A5017818529- Topic Modeling
- Natural Language Processing Techniques
- Voice and Speech Disorders
- Machine Learning in Healthcare
- Dementia and Cognitive Impairment Research
- Functional Brain Connectivity Studies
- Mental Health via Writing
- Neurotransmitter Receptor Influence on Behavior
- Cannabis and Cannabinoid Research
- Substance Abuse Treatment and Outcomes
- Parkinson's Disease Mechanisms and Treatments
- Pain Management and Treatment
- Infant Health and Development
- Cognitive Functions and Memory
- Neurological disorders and treatments
- Schizophrenia research and treatment
- Recommender Systems and Techniques
- Neurobiology of Language and Bilingualism
- Long-Term Effects of COVID-19
- Psychosomatic Disorders and Their Treatments
- Text Readability and Simplification
- Lexicography and Language Studies
- Sentiment Analysis and Opinion Mining
- Misinformation and Its Impacts
- Complex Network Analysis Techniques
IBM (United States)
2018-2025
IBM Research - Thomas J. Watson Research Center
2020
University of Rochester
2013
The aim of this study is to use classification methods predict future onset Alzheimer's disease in cognitively normal subjects through automated linguistic analysis.To performance as an early biomarker AD, we performed predictive modeling diagnosis AD from a baseline Framingham Heart Study participants. variables were derived written responses the cookie-theft picture-description task. We compared with clinical and neuropsychological variables. included 703 samples 270 participants out which...
Prior research has identified associations between social media activity and psychiatric diagnoses; however, diagnoses are rarely clinically confirmed. Toward the goal of applying novel approaches to improve outcomes, using real patient data is necessary. We collected 3,404,959 Facebook messages 142,390 images across 223 participants (mean age = 23.7; 41.7% male) with schizophrenia spectrum disorders (SSD), mood (MD), healthy volunteers (HV). analyzed features uploaded up 18 months before...
This paper highlights the design philosophy and architecture of Health Guardian, a platform developed by IBM Digital team to accelerate discoveries new digital biomarkers development health technologies. The Guardian allows for rapid translation artificial intelligence (AI) research into cloud-based microservices that can be tested with data from clinical cohorts understand disease enable early prevention. connected mobile applications, wearables, or Internet things (IoT) devices collect...
The diagnosis and treatment of psychiatric disorders depends on the analysis behavior through language by a clinical specialist. This is subjective in nature could benefit from automated, objective acoustic linguistic processing methods. integrated approach would convey richer representation patient speech, particularly for expression emotion. In this work, we explore potential prosodic metrics to infer variables predict psychosis, condition which produces measurable derailment tangentiality...
Amyotrophic lateral sclerosis (ALS) is a degenerative disease which causes death of neurons controlling voluntary muscles. It currently assessed with subjective clinical measurements, but it would benefit from alternative surrogate biomarkers that can better estimate progression. This work analyzes speech and fine motor coordination subjects recruited by the Answer ALS foundation using data mobile app. In addition, variables such as speech, writing total ALSFRS-R scores are also acquired...
Foundation models applied to bio-molecular space hold promise accelerate drug discovery. Molecular representation is key building such models. Previous works have typically focused on a single or view of the molecules. Here, we develop multi-view foundation model approach, that integrates molecular views graph, image and text. Single-view are each pre-trained dataset up 200M molecules then aggregated into combined representations. Our validated diverse set 18 tasks, encompassing...
Suicidal rates have been increasing since 2000 according the latest report of Centers for disease control and prevention. Today internet opens a channel where people can communicate information remains registered, making acoustic, semantic syntactic analyses especially appealing to find hidden cues that be used detect signs different mental conditions. Here we analyze poems from poets who committed suicide develop method suicidal signs. We use bipartite graph matching algorithms after data...
We apply multi-rate HMMs, a tree structured HMM model, to the word-alignment problem. Multi-rate HMMs allow us model reordering at both morpheme level and word in hierarchical fashion. This approach leads better machine translation results than morphemeaware that does not explicitly reordering.
Valid biomarkers that can predict longitudinal clinical outcomes at low cost are a holy grail in psychiatric research, promising to ultimately be used optimize and tailor intervention prevention efforts.
The emergence of COVID-19 offered a unique opportunity to study chronic pain patients as they responded sudden changes in social environments, increased community stress, and reduced access care. We report findings from n=70 Spinal Cord Stimulation (SCS) before during initial pandemic stages resulting advances home monitoring artificial intelligence that produced novel insights despite pandemic-related disruptions. From multi-dimensional array frequently monitored signals—including mobility,...
In the hardware technical support domain, scaling technician skills remains a prevalent problem. Given large portfolio of products service providers need to maintain, it is not possible for every be expert at repairing product. Augmented reality addresses this problem through virtual procedures, which are interactive 3D visual representations text-based knowledge articles that describe how perform step-by-step repair actions. Virtual thus, equip technicians with they wide range products....
Academic advances of AI models in high-precision domains, like healthcare, need to be made explainable order enhance real-world adoption. Our past studies and ongoing interactions indicate that medical experts can use systems with greater trust if there are ways connect the model inferences about patients explanations tied back context use. Specifically, risk prediction is a complex problem diagnostic interventional importance clinicians wherein they consult different sources make decisions....