Design and test of neural networks and statistical classifiers in computer-aided movement analysis: a case study on gait analysis

Lafuente, Rubén; Belda, J.M.; Sánchez-Lacuesta, J.; Soler, Cecilia Ferrer; Prat, Jaime

doi:10.1016/s0268-0033(97)00082-x

Cited by 45 publications

(24 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The average accuracy of classification rate among the three walking conditions was 83.3%. Lafuente et al (1998) reverted to the standard network structure (i.e. with one hidden layer), utilized likewise by Holzreiter and Kohle (1993) when attempting a classification into four gait categories.…”

Section: Discussionmentioning

confidence: 99%

“…Another advantage is the ability to capture patterns in the data within their internal parameters known as weights and biases which may influence gait patterns (Hinton, 1992). A significant number of papers (Holzreit & Kohle, 1993;Barton & Lees, 1997Gioftsos & Grieve, 1995;Lafuente et al, 1998; as well as our own research outlined in the next section) have shown ANNs to be useful in distinguishing gait patterns. Recent efforts generally fall into three categories of application: i) classification of human gait, ii) biomechanical modelling, iii) prediction of gait variables and parameters (Chau, 2001).…”

Section: Gait Classification Using Annsmentioning

confidence: 95%

See 1 more Smart Citation

Artificial Neural Networks (ANN) Applied for Gait Classification and Physiotherapy Monitoring in Post Stroke Patients

Kaczmarczyk¹,

Wit²,

Krawczyk³

et al. 2011

Artificial Neural Networks - Methodological Advances and Biomedical Applications

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Section: Gait Classification Using Annsmentioning

confidence: 95%

Artificial Neural Networks (ANN) Applied for Gait Classification and Physiotherapy Monitoring in Post Stroke Patients

Kaczmarczyk¹,

Wit²,

Krawczyk³

et al. 2011

Artificial Neural Networks - Methodological Advances and Biomedical Applications

View full text Add to dashboard Cite

“…The GRFs amplitudes were normalized based on body mass [19] [20] [21] [22]. The gait cycle/stride time and phases are determined based on the vertical component of GRFs [23].…”

Section: Data Analysis and Processingmentioning

confidence: 99%

Application of Multiple Sensor Data Fusion for the Analysis of Human Dynamic Behavior in Space: Assessment and Evaluation of Mobility-Related Functional Impairments

Sarkodie-Gyan¹,

Yu²,

Bogale³

et al. 2017

JBiSE

View full text Add to dashboard Cite

The authors have applied a systems analysis approach to describe the musculoskeletal system as consisting of a stack of superimposed kinematic hierarchical segments in which each lower segment tends to transfer its motion to the other superimposed segments. This segmental chain enables the derivation of both conscious perception and sensory control of action in space. This applied systems analysis approach involves the measurements of the complex motor behavior in order to elucidate the fusion of multiple sensor data for the reliable and efficient acquisition of the kinetic, kinematics and electromyographic data of the human spatial behavior. The acquired kinematic and related kinetic signals represent attributive features of the internal reconstruction of the physical links between the superimposed body segments. Indeed, this reconstruction of the physical links was established as a result of the fusion of the multiple sensor data. Furthermore, this acquired kinematics, kinetics and electromyographic data provided detailed means to record, annotate, process, transmit, and display pertinent information derived from the musculoskeletal system to quantify and differentiate between subjects with mobility-related disabilities and able-bodied subjects, and enabled an inference into the active neural processes underlying balance reactions. To gain insight into the basis for this long-term dependence, the authors have applied the fusion of multiple sensor data to investigate the effects of Cerebral Palsy, Multiple Sclerosis and Diabetic Neuropathy conditions, on biomechanical/neurophysiological changes that may alter the ability of the human locomotor system to generate ambulaHow to cite this paper:

show abstract

“…A seleção do conjunto adequado de parâmetros para a quantificação do contorno do tumor ainda é um problema em aberto. Alguns trabalhos utilizam métodos estatísticos lineares, como a Análise Linear Discriminante ou Análise de Componente Principal, para avaliar o potencial de discriminação dos parâmetros (Lafuente J et al, 1997).…”

Section: Introductionunclassified

Classificação De Imagens De Ultra-Som De Tumor De Mama Através De Técnicas De Inteligência Computacional

Macrini¹,

Alvarenga²,

Pereira³

et al. 2016

Anais Do 8. Congresso Brasileiro De Redes Neurais

View full text Add to dashboard Cite

Mammography is the most important screening exam for early diagnosis of breast cancer. However, it still presents a lack of specificity, since a considerable number of suspicious lesions have to undergo surgical biopsy. Therefore, ultrasound (US) breast images have been used to improve diagnostics and decrease the number of unneeded biopsies for patients with palpable mass and inconclusive mammograms. Malignant breast tumors tend to present irregular and blurred contours while benign ones are usually round, smooth and well-defined. Accordingly, investigating the tumor contour may decisively help in establishing the diagnosis. On this basis, a non-linear method, Mutual Information (MI), and a linear one (Linear Discriminant Analysis -LDA) were implemented to rank morphological features in discriminating breast tumors in US images. Compared to LDA, the MI was able to identify the diagnostic contribution of Mshape feature. A feedforward neural network was implemented using as input seven morphological parameters to classify the breast US images. The best performance (88 %) was obtained with the five first parameters selected by MI.Keywords Mutual Information, parameters selection, ultrasound images, breast cancer, neural networks.Resumo -A mamografia tem sido apontada como a única técnica diagnóstica que comprovadamente contribui, através de um programa de acompanhamento periódico, para a detecção precoce e redução da mortalidade por câncer de mama. Porém, uma quantidade considerável de lesões suspeitas ainda tem sido encaminhadas para biópsia. Imagens de ultrasom (US) têm sido usadas para melhorar o diagnóstico e diminuir o número de biópsias desnecessárias em pacientes com massas palpáveis e mamografias inconclusivas. Tumores malignos de mama tendem a apresentar contornos irregulares e borrados, enquanto tumores benignos são normalmente redondos, lisos e bem-definidos. A investigação do contorno do tumor pode ajudar no diagnóstico. Neste artigo, um método linear e um não-linear foram utilizados para ordenar parâmetros morfométricos usados para quantificar o contorno e a forma de tumores de mama em imagens por US. Comparado com o método linear, o método não-linear pôde identificar a contribuição não-linear do parâmetro morfométrico Mshape em relação ao diagnóstico. Uma rede neural feedforward foi implementada, tendo como entrada sete parâmetros morfométricos para classificar as imagens. O melhor desempenho (88%) foi obtido com os primeiros cinco parâmetros selecionados pelo método não-linear.Palavras-chave -Informação Mútua, seleção de parâmetros, imagens de ultra-som, câncer de mama, redes neurais.

show abstract

Design and test of neural networks and statistical classifiers in computer-aided movement analysis: a case study on gait analysis

Cited by 45 publications

References 23 publications

Artificial Neural Networks (ANN) Applied for Gait Classification and Physiotherapy Monitoring in Post Stroke Patients

Artificial Neural Networks (ANN) Applied for Gait Classification and Physiotherapy Monitoring in Post Stroke Patients

Application of Multiple Sensor Data Fusion for the Analysis of Human Dynamic Behavior in Space: Assessment and Evaluation of Mobility-Related Functional Impairments

Classificação De Imagens De Ultra-Som De Tumor De Mama Através De Técnicas De Inteligência Computacional

Contact Info

Product

Resources

About