Ipek Ensari

Hi, I am Ipek. I am an Assistant Professor within the Department of Artificial Intelligence and Human Health at the Icahn School of Medicine and Hasso Plattner Institute for Digital Health at Mount Sinai. Previously, I was an Associate Research Scientist at the Data Science Institute at Columbia University in New York City. My research program focuses on mHealth and machine learning-based approaches to develop clinically relevant, ecologically valid digital patient-reported outcome measures, and tools for chronic symptom self-management, grounded in women’s health conditions. (photo credit: Pam Pasco @pamashley)

Recent Updates and Announcements:

  • We are looking to collaborate with pelvic floor physical therapists in the NYC area for an observational study with women with pelvic pain disorders. If you are interested in collaborating with our team, you can learn more through the d2prom study website.
  • Our recent work on developing reinforcement learning based personalized exercise recommendations for endometriosis pain management was presented at the ML4Health conference in New Orleans. The paper was presented by the fantastic Dominik Meier at Hasso Plattner Institute Potsdam, whose primary adviser is Dr. Stefan Konigorski. This research has been in the works for a long time and I am very excited that I get to collaborate with an excellent team to conduct the clinical intervention study. The pre-print paper is available on the MLR proceedings website.
  • Postdoctoral fellow Eugenia Alleva, MD, masterfully presented some of the latest research on dysmenorrhea phenotyping via prompt-based learning at AMIA 2023 in New Orleans. She discussed how optimizing prompt-based tuning via keyword-optimized template insertion (KOTI) enables the adaptation of pre-trained clinical language models with only a few training examples. You can check out the pre-print to learn more about how KOTI can be applied to different clinical tasks.

About

ipek ensari

photo credit: Robert Florida

My training background spans both patient-oriented clinical research (in the lab) and data-oriented approaches to ambulatory (out of the lab) measurement, estimation and prediction of physical activity, sleep, physiological outcomes (e.g., heart rate, blood pressure), and psychological outcomes (e.g., stress, depressive symptoms). I evaluate these behavioral outcomes both from the perspective of the patient (e.g., intervention delivery), and from a data architecture perspective (e.g., use of appropriate trackers, extraction and summarization of mobile data). My research revolves around the philosophy of “digital data for good”- maximizing the health benefits provided to the user/patient from their digital data obtained through in-the-wild sensing and self-tracking via mHealth apps. This involves taking a user-centered approach to the design of the mHealth technology, as well as provision of interpretable communication of their digital data. I advise graduate students on wearable/mobile data-related research, and lead various STEM and data science-centered community outreach events (Citizen Science Interactive).  In 2020, I collaborated with Camille Hollet-French, a Vancouver-based filmmaker, to make a film about endometriosis. The end product is ENDOMIC (@abloodycrisis), which had its world premiere at the 2021 Slamdance Film Festival.

Research

In our laboratory we conduct research on methods for combining actively-tracked and passively-collected mHealth data to investigate data methods for designing patient-centered mHealth measures, elucidating elucidating phenotypic and endotypic variation in women’s reproductive disorders (e.g., using electronic health records), and designing mHealth intervention tools for chronic symptom management. We are recruiting graduate students and postdoctoral fellows to various new projects on these topics.

1. Women’s reproductive health, participatory mHealth research

A central aim of my research is to re-tool patient-generated data via mHealth technology to better characterize conditions that are traditionally poorly documented and not well understood

2. Digital phenotyping: Elucidating clinically meaningful variations in patient profiles

I use informatics and data-driven approaches to delineate symptom trajectories in diseases with a dynamic course (e.g., endometriosis, multiple sclerosis), and identification of self-management approaches for their effective management

3. Symptom self-management and computational methods for patient-generated health data:

As a trained kinesiologist, a focus on physical activity and its health outcomes is a central theme across my projects

Read More
Publications

ENDOMIC

An exhaustive meta-analytic review documenting a mysterious “women’s” issue, otherwise known as endometriosis.[1] 

[1]term used to describe a clinical etiology that thus far has only been identified in primates with a female reproductive system, an anatomical structure of decidedly lower importance in comparison to those of the male primate.

In the fall of 2020, I participated in the 13th Annual Imagine Film Festival’s Symbiosis short film competition. I partnered with Camille Hollett-French to make a film that expresses our frustrations about endometriosis from multiple perspectives. Camille is a Vancouver-based filmmaker who herself has endometriosis. The end product is ENDOMIC, a “documentary” through the lens of satire about modern medicine’s and society’s take on endometriosis. Using a mock-meets-doc format, the film weaves in real patient experiences in parallel with a narrative exploration of its history in the scientific literature, disparities in diagnosis and treatment, and lack of research funding, to raise awareness and help sense-making for patients.

