LDR   07074ngm^^22003613a^4500
001        AA00000299_00001
005        20210519092825.0
006        m^^^^^o^^i^^^^^^^^
007        cr^^na---ma^mp
008        210427n^^^^^^^^xx^nnn^^^^^^^^o^^^^ueng^d
245 00 |a Biomedical and Health Informatics Part II: Graduate Scholarship |h [electronic resource].
260        |c 04/14/2021.
520 3    |a Integrative Survival Analysis of Breast Cancer with Gene Expression and DNA Methylation Data by Guanghui Liu, Isabelle Bichindaritz. Survival analysis has currently become a hot topic because it has been proven to be useful for understanding the relationships between patients’ variables and covariates (e.g. clinical and genetic features) and the effective-ness of various treatment options. Integrative multi-feature fusion analysis on biomedical data has gained much attention recently. Here, we present an adaptive multi-task learning method, which combines the Cox loss task with the ordinal loss task, for survival prediction of breast cancer patients using multi-modal learning instead of performing survival analysis on each feature data set. First, we use feature selection algorithm to reduce the mRNA and methylation feature dimensions and extract cluster eigengenes respectively. Then, we add an auxiliary ordinal loss to the original Cox model to improve the ability to optimize the learning process in training and regularization. The auxiliary loss helps to reduce the vanishing gradient problem for earlier layers and helps to decrease the loss of the primary task. Meanwhile, we use an adaptive weights approach to multi-task learning which weighs multiple loss functions by considering the homoscedastic uncertain-ty of each task. Finally, we build an ordinal cox hazards model for survival analysis and use long short-term memory (LSTM) method to predict patients’ survival risk. We use the cross-validation method and the concordance index (C-index) for assessing the prediction effect. Stringent cross-verification testing processes for the benchmark data set and two additional datasets demonstrate that the developed approach is effective, achieving very competitive performance with existing approaches.
520 3    |a A Mobile-Coach Higher Education Institution Research Platform Prototype Incorporating Diversity, Equity, and Inclusion: The Healthcare-as-a-Service Platform Initiative Using Lean Six Sigma and AI with SBIR/STTR Programs by Ty Austin. How can mobile health (mHealth)-based Federal Small Business Innovation Research (SBIR) and Small Business Technology Transfer (STTR) programs be of value to biomedical data science (BDS) programs seeking diversity, equity and inclusion (DEI), while searching SBIR.gov data with keywords related to covid-19, diabetes, digestion, kidney and cardiovascular diseases (C2DKCDs), and a value proposition goal to produce better higher education institution (HEI) student services with an AI chatbot developed using using the Lean Six Sigma (LSS) methodology? This project sought to leverage the Centre for Digital Health Interventions' Mobile-Coach.eu (MC) platform - a novel, award winning open source behavioral intervention platform - as the foundation for creating a HEI platform design and prototype, called MC-HEIDEIP, to work with SBIR and STTR programs, their data and sustainability at a county level. The project focus was seeking to incorporate an AI chatbot with diversity, equity and inclusion (DEI) objectives using the LSS methodology in the SUNY Oswego Biomedical and Health Informatics (BHI) program. It reviewed a 2020 BHI LSS DEI project, and examined the program’s obstacles, failures, successes, data analysis, and potential effects on population health disparities and sustainability with DNA-based innovations from Perceptual Control Theory, Glasser Choice Theory, and Foucault’s ideas on genealogy. This project proposed the study of mHealth in BDS programs in New York with selected counties in other states, and the effect on population health disparities and its sustainability. In particular, it sought to use asynchronous mHealth learning and applied research about C2DKCDs over a 3-year period. It also included articles and documents from other organizations that are currently using mHealth as a way to further elaborate and support SBIR/STTR submissions and awards near, or within, a HEI campus environment.
520 3    |a DNA Methylation for the Diagnosis and Characterization of Psychiatric Disorders by Bharat Yaddanapalli and Paola Marin. Psychiatric disorders are characterized by a combination of abnormal thoughts, perceptions, emotions, behavior, and relationships with others. They include depression, bipolar disorder, schizophrenia etc. Globally, an estimated 264 million people are affected by depression, 45 million people with bipolar disorder, 20 million people with schizophrenia. The burden of mental disorders continues to grow with significant impacts on health and major social, human rights, and economic consequences in all the countries of the world. Diagnosis of these psychiatric disorders and differentiating them among themselves is a major concern as the diagnosis is majorly based on the immense knowledge of the physician to analyze the symptoms based on what the patients portray and on their own observations, so the diagnosis won’t be correct 100% of the time and this affects the treatment of the patient as each disorder have different medications and treatments. Hence a valid and accurate analysis method is required to differentiate between these disorders. DNA methylation a biological procedure that happens in the body by virtue affects the gene expression and can used as a standard technique to differentiate between cancer genes and normal genes, recently it has been found out that it can also play a major role in differentiating the different types of psychiatric disorders as divergent DNA methylation positions will cause different effects. To analyze this large data, it is important to use machine learning and other techniques and this vast data is certainly exceedingly difficult to study. Here in this project, I have used R and Python to evaluate the data and got appropriate results that can help in easy identification of the distinct psychiatric disorders.
520 3    |a Session Chair: Isabelle Bichindaritz
533        |a Electronic reproduction. |c SUNY Oswego Institutional Repository, |d 2021. |f (Oswego Digital Library) |n Mode of access: World Wide Web. |n System requirements: Internet connectivity; Web browser software.
535 1    |a SUNY Oswego.
541        |a Collected for SUNY Oswego Institutional Repository by the online self-submittal tool. Submitted by Zach Vickery.
650        |a Quest 2021.
650        |a Biomedical and Health Informatics.
700        |a SUNY Oswego.
700 1    |a Liu, Guanghui. |4 spk
700 1    |a Austin, Ty. |4 spk
700 1    |a Yaddanpalli, Bharat. |4 spk
700 1    |a Marin, Paola. |4 spk
700 1    |a Bichindaritz, Isabelle. |4 spk
830    0 |a Oswego Digital Library.
830    0 |a Quest.
852        |a OswegoDL |c Quest
856 40 |u https://digitallibrary.oswego.edu/AA00000299/00001 |y Electronic Resource
992 04 |a https:/digitallibrary.oswego.edu/content/AA/00/00/02/99/00001/Project Poster - Vincent Preikstasthm.jpg
997        |a Quest


The record above was auto-generated from the METS file.