# Spatiotemporal Ground Reaction Force Analysis using Convolutional Neural Networks to Analyze Parkinsonian Gait

arXiv: https://arxiv.org/abs/2102.00628 · HTML: https://mlsanity.com/paper/2102.00628/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 6 of its 48 references that are in the map.
- First posted on arXiv 1 Feb 2021.

## Builds on (most cited first, 6 of 6)

- [Fully Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1411.4038/) (2014)
- [Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition](https://mlsanity.com/paper/1406.4729/) (2014)
- [A systematic study of the class imbalance problem in convolutional neural networks](https://mlsanity.com/paper/1710.05381/) (2017)
- [The Real-World-Weight Cross-Entropy Loss Function: Modeling the Costs of Mislabeling](https://mlsanity.com/paper/2001.00570/) (2020)
- [Potential, Challenges and Future Directions for Deep Learning in Prognostics and Health Management Applications](https://mlsanity.com/paper/2005.02144/) (2020)
- [FPGA-based Accelerators of Deep Learning Networks for Learning and Classification: A Review](https://mlsanity.com/paper/1901.00121/) (2019)

## Abstract (arXiv)

Parkinson's disease (PD) is a non-curable disease that commonly found among elders that greatly reduce their quality of life. PD primarily affects the gait pattern and slowly changes the walking gait from the normality to disability. The early diagnosing of PD is important for treatments and gait pattern analysis is used as a technique to diagnose PD. The present paper has identified the raw spatiotemporal ground reaction force (GRF) as a key parameter to identify the changes in human gait patterns associated with PD. The changes in GRF are identified using a convolutional neural network through pre-processing, conversion, recognition, and performance evaluation. The proposed algorithm is capable of identifying the severity of the PD and distinguishing the parkinsonian gait from the healthy gait. The technique has shown a 97% of accuracy in automatic decision-making process.