The Experts below are selected from a list of 76173 Experts worldwide ranked by ideXlab platform
Jarek Krajewski - One of the best experts on this subject based on the ideXlab platform.
-
Generalized Two-Stage Rank Regression Framework for Depression Score Prediction from Speech
IEEE Transactions on Affective Computing, 2020Co-Authors: Nicholas Cummins, Julien Epps, Vidhyasaharan Sethu, James R. Williamson, Thomas F. Quatieri, Jarek KrajewskiAbstract:This paper introduces a novel speech-based Depression Score prediction paradigm, the 2-stage ranking prediction framework, and highlights the benefits it brings to Depression prediction. Conventional regression approaches aim to discern a single functional relationship between speech features and Depression Scores, making an implicit assumption about the existence of a single fixed relationship between the features and Scores. However, as the relationship between severity of Depression and the clinical Score may vary over the range of the assessment scale, this style of analysis may not be suited to Depression prediction. The proposed framework on the other hand, imposes a series of partitions on the feature space, with each partition corresponding to a distinct predefined range of Depression Scores, and predicts the Score based on measures of membership to each partition. This approach provides additional flexibility by allowing different rankings to be learnt for different Depression Scores, and relaxes assumptions made by conventional regression approaches. Results demonstrate the framework's suitability for Depression Score prediction: different 2-stage implementations, based on heterogeneous feature extraction and modelling approaches, produce state-of-the-art results on the AVEC-2013 dataset. It is also demonstrated that, unlike fusion of conventional regression systems, the fusion of two-stage systems consistently improves prediction performance.
-
weighted pairwise gaussian likelihood regression for Depression Score prediction
International Conference on Acoustics Speech and Signal Processing, 2015Co-Authors: Nicholas Cummins, Julien Epps, Vidhyasaharan Sethu, Jarek KrajewskiAbstract:This paper presents a technique in which feature vectors are mapped onto ordinal ranges of clinical Depression Scores using weighted pairwise Gaussians. The position of a test vector with respect to these partitions is used to perform Depression Score prediction. Results found on a set of spectral and formant based speech characteristics indicate the potential of this technique for performing Depression Score prediction. Key results on the AVEC 2013 development set indicate that the inclusion of weights and Bayesian adaptation improves system performance by 16.5% – 18.5% when compared to using an unweighted non-adapted system. Fusing results from Bayesian adapted models corresponding to different feature spaces offers up to 8% further improvement. Further, fusion consistently improves performance on both the AVEC 2013 development and test set, in contrast to conventional regressor fusion.
-
ICASSP - Weighted pairwise Gaussian likelihood regression for Depression Score prediction
2015 IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP), 2015Co-Authors: Nicholas Cummins, Julien Epps, Vidhyasaharan Sethu, Jarek KrajewskiAbstract:This paper presents a technique in which feature vectors are mapped onto ordinal ranges of clinical Depression Scores using weighted pairwise Gaussians. The position of a test vector with respect to these partitions is used to perform Depression Score prediction. Results found on a set of spectral and formant based speech characteristics indicate the potential of this technique for performing Depression Score prediction. Key results on the AVEC 2013 development set indicate that the inclusion of weights and Bayesian adaptation improves system performance by 16.5% – 18.5% when compared to using an unweighted non-adapted system. Fusing results from Bayesian adapted models corresponding to different feature spaces offers up to 8% further improvement. Further, fusion consistently improves performance on both the AVEC 2013 development and test set, in contrast to conventional regressor fusion.
P. Ebadi - One of the best experts on this subject based on the ideXlab platform.
-
P03-16 - Higher Depression Score is associated with the level of antinuclear-antibody, soluble-gp130, soluble-leptin-receptor, hookah and cigarette smoking in systemic lupus erythematosus
European Psychiatry, 2011Co-Authors: K. Bagheri, P. EbadiAbstract:Introduction Depression stands out as an important health issue. Systemic lupus erythematosus (SLE) is a prototypic autoimmune disorder with possible involvement of each body organ and system. Objectives to determine prevalence and level of Depression and its relationship with autoantibody titers, individual, environmental, laboratory and lifestyle parameters in SLE-patients Aims to evaluate the causes or effects of Depression in SLE-patients. Methods 50 variables containing sociodemographics, health-related habits, Depression, serum molecules and blood parameters were evaluated in SLE-patients and healthy controls. Results Although no significant differences for 4 sociodemographic and 18 health-related lifestyle parameters were found between patients and controls but Depression Score and its prevalence in SLE-patients was higher. SLE-patients had higher Depression Score than controls. 94.12% of SLE-patients had no or mild Depression, while the rest had mild-moderate or moderate Depression. Furthermore, more depressed SLE-patients had higher level of autoantibodies. Between lupus-related-parameters such as autoantibodies, complement level and other hematological parameters, only the correlation of antinuclear-antibody (ANA) titer and Depression Score remained significant after controlling for variables. Interestingly, Depression Score was inversely correlated with the level of soluble-leptin-receptor (sLeptinR) and soluble-glyeoprotein-130 (sgp130) after controlling for variables in SLE-patients. Depression of SLE-patients had a direct association with alcohol, cigarette and hookah consumption. Conclusions Depression Score and its prevalence are higher in SLE-patients than controls. These Depression levels are correlated with ANA or some habits like hookah smoking in patient group. Inverse association between Depression and some soluble molecules like sgp130 or sLeptinR in SLE-patients also remained to be elucidated.
