A5080030626- Emotion and Mood Recognition
- Speech and dialogue systems
- Mental Health via Writing
- Speech and Audio Processing
- German Literature and Culture Studies
- Speech Recognition and Synthesis
- Music and Audio Processing
- Sentiment Analysis and Opinion Mining
- Social Robot Interaction and HRI
- Communication in Education and Healthcare
- Angiogenesis and VEGF in Cancer
- Digital Mental Health Interventions
- Cancer Treatment and Pharmacology
- Cancer Genomics and Diagnostics
- Colorectal Cancer Treatments and Studies
- Neural Networks and Applications
- Mental Health Research Topics
- Language, Discourse, Communication Strategies
- Language, Metaphor, and Cognition
- Genetic factors in colorectal cancer
- Human Pose and Action Recognition
- Neural Networks and Reservoir Computing
- Voice and Speech Disorders
- Topic Modeling
- Phonetics and Phonology Research
META Health
2023-2024
University of Southern California
2014-2023
Creative Technologies (United States)
2014-2023
Karlsruhe Institute of Technology
2019-2022
Children's Hospital of Los Angeles
2021
Operation PAR
2021
Shriners Hospitals for Children - Erie
2021
American Society of Plastic Surgeons
2021
Southern California University for Professional Studies
2012-2021
Neurobehavioral Systems
2021
Speech processing algorithms are often developed demonstrating improvements over the state-of-the-art, but sometimes at cost of high complexity. This makes algorithm reimplementations based on literature difficult, and thus reliable comparisons between published results current work hard to achieve. paper presents a new collaborative freely available repository for speech called COVAREP, which aims fast easy access facilitating research in field. We envisage that COVAREP will allow more...
The AVAGAST study showed that adding bevacizumab to chemotherapy in patients with advanced gastric cancer improves progression-free survival and tumor response rate but not overall survival. To examine the hypothesis angiogenic markers may have predictive value for efficacy cancer, included a prospective, mandatory biomarker program.Patients previously untreated, locally or metastatic were randomly assigned (n = 387) placebo combination chemotherapy. Blood tissue samples collected at...
The Audio/Visual Emotion Challenge and Workshop (AVEC 2016) "Depression, Mood Emotion" will be the sixth competition event aimed at comparison of multimedia processing machine learning methods for automatic audio, visual physiological depression emotion analysis, with all participants competing under strictly same conditions. goal is to provide a common benchmark test set multi-modal information bring together recognition communities, as well video compare relative merits various approaches...
We present SimSensei Kiosk, an implemented virtual human interviewer designed to create engaging face-to-face interaction where the user feels comfortable talking and sharing information. Kiosk is also interactional situations favorable automatic assessment of distress indicators, defined as verbal nonverbal behaviors correlated with depression, anxiety or post-traumatic stress disorder (PTSD). In this paper, we summarize design methodology, performed over past two years, which based on...
The Audio/Visual Emotion Challenge and Workshop (AVEC 2017) "Real-life depression, affect" will be the seventh competition event aimed at comparison of multimedia processing machine learning methods for automatic audiovisual depression emotion analysis, with all participants competing under strictly same conditions. goal is to provide a common benchmark test set multimodal information bring together recognition communities, as well compare relative merits various approaches from real-life...
A common barrier to healthcare for psychiatric conditions is the stigma associated with these disorders. Perceived prevents many from reporting their symptoms. Stigma a particularly pervasive problem among military service members, preventing them symptoms of combat-related like posttraumatic stress disorder (PTSD). However, research shows increased by members when anonymous assessments are used. For example, report more PTSD they anonymously answer Post-Deployment Health Assessment (PDHA)...
This article describes and evaluates the focus variation method, an optical 3D measurement technique.The goal is to analyse performance of method on a series typical tasks including roughness measurements, form wear measurements.First, comparison measurements between proposed tactile device newly developed standard made.Results show that both systems deliver R values are comparable each other with differences few nanometers.Afterwards performed calibration hemi-spherical calottes, showing...
We evaluated the prognostic and predictive use of circulating VEGF-A levels in phase III trials bevacizumab colorectal cancer, lung renal cell carcinoma.Baseline plasma samples from 1,816 patients were analyzed for using an ELISA, which recognizes major isoforms with equivalent sensitivity. HR 95% confidence intervals (CI) study end points estimated Cox regression analysis. A subset matched archival tumor was expression situ hybridization.Higher showed trends toward adverse significance...
We demonstrate a method for generating 3D virtual character performance from the audio signal by inferring acoustic and semantic properties of utterance. Through prosodic analysis signal, we perform an stress pitch, relate it to spoken words identify agitation state. Our rule-based system performs shallow utterance text determine its semantic, pragmatic rhetorical content. Based on these analyses, generates facial expressions behaviors including head movements, eye saccades, gestures, blinks...
Sayan Ghosh, Mathieu Chollet, Eugene Laksana, Louis-Philippe Morency, Stefan Scherer. Proceedings of the 55th Annual Meeting Association for Computational Linguistics (Volume 1: Long Papers). 2017.
We investigate the capabilities of automatic nonverbal behavior descriptors to identify indicators psychological disorders such as depression, anxiety, and post-traumatic stress disorder. seek confirm enrich present state art, predominantly based on qualitative manual annotations, with quantitative descriptors. In this paper, we propose four that can be automatically estimated from visual signals. introduce a new dataset called Distress Assessment Interview Corpus (DAIC) which includes 167...
Death by suicide demonstrates profound personal suffering and societal failure. While basic sciences provide the opportunity to understand biological markers related suicide, computer science provides opportunities thought markers. In this novel prospective, multimodal, multicenter, mixed demographic study, we used machine learning measure fuse two classes of suicidal markers: verbal nonverbal. Machine algorithms were with subjects’ words vocal characteristics classify 379 subjects recruited...
We seek to investigate voice quality characteristics, in particular on a breathy tense dimension, as an indicator for psychological distress, i.e. depression and post-traumatic stress disorder (PTSD), within semi-structured virtual human interviews. Our evaluation identifies significant differences between the of psychologically distressed participants not-distressed this limited corpus. capability automatic algorithms classify speech speaker-independent experiments. Additionally, we examine...
Reduced frequency range in vowel production is a well documented speech characteristic of individuals with psychological and neurological disorders. Affective disorders such as depression post-traumatic stress disorder (PTSD) are known to influence motor control particular production. The assessment documentation reduced space expressivity often either rely on subjective assessments or analysis under constrained laboratory conditions (e.g. sustained production, reading tasks). These...
Automatically assessing emotional valence in human speech has historically been a difficult task for machine learning algorithms. The subtle changes the voice of speaker that are indicative positive or negative states often "overshadowed" by characteristics relating to intensity activation. In this work we explore representation approach automatically derives discriminative representations speech. particular, investigate two strategies improve classifier performance: (1) utilization...
No validated predictive biomarkers for antiangiogenic treatment of metastatic renal cell carcinoma (mRCC) exist. Tumor vascular endothelial growth factor A (VEGF-A) level may be useful. We determined tumor uptake <sup>89</sup>Zr-bevacizumab, a VEGF-A–binding PET tracer, in mRCC patients before and during pilot study. <b>Methods:</b> Patients underwent <sup>89</sup>Zr-bevacizumab scans at baseline 2 6 wk after initiating either bevacizumab (10 mg/kg every wk) with interferon-α (3–9 million IU...