Cathy Goldstein

By |

My research is primarily in the use of wearable technologies in the ambulatory setting to evaluate the role of sleep and circadian rhythms in health (including during pregnancy) and chronic disease (multiple sclerosis, gastrointestinal).

What are some of your most interesting projects?

My most interesting project was a collaboration with mathematics (Drs. Walch and Forger). We developed algorithms to model sleep and circadian phase and tested the predictions against gold-standard polysomnogram and dim light melatonin onset. We were the first group to develop our own sleep classifier adapted for use on an off the shelf device and disclose the code as open source.

How did you end up where you are today?

I am fellowship trained in sleep medicine and the completion of my training aligned with the release of smartwatches and fitness trackers that reported sleep. I suspected early on that these devices would have promise in longitudinal sleep research.

Michael Sjoding

By |

Application of machine learning and artificial intelligence in healthcare, particularly in the field of pulmonary and critical care medicine. Deep learning applied to radiologic imaging studies. Physician and artificial intelligence interactions and collaborations. Identifying and addressing algorithmic bias.

Maggie Makar

By |

My research focuses on developing reliable and efficient machine learning methods for causal inference as well as predictive models that leverage causal reasoning. My work typically involves applications to healthcare.

Alexander Rodríguez

By |

Alex’s research interests include machine learning, time series, multi-agent systems, uncertainty quantification, and scientific modeling. His recent focus is on developing trustworthy AI systems that can offer insightful guidance for critical decisions, especially in applications involving complex spatiotemporal dynamics. His work is primarily motivated by real-world problems in public health, environmental health and community resilience.

Jacob Allgeier

By |

My goal as an ecologist is to apply ecological theory to help solve real-world conservation issues. Specifically, I seek to identify the mechanisms by which behavioral, population, and community dynamics mediate nutrient and energy pathways. The objective is to improve our ability to predict ecological outcomes, and enhance conservation efficacy such as the sustainability of ecosystem services (e.g., fisheries). Much of this research takes place in tropical coastal ecosystems (mangroves, seagrass beds, and coral reefs) where I study gradients created by anthropogenic impacts to test theory directly within the context of environmental change and biodiversity loss. My research is broad and multifaceted, and includes a combination of extensive field-based research and computational analyses.The type of data we collect in the field has endless potential to be better understood through collaborations with MIDAS. I rely on (and very much enjoy) integrative collaborations across a variety of fields.

What are some of your most interesting projects?

We have recently generated one of the most extensive high-resolution dataset of fish movement in any system that we are aware of. We are using these data to understand the role of consumers in moving nutrient and energy through these ecosystems, and also to better understand the relative ecological importance of individual-level vs species-level variation.

What is the most significant scientific contribution you would like to make?

I would like to improve our ability to predict fish production in tropical coastal ecosystems to improve food security. I would also like to help improve our ability to manage seagrass ecosystems to maximize carbon sequestration and storage.

What makes you excited about your data science and AI research?

A central goal of my lab is to collect extremely high-end and extensive empirical data such that it can inform models that help us forecast ecological processes at the scales of entire ecosystems. We are currently using a suite of techniques to do so, including the use of individual-based modeling in particular. The type of data we are generating is absolutely ripe for being used with high-power data science and AI research. I honestly believe the applications are endless and would be extremely excited to team up with folks to build on these exciting possibilities.

What are some interesting facts about yourself?

Backpacking and woodworking are how I unwind. I love being outside, and I love doing field work.

Jacob Underwater

Studying our artificial reefs in Haiti

Irina Gaynanova

By |

Dr. Gaynanova’s research focuses on the development of statistical methods for analysis of modern high-dimensional biomedical data. Her methodological interests are in data integration, machine learning and high-dimensional statistics, motivated by challenges arising in analyses of multi-omics data (e.g., RNASeq, metabolomics, micribiome) and data from wearable devices (continuous glucose monitors, ambulatory blood pressure monitors, activity trackers).Dr. Gaynanova’s research has been funded by the National Science Foundation, and recognized with a David P. Byar Young Investigator Award and an NSF CAREER Award. She currently serves as an Associate Editor for Journal of the American Statistical Association, Biometrika and Data Science in Science.

