Now showing 1 - 3 of 3
No Thumbnail Available
Publication

Artificial hydrocarbon networks for freezing of gait detection in Parkinson’s disease

2020 , Martinez-Villaseñor, Lourdes , Ponce, Hiram , Nuñez Martínez, José Pablo

Freezing of gait (FoG) is one of the most impairing phenomenon experienced by Parkinson's disease (PD) patients. This phenomenon is associated with falls and is an important factor that limits autonomy and impairs quality of life of PD patients. Pharmacological treatment is difficult and do not always help to deal with this problem. Robust FoG detection systems can help monitoring and identifying when a patient needs aid providing external cueing to deal with FoG episodes. In this paper, we describe a comparative analysis of traditional machine learning techniques against Artificial Hydrocarbon Networks (AHN) for FoG detection. We compared four supervised machine learning classifiers and AHN for FoG event detection using a publicly available dataset, obtaining 88% of F-score metric with AHN. We prove that AHN are suitable for FoG detection. © 2020 IEEE.

No Thumbnail Available
Publication

A vision-based approach for fall detection using multiple cameras and convolutional neural networks: A case study using the UP-Fall detection dataset

2019 , Espinosa Loera, Ricardo Abel , Ponce, Hiram , Moya-Albor, Ernesto , Martinez-Villaseñor, Lourdes , Brieva, Jorge , Gutiérrez, Sebastián

The automatic recognition of human falls is currently an important topic of research for the computer vision and artificial intelligence communities. In image analysis, it is common to use a vision-based approach for fall detection and classification systems due to the recent exponential increase in the use of cameras. Moreover, deep learning techniques have revolutionized vision-based approaches. These techniques are considered robust and reliable solutions for detection and classification problems, mostly using convolutional neural networks (CNNs). Recently, our research group released a public multimodal dataset for fall detection called the UP-Fall Detection dataset, and studies on modality approaches for fall detection and classification are required. Focusing only on a vision-based approach, in this paper, we present a fall detection system based on a 2D CNN inference method and multiple cameras. This approach analyzes images in fixed time windows and extracts features using an optical flow method that obtains information on the relative motion between two consecutive images. We tested this approach on our public dataset, and the results showed that our proposed multi-vision-based approach detects human falls and achieves an accuracy of 95.64% compared to state-of-the-art methods with a simple CNN network architecture. © 2019 Elsevier Ltd

No Thumbnail Available
Publication

A Method to Improve Speed of Training Algorithm in Artificial Hydrocarbon Networks

2019 , Campos Souza, Paulo V. de , Ponce, Hiram , Martinez-Villaseñor, Lourdes

Artificial hydrocarbon networks (AHN) is a supervised machine learning method inspired on chemical carbon networks that simulate heuristic chemical rules involved within organic molecules to represent the structure and behavior of data. However, training AHN depends on a relevant number of parameters. In that sense, the original training algorithm presents some issues to find suitable parameters in a reasonable amount of time. Thus, this paper proposes a new training algorithm for AHN based on the concept of extreme learning machines, to update weight parameters related to the molecular functions. To evaluate the effectiveness of the proposed algorithm, binary classification and regression tests are performed over real public datasets from a central data repository specialized in machine learning problems. The results obtained validated that the updating of the weight parameters using the new training algorithm in the molecular structures is efficient and maintains the expected results of model accuracy. In addition, this work increased up to 24.88% the speed of the training phase in contrast to the original algorithm. © 2019 IEEE.