Ipek Ensari

Dances with Films 2021

ENDOMIC is unique in several aspects. First, we crafted its satirical story-telling using content from published peer-reviewed research and real patient experiences (even though some of them sound fictional at times!). Next, it features some of my most recent academic work demonstrating how mobile health technology and direct patient input can be used to improve our understanding of endometriosis. The film also discusses inclusive medicine and healthcare, a first in this realm, and about a third of the production and acting crew include members with endometriosis in real life.

You can follow ENDOMIC (@abloodycrisis) on InstagramTwitter, and Facebook. Click below to read more about the film’s world premiere at the 2021 Slamdance Film Festival, media coverage, and reviews.

ENDOMIC

For more information on the background and methodology of the Impact Assessment Scale (IAS) development for Entertainment to Affect Change (E2AC), you can check out the project’s Github repository. The site is regularly updated whenever there is a new iteration of the data analyses to validate the factor structure and other properties of the IAS.

News and Events

Ipek Ensari nyhacker mobile health data talk

  • I gave an invited talk at the Hasso Plattner Institute at Potsdam University on July 13th, 2023. I had a great time meeting with the students and faculty, and visiting the Digital Health Cluster. The recording of the talk is available on the department website.
  • The New Wave of AI in Healthcare conference in NYC organized by the HPIMS took place on May 23rd-24th. The briefs and recordings of the talks are available via the NYAS’s website.
  • I recently came across Endonews, a website that features the latest peer-reviewed research on endometriosis, as well as interviews. It is a good read, with a lay summary and key insights on every featured article.
  • My talk at the nyhackR meetup is now available on youtube. Thank you to everyone who attended in person and virtually! I greatly enjoyed meeting everyone and the questions during the Q&A.
  • We were back at the Columbia University Data Science Day in Spring 2023, held at Lerner Hall on campus. I have a very personal connection to this annual event and always feel thrilled when we get to present our research with this community. This year, I presented some of our latest methodological work on functional data analysis for clustering wearable data for characterizing heterogeneous, longitudinal physical activity patterns.
  • I was awarded an R01 grant by the National Institutes of Health (NICHD) in September. This project will investigate a functional data analytic framework to design and evaluate mHealth measures of pain, quality of life, and treatment response using patient generated health data with high complexity and temporality. This study will test a novel application of Distributed Lag Models (DLMs) in this mHealth context to estimate “critical windows” of tracking analogous to the “window of susceptibility” (Further project details).
  • I spoke at the 2022 New York R Conference in June! I have been attending these events for several years so I was thrilled to be invited as a speaker. The event was held during June 8-10th as a hybrid conference. The recording of my talk on functional clustering using wearable data is available on youtube.
  • ipek ensari columbia data science day functional phenotypingI presented some of my latest work on integrating functional data methods with unsupervised learning for clustering daily physical activity trajectories at DSI’s Annual Data Science Day on April 6th.
  • I had the great pleasure to join the Symbiosis Competition judges panel at this year’s Imagine Science Film Festival. The event was held October 15th-22nd in NYC. Symbiosis is where I met my co-director and partner in crime Camille Hollett-French last year. Moreover, I got to meet the festival directors and organizers in-person for the first time this year. Overall an exciting time. Congrats to this year’s Symbiosis winners Arianna and Rashu!
  • ENDOMIC has been announced among the official selections by the 12th Awareness Film Festival! We are thrilled to be a part of this festival, whose mission is to inform and inspire audiences by spotlighting film-makers committed to making positive change throughout the world. This year’s event will be held October 21st-November 1st in 2021 Los Angeles.
  • My presentation on digital health applications of unsupervised learning at the 2021 Annual DSI Data Science Day is now available online.
  • A piece by Karina Alexanyan on my recent collaboration on ENDOMIC as inspired by my research is now live. It was great pleasure to chat with Karina and I am excited about how the piece turned out.
  • My latest work on digital phenotyping of sleep using unsupervised machine learning was presented at the DSI Health Analytics Poster Session was held on April 2nd, 2021.
  • With my colleagues at Columbia, I was recently awarded seed funds from the Data Science Institute to study digital sleep phenotypes and examine the associations of daily exposure to minority stressors (such as experiences of discrimination and anticipated discrimination) with sleep health among Black and Latinx sexual and gender minority adults. Read more on the aims of the project aims and the DSI Seed Funds
  • Speaking at the DSI Summer Career Webinar Series
  • Virtual Data Science Day 2020 at DSI 
  • Guest Post on Thrive Global: “Making your wearables work for you

Contact

Ipek Ensari nyas For media and consulting inquiries, please reach out via Linkedin or at ipek.ensari@gmail.com. For research-related inquiries, you can contact me at ipek.ensari@mssm.edu. Thank you for visiting my website!