-
P03-15 - Higher Depression Score in recurrent miscarriage patients is associated with the level of anti-double-stranded-DNA-antibody, soluble-gp130, soluble-leptin-receptor and hookah smoking
European Psychiatry, 2011Co-Authors: K. Bagheri, P. EbadiAbstract:Introduction Depression stands out as an important health issue that affects the entire family. Objectives to assess the prevalence and level of Depression and its relationship with autoantibodies, individual, environmental, laboratory and lifestyle parameters in miscarriage-patients Aims to evaluate the causes or effects of Depression in recurrent miscarriage-patients. Methods 50 variables containing sociodemographics, health-related habits, Depression, serum molecules and blood parameters were evaluated in recurrent miscarriage-patients and healthy women with natural childbirth. Results Although no significant differences for 4 sociodemographic and 18 health-related lifestyle parameters were found between patients and controls but Depression Score and its prevalence was higher in patients than controls. Depression Scores were higher in miscarriage-patients (ranging 3–36, mean = 11.92 ± 1.37) than controls (ranging 0–8, mean = 3.05 ± 0.63) which 80.96% of them had no or mild, 2.38% mild-moderate, 4.76% moderate and 11.9% severe Depression. There was also a significant association between anti-double-stranded-DNA-antibody (anti-dsDNA) or rheumatoid-factor (RF) level and Depression in miscarriage-patients. Interestingly, Depression Score was inversely correlated with the level of soluble-leptin-receptor (sLeptinR) and soluble-glyeoprotein-130 (sgp130) after controlling for covariates in patients. Depression of miscarriage-patients was correlated with hookah consumption rate. Surprising that, the association between abortion histories and Depression in miscarriage-patients become not significant after controlling for age. Conclusions Depression Score and its prevalence are higher in miscarriage-patients than controls. The Depression levels are correlated with certain autoantibodies such as anti-dsDNA or some habits like hookah smoking in miscarriage-patients. Inverse association between Depression and some soluble molecules like sgp130 or sLeptinR in recurrent miscarriage-patients also remained to be elucidated.
Nicholas Cummins - One of the best experts on this subject based on the ideXlab platform.
-
Generalized Two-Stage Rank Regression Framework for Depression Score Prediction from Speech
IEEE Transactions on Affective Computing, 2020Co-Authors: Nicholas Cummins, Julien Epps, Vidhyasaharan Sethu, James R. Williamson, Thomas F. Quatieri, Jarek KrajewskiAbstract:This paper introduces a novel speech-based Depression Score prediction paradigm, the 2-stage ranking prediction framework, and highlights the benefits it brings to Depression prediction. Conventional regression approaches aim to discern a single functional relationship between speech features and Depression Scores, making an implicit assumption about the existence of a single fixed relationship between the features and Scores. However, as the relationship between severity of Depression and the clinical Score may vary over the range of the assessment scale, this style of analysis may not be suited to Depression prediction. The proposed framework on the other hand, imposes a series of partitions on the feature space, with each partition corresponding to a distinct predefined range of Depression Scores, and predicts the Score based on measures of membership to each partition. This approach provides additional flexibility by allowing different rankings to be learnt for different Depression Scores, and relaxes assumptions made by conventional regression approaches. Results demonstrate the framework's suitability for Depression Score prediction: different 2-stage implementations, based on heterogeneous feature extraction and modelling approaches, produce state-of-the-art results on the AVEC-2013 dataset. It is also demonstrated that, unlike fusion of conventional regression systems, the fusion of two-stage systems consistently improves prediction performance.