Tian An Wong

By |

Analysis of policing technology and police data, including impact assessment of surveillance technology, media sentiment analysis, and fatal police violence. Methods include topological data analysis, natural language processing, multivariate time series analysis, difference-in-differences, and complex networks.

Cristian Minoccheri

By |

Dr. Minoccheri’s research interests focus on using mathematical tools to enhance existing machine learning methods and develop novel ones. A central topic is the use of tensor methods, multilinear algebra, and invariant theory to leverage higher order structural properties in data mining, classification, and deep learning. Other research interests include interpretable machine learning and transparent models. The main applications are in the computational medicine domain, such as phenotyping, medical image segmentation, drug design, patients’ prognosis.

Ying Xu

By |

Xu’s research is focused on the educational applications of artificial intelligence, in particular, natural language processing and speech technologies. She explores how these conversational technologies play the role of social partners or learning companions for children, and leverages AI to empower teachers to co-create learning resources to support their instructional goals. In addition, Xu’s research also aims to identify and actively challenge biases inherent in AI technologies used for educational purposes, with the goal of making these technologies more responsive and responsible to children, parents, and teachers from diverse backgrounds. To carry out her research, Xu closely collaborates with national media producers, including PBS KIDS and Sesame Workshop, as well as industrial partners and local community organizations. Her work has been supported by funding from the National Science Foundation, Schmidt Futures, and the Corporation for Public Broadcasting.

What are some of your most interesting projects?

Most of my work has been focused on partnering with public media to explore how AI can facilitate more active and educationally beneficial ways for children to engage with digital media. For instance, I collaborated with PBS KIDS to develop interactive television shows that allow children to talk to their favorite characters as they watch STEM-related programs. Think about children spending nearly two hours every day watching television. And considering that public media programs are valuable and also accessible learning resources, especially for children from less privileged backgrounds. If we could transform these hundreds of hours of screentime into active STEM learning experiences, that could have profound implications. We’ve carried out multiple studies to test whether these interactive videos indeed help children learn. One consistent finding is that when having interactions with the media character, children comprehend the science concepts better and are also more motivated to think about science problems than the children who watched the broadcast version that does not have the AI-assisted interactions. I remember testing this interactive program with preschoolers and observing their enthusiastic conversations with Elinor, which is the main character of a show. We also found that, when children used our interactive videos at their homes, their parents were more likely to participate in the discussion with their children. This heightened parent involvement could potentially have lasting impact on children’s STEM learning in the long run. With the support from the National Science Foundation and the Corporation for Public Broadcasting, we are working on making our interactive television shows publicly available on PBS KIDS platforms.

What is the most significant scientific contribution you would like to make?

I hope that my research can clarify some of our questions regarding how AI might impact child development. Specifically: How do children interact with, perceive, and learn from conversational technologies or non-human entities in general? Can these technologies actually become social partners for children? Ultimately, I hope my research can unpack the complex interplay among children, their social contexts, and technology. Only then will we be able to harness the unique learning experiences conversational agents can provide and ensure that this technology is integrated into children’s existing social contexts and relationships in ways that enhance their development.


Accomplishments and Awards

Mark Draelos

Mark Draelos

By |

My work focuses on image-guided medical robots with an emphasis on clinical translation. My interests include medical robotics, biomedical imaging, data visualization, medical device development, and real-time algorithms.

A major ongoing project is the development of robotic system for automated eye examination. This system relies on machine learning models for tracking and eventually for interpretation of collected data. Other projects concern the live creation of virtual reality scenes from volumetric imaging modalities like optical coherence tomography and efficient acquisition strategies for such purposes.