehensively scrutinize the output of the AI model used in the study by Ozates et al 10 concerning patients with CP during walking. To achieve this objective, we employed a three-step assessment approach.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Evaluating feasibility of lower extremity joint moments predicted by an artificial intelligence model during walking in patients with cerebral palsy

Salami,

Ozates,

Arslan

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

Several studies have highlighted the advantages of employing artificial intelligence (AI) models in gait analysis. However, the practicality of AI models into clinical gait routines remains uncertain. In this study, we propose a three-step approach to evaluate the feasibility of a previously developed AI model. This model predicts joint moments during walking for 622 patients with cerebral palsy using joint kinematics as the input. Root mean square errors between lab-measured and predicted joint moments are labeled as Green (acceptable), Yellow (acceptable with caution), or Red (unacceptable). Kinematics are classified accordingly, and statistical analyses determine their impact on the AI model output. A linear discrimination analysis (LDA) model predicts labels for newly predicted joint moments based on kinematics. The knee moment showed the largest Green label population (73%), while the ankle moment has the smallest (34%). Gait profile score show significant differences between all labels except the Green vs Red for the ankle joint. The LDA model achieves75% accuracy for knee joint moment prediction, with a Green sensitivity of 94.7%. Severer patient’s condition leads to increase in Red population. While the AI model shows promise for predicting knee and hip moments, further development is necessary before its integration into clinical routines.

show abstract

Section: Datamentioning

confidence: 99%

“…Only data from the stance phase were utilized throughout the process. For further details, we refer you to Ozates et al 10 .…”

Section: Datamentioning

confidence: 99%