-
weighted pairwise gaussian likelihood regression for Depression Score prediction
International Conference on Acoustics Speech and Signal Processing, 2015Co-Authors: Nicholas Cummins, Julien Epps, Vidhyasaharan Sethu, Jarek KrajewskiAbstract:This paper presents a technique in which feature vectors are mapped onto ordinal ranges of clinical Depression Scores using weighted pairwise Gaussians. The position of a test vector with respect to these partitions is used to perform Depression Score prediction. Results found on a set of spectral and formant based speech characteristics indicate the potential of this technique for performing Depression Score prediction. Key results on the AVEC 2013 development set indicate that the inclusion of weights and Bayesian adaptation improves system performance by 16.5% – 18.5% when compared to using an unweighted non-adapted system. Fusing results from Bayesian adapted models corresponding to different feature spaces offers up to 8% further improvement. Further, fusion consistently improves performance on both the AVEC 2013 development and test set, in contrast to conventional regressor fusion.
-
ICASSP - Weighted pairwise Gaussian likelihood regression for Depression Score prediction
2015 IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP), 2015Co-Authors: Nicholas Cummins, Julien Epps, Vidhyasaharan Sethu, Jarek KrajewskiAbstract:This paper presents a technique in which feature vectors are mapped onto ordinal ranges of clinical Depression Scores using weighted pairwise Gaussians. The position of a test vector with respect to these partitions is used to perform Depression Score prediction. Results found on a set of spectral and formant based speech characteristics indicate the potential of this technique for performing Depression Score prediction. Key results on the AVEC 2013 development set indicate that the inclusion of weights and Bayesian adaptation improves system performance by 16.5% – 18.5% when compared to using an unweighted non-adapted system. Fusing results from Bayesian adapted models corresponding to different feature spaces offers up to 8% further improvement. Further, fusion consistently improves performance on both the AVEC 2013 development and test set, in contrast to conventional regressor fusion.
K. Bagheri - One of the best experts on this subject based on the ideXlab platform.
-
P03-16 - Higher Depression Score is associated with the level of antinuclear-antibody, soluble-gp130, soluble-leptin-receptor, hookah and cigarette smoking in systemic lupus erythematosus
European Psychiatry, 2011Co-Authors: K. Bagheri, P. EbadiAbstract:Introduction Depression stands out as an important health issue. Systemic lupus erythematosus (SLE) is a prototypic autoimmune disorder with possible involvement of each body organ and system. Objectives to determine prevalence and level of Depression and its relationship with autoantibody titers, individual, environmental, laboratory and lifestyle parameters in SLE-patients Aims to evaluate the causes or effects of Depression in SLE-patients. Methods 50 variables containing sociodemographics, health-related habits, Depression, serum molecules and blood parameters were evaluated in SLE-patients and healthy controls. Results Although no significant differences for 4 sociodemographic and 18 health-related lifestyle parameters were found between patients and controls but Depression Score and its prevalence in SLE-patients was higher. SLE-patients had higher Depression Score than controls. 94.12% of SLE-patients had no or mild Depression, while the rest had mild-moderate or moderate Depression. Furthermore, more depressed SLE-patients had higher level of autoantibodies. Between lupus-related-parameters such as autoantibodies, complement level and other hematological parameters, only the correlation of antinuclear-antibody (ANA) titer and Depression Score remained significant after controlling for variables. Interestingly, Depression Score was inversely correlated with the level of soluble-leptin-receptor (sLeptinR) and soluble-glyeoprotein-130 (sgp130) after controlling for variables in SLE-patients. Depression of SLE-patients had a direct association with alcohol, cigarette and hookah consumption. Conclusions Depression Score and its prevalence are higher in SLE-patients than controls. These Depression levels are correlated with ANA or some habits like hookah smoking in patient group. Inverse association between Depression and some soluble molecules like sgp130 or sLeptinR in SLE-patients also remained to be elucidated.
