BEGIN:VCALENDAR VERSION:2.0 METHOD:PUBLISH PRODID:-//Telerik Inc.//Sitefinity CMS 14.4//EN BEGIN:VTIMEZONE TZID:Mountain Standard Time BEGIN:STANDARD DTSTART:20231102T020000 RRULE:FREQ=YEARLY;BYDAY=1SU;BYHOUR=2;BYMINUTE=0;BYMONTH=11 TZNAME:Mountain Standard Time TZOFFSETFROM:-0600 TZOFFSETTO:-0700 END:STANDARD BEGIN:DAYLIGHT DTSTART:20230301T020000 RRULE:FREQ=YEARLY;BYDAY=2SU;BYHOUR=2;BYMINUTE=0;BYMONTH=3 TZNAME:Mountain Daylight Time TZOFFSETFROM:-0700 TZOFFSETTO:-0600 END:DAYLIGHT END:VTIMEZONE BEGIN:VEVENT DESCRIPTION:\n\nHossein Estiri\, PhD \;\nAssistant Professor\nDirector\ , Clinical Augmented Intelligence (CLAI) Group \nMassachusetts General Hos pital Harvard Medical School\nTemporal Machine Learning for Phenotype Mode ling with EHR Data\nAs a social scientist\, data scientist\, and clinical informaticist\, his most recent work has focused on architecting visual an alytics application to explore data quality in Electronic Health Records d ata and characterize patients\, using Statistical Learning techniques and Data Science methodologies.\nHe has developed informatics systems and expl ainable machine learning methodologies for improving the care for patients by enhancing knowledge discovery from the massive amounts of clinical dat a. A considerable portion of his research in clinical informatics has been on understanding the issues in clinical data and exploring the possibilit ies for their secondary use in research and medical knowledge discovery. W hile much of the machine learning in healthcare has been steered by comput er scientists\, who often prioritize prediction accuracy over interpretabi lity\, my research focuses on developing computational methods that priori tize clinician interpretability over prediction.\nDr. Estiri has developed the novel thinking learning machine learning framework (MLHO) that enable s adaptive knowledge discovery from clinical data for the identification o f temporal diagnostic/prognostic biomarkers of health outcomes. The tempor al methodology he has developed for cohort identification and prediction w ith EHR data demonstrates clear evidence of the feasibility of the approac hes to model complex evolving phenotypes\n\n DTEND:20230309T200000Z DTSTAMP:20240329T065212Z DTSTART:20230309T190000Z LOCATION: SEQUENCE:0 SUMMARY:DBMI Seminar: Hossein Estiri\, PhD UID:RFCALITEM638472703322129648 X-ALT-DESC;FMTTYPE=text/html:
\n
\nHossein Estiri\, Ph
D \;
\nAssistant Professor
\nDirector\, Clinical Au
gmented Intelligence (CLAI) Group
\nMassachusetts General Hospital H
arvard Medical School
Temporal Machine Learning for Phenot
ype Modeling with EHR Data
\nAs a social scientist\, data sc
ientist\, and clinical informaticist\, his most recent work has focused on
architecting visual analytics application to explore data quality in Elec
tronic Health Records data and characterize patients\, using Statistical L
earning techniques and Data Science methodologies.
He has develope d informatics systems and explainable machine learning methodologies for i mproving the care for patients by enhancing knowledge discovery from the m assive amounts of clinical data. A considerable portion of his research in clinical informatics has been on understanding the issues in clinical dat a and exploring the possibilities for their secondary use in research and medical knowledge discovery. While much of the machine learning in healthc are has been steered by computer scientists\, who often prioritize predict ion accuracy over interpretability\, my research focuses on developing com putational methods that prioritize clinician interpretability over predict ion.
\nDr. Estiri has developed the novel thinking learning machine
learning framework (MLHO) that enables adaptive knowledge discovery from c
linical data for the identification of temporal diagnostic/prognostic biom
arkers of health outcomes. The temporal methodology he has developed for c
ohort identification and prediction with EHR data demonstrates clear evide
nce of the feasibility of the approaches to model complex evolving phenoty
pes
\n
\n