Schiz: Difference between revisions
Computational Linguistics and Information Processing
(Created page with "This project, a collaboration with Deanna Kelly [http://www.medschool.umaryland.edu/profiles/Kelly-Deanna/] at the University of Maryland School of Medicine, targets advances...") |
No edit summary |
||
| (4 intermediate revisions by one other user not shown) | |||
| Line 1: | Line 1: | ||
This project, a collaboration with Deanna Kelly | '''Computational Modeling to Identify Symptom Changes in Schizophrenia and Depression | ||
[http://www.medschool.umaryland.edu/profiles/Kelly-Deanna/] | ''' | ||
at the University of Maryland School of | |||
Medicine, targets advances in identifying and monitoring mental illness, an enormous societal challenge. In addition to high costs, continued access to care and regular treatment and monitoring is fragmented, and more than 113 million Americans do not even have ready access to a clinician qualified to perform psychiatric or psychological evaluations. The project brings together Dr. Kelly’s expertise in the treatment and monitoring of severe mental illness, particularly schizophrenia, with Dr. Resnik’s expertise in the use of linguistic analysis and computational modeling of mental status, including work in depression and PTSD. | This project, a collaboration with Deanna Kelly [http://www.medschool.umaryland.edu/profiles/Kelly-Deanna/] | ||
at the University of Maryland School of Medicine, targets advances in identifying and monitoring mental illness, an enormous societal challenge. In addition to high costs, continued access to care and regular treatment and monitoring is fragmented, and more than 113 million Americans do not even have ready access to a clinician qualified to perform psychiatric or psychological evaluations. The project brings together Dr. Kelly’s expertise in the treatment and monitoring of severe mental illness, particularly schizophrenia, with Dr. Resnik’s expertise in the use of linguistic analysis and computational modeling of mental status, including work in depression and PTSD. | |||
Combining within-clinic data collection and a data donation infrastructure built in collaboration with Qntfy [http://qntfy.com] at umd.ourdatahelps.org [http://umd.ourdatahelps.org], we are collecting a unique new dataset that includes clinical variables, within-clinic prompted language responses, and naturally occurring social media interaction. Using this dataset, new computational techniques for predictive modeling re being investigated, with a focus on symptom changes within clinically relevant symptom domains. | |||
This project has been funded by the UMB-UMCP Seed Grant Program. | |||
[https://www.umaryland.edu/ord/resources-for-investigators/find-funding-opportunities/umb-umcp-seed-grant-program/] | |||
Latest revision as of 15:56, 10 December 2017
Computational Modeling to Identify Symptom Changes in Schizophrenia and Depression
This project, a collaboration with Deanna Kelly [1] at the University of Maryland School of Medicine, targets advances in identifying and monitoring mental illness, an enormous societal challenge. In addition to high costs, continued access to care and regular treatment and monitoring is fragmented, and more than 113 million Americans do not even have ready access to a clinician qualified to perform psychiatric or psychological evaluations. The project brings together Dr. Kelly’s expertise in the treatment and monitoring of severe mental illness, particularly schizophrenia, with Dr. Resnik’s expertise in the use of linguistic analysis and computational modeling of mental status, including work in depression and PTSD.
Combining within-clinic data collection and a data donation infrastructure built in collaboration with Qntfy [2] at umd.ourdatahelps.org [3], we are collecting a unique new dataset that includes clinical variables, within-clinic prompted language responses, and naturally occurring social media interaction. Using this dataset, new computational techniques for predictive modeling re being investigated, with a focus on symptom changes within clinically relevant symptom domains.
This project has been funded by the UMB-UMCP Seed Grant Program. [4]