-
P03-15 - Higher Depression Score in recurrent miscarriage patients is associated with the level of anti-double-stranded-DNA-antibody, soluble-gp130, soluble-leptin-receptor and hookah smoking
European Psychiatry, 2011Co-Authors: K. Bagheri, P. EbadiAbstract:Introduction Depression stands out as an important health issue that affects the entire family. Objectives to assess the prevalence and level of Depression and its relationship with autoantibodies, individual, environmental, laboratory and lifestyle parameters in miscarriage-patients Aims to evaluate the causes or effects of Depression in recurrent miscarriage-patients. Methods 50 variables containing sociodemographics, health-related habits, Depression, serum molecules and blood parameters were evaluated in recurrent miscarriage-patients and healthy women with natural childbirth. Results Although no significant differences for 4 sociodemographic and 18 health-related lifestyle parameters were found between patients and controls but Depression Score and its prevalence was higher in patients than controls. Depression Scores were higher in miscarriage-patients (ranging 3–36, mean = 11.92 ± 1.37) than controls (ranging 0–8, mean = 3.05 ± 0.63) which 80.96% of them had no or mild, 2.38% mild-moderate, 4.76% moderate and 11.9% severe Depression. There was also a significant association between anti-double-stranded-DNA-antibody (anti-dsDNA) or rheumatoid-factor (RF) level and Depression in miscarriage-patients. Interestingly, Depression Score was inversely correlated with the level of soluble-leptin-receptor (sLeptinR) and soluble-glyeoprotein-130 (sgp130) after controlling for covariates in patients. Depression of miscarriage-patients was correlated with hookah consumption rate. Surprising that, the association between abortion histories and Depression in miscarriage-patients become not significant after controlling for age. Conclusions Depression Score and its prevalence are higher in miscarriage-patients than controls. The Depression levels are correlated with certain autoantibodies such as anti-dsDNA or some habits like hookah smoking in miscarriage-patients. Inverse association between Depression and some soluble molecules like sgp130 or sLeptinR in recurrent miscarriage-patients also remained to be elucidated.
Julien Epps - One of the best experts on this subject based on the ideXlab platform.
-
Generalized Two-Stage Rank Regression Framework for Depression Score Prediction from Speech
IEEE Transactions on Affective Computing, 2020Co-Authors: Nicholas Cummins, Julien Epps, Vidhyasaharan Sethu, James R. Williamson, Thomas F. Quatieri, Jarek KrajewskiAbstract:This paper introduces a novel speech-based Depression Score prediction paradigm, the 2-stage ranking prediction framework, and highlights the benefits it brings to Depression prediction. Conventional regression approaches aim to discern a single functional relationship between speech features and Depression Scores, making an implicit assumption about the existence of a single fixed relationship between the features and Scores. However, as the relationship between severity of Depression and the clinical Score may vary over the range of the assessment scale, this style of analysis may not be suited to Depression prediction. The proposed framework on the other hand, imposes a series of partitions on the feature space, with each partition corresponding to a distinct predefined range of Depression Scores, and predicts the Score based on measures of membership to each partition. This approach provides additional flexibility by allowing different rankings to be learnt for different Depression Scores, and relaxes assumptions made by conventional regression approaches. Results demonstrate the framework's suitability for Depression Score prediction: different 2-stage implementations, based on heterogeneous feature extraction and modelling approaches, produce state-of-the-art results on the AVEC-2013 dataset. It is also demonstrated that, unlike fusion of conventional regression systems, the fusion of two-stage systems consistently improves prediction performance.
-
weighted pairwise gaussian likelihood regression for Depression Score prediction
International Conference on Acoustics Speech and Signal Processing, 2015Co-Authors: Nicholas Cummins, Julien Epps, Vidhyasaharan Sethu, Jarek KrajewskiAbstract:This paper presents a technique in which feature vectors are mapped onto ordinal ranges of clinical Depression Scores using weighted pairwise Gaussians. The position of a test vector with respect to these partitions is used to perform Depression Score prediction. Results found on a set of spectral and formant based speech characteristics indicate the potential of this technique for performing Depression Score prediction. Key results on the AVEC 2013 development set indicate that the inclusion of weights and Bayesian adaptation improves system performance by 16.5% – 18.5% when compared to using an unweighted non-adapted system. Fusing results from Bayesian adapted models corresponding to different feature spaces offers up to 8% further improvement. Further, fusion consistently improves performance on both the AVEC 2013 development and test set, in contrast to conventional regressor fusion.
-
ICASSP - Weighted pairwise Gaussian likelihood regression for Depression Score prediction
2015 IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP), 2015Co-Authors: Nicholas Cummins, Julien Epps, Vidhyasaharan Sethu, Jarek KrajewskiAbstract:This paper presents a technique in which feature vectors are mapped onto ordinal ranges of clinical Depression Scores using weighted pairwise Gaussians. The position of a test vector with respect to these partitions is used to perform Depression Score prediction. Results found on a set of spectral and formant based speech characteristics indicate the potential of this technique for performing Depression Score prediction. Key results on the AVEC 2013 development set indicate that the inclusion of weights and Bayesian adaptation improves system performance by 16.5% – 18.5% when compared to using an unweighted non-adapted system. Fusing results from Bayesian adapted models corresponding to different feature spaces offers up to 8% further improvement. Further, fusion consistently improves performance on both the AVEC 2013 development and test set, in contrast to conventional regressor fusion.
