2021-22 Takeda Fellows: Leaning on AI to advance medicine for humans | MIT News

Cortez Deacetis

In tumble 2020, MIT’s College of Engineering and Takeda Prescription drugs Firm Constrained released the MIT-Takeda Software, a collaboration to assistance users of the MIT local community operating at the intersection of synthetic intelligence and human wellness. Housed at the Abdul Latif Jameel Clinic for Device Understanding in Health, the collaboration aims to use artificial intelligence to each gain human wellbeing and aid in drug development. Combining know-how with chopping-edge wellness research, the program’s members hope to improve health and fitness outcomes throughout the globe.

Consequently considerably, the partnership has supported joint investigate endeavours concentrated on topics these kinds of as automatic inspection in sterile pharmaceutical manufacturing and equipment finding out for liver phenotyping.

Just about every year, the software also cash graduate fellowships to guidance learners pursuing exploration on a wide range of troubles tied to wellness and AI. This year’s Takeda fellows, explained underneath, are working on investigation involving digital well being document algorithms, distant sensing facts as it relates to environmental well being danger, and neural networks for the enhancement of antibiotics.

Monica Agrawal

Agrawal is a PhD pupil in the Office of Electrical Engineering and Personal computer Science (EECS). Her exploration focuses on the growth of device studying algorithms that could unlock the opportunity of electronic health and fitness information to power personalised, genuine-earth studies of comparative performance. She is tackling the problem from three interconnected angles: understanding the basic building blocks of scientific textual content, enabling the structuring of medical timelines with only negligible labeled facts, and redesigning scientific documentation to incentivize substantial-quality structured info at the time of development. Agrawal earned the two a BS and an MS in computer science from Stanford College.

Peng Cao

A PhD pupil in EECS, Peng Cao’s investigate is centered on building a new tactic to checking oxygen saturation by examining the radio frequency alerts that bounce off a person’s body. To this stop, she is extracting respiration alerts from the radio alerts and then schooling a neural network to infer oxygen stages from it. Peng acquired a BS in laptop or computer science from Peking University in China.

Bianca Lepe

A PhD pupil in organic engineering, Bianca Lepe is operating to benchmark current and defining subsequent-era vaccine candidates for tuberculosis. She is employing publicly accessible facts put together with machine mastering algorithms to detect the Mtb proteins that are well-suited as subunit vaccine antigens throughout the diversity of the human leukocyte antigen alleles. Lepe acquired a BS in biological engineering and business from Caltech an MS in techniques and artificial biology from the University of Edinburgh in Scotland and an MPhil in engineering plan from the College of Cambridge in England.

Caroline McCue

Caroline McCue is a PhD scholar in mechanical engineering who is establishing a program that could simplify and velocity up the method of cell passaging. A lot more specially, she is creating and screening a platform that triggers cell detachment in response to straightforward external stimuli, these types of as a change in voltage or in mechanical qualities. She options to examination the efficacy of this system by applying equipment studying to quantify the adhesion of Chinese hamster ovary cells to these surfaces. McCue earned a BS in mechanical engineering from the College of Maryland.

Somesh Mohapatra

A PhD scholar in the Section of Resources Science and Engineering, Somesh Mohapatra is also pursuing an MBA at the MIT Sloan Faculty of Administration as aspect of the Leaders for Global Functions Method. His doctoral investigate, in close collaboration with experimentalists at MIT, focuses on developing biomacromolecules applying interpretable equipment finding out and simulations. Exclusively, Mohapatra leverages macromolecule graph representations to create equipment mastering designs for quantitative prediction, optimization, and attribution methods. He then applies these equipment to elucidate style concepts and to make improvements to efficiency and synthetic accessibility of features macromolecules, ranging from peptides and glycans to electrolytes and thermosets. Mohapatra acquired his BTech in metallurgical and materials engineering from the Indian Institute of Know-how Roorkee in India.

Luke Murray

Luke Murray is a PhD pupil in EECS. He is producing MedKnowts, a process that combines equipment mastering and human pc interaction tactics to reduce the effort demanded to synthesize information for health-related decision-generating, and writer superior-top quality, structured, clinical documentation. MedKnowts unifies these two currently splintered workflows by delivering a seamless interface that re-imagines documentation as a purely natural byproduct of scientific reasoning, somewhat than as a compliance prerequisite. Murray acquired his BS in computer science from Brown University.

Ufuoma Ovienmhada

Ufuoma Ovienmhada SM ’20 is a PhD student in aeronautics and astronautics. Her study employs a combined-procedures approach (community-centered style and design, systems engineering, and device learning) to satellite remote sensing knowledge to generate instruments that evaluate how human wellness hazard relates to environmental dangers. Ovienmhada earned her BS in mechanical engineering from Stanford College and her SM in media arts and sciences from MIT.​

Lagnajit Pattanaik

Lagnajit “Lucky” Pattanaik is a PhD university student in chemical engineering. He seeks to shift the paradigm of predictive organic chemistry from qualitative to quantitative. More specially, his exploration is focused on the enhancement of machine studying approaches for predicting 3D constructions of molecules and reactions, which includes transition point out geometries and the geometrical conformations that molecules consider in remedy. He acquired a BS in chemical engineering from Ohio State University.

Na Sunshine 

A PhD university student in EECS, Na Sunshine is doing work in the rising discipline of neuro-immuno-genomics. More specially, she is establishing equipment studying procedures to far better fully grasp the interactions between two incredibly complex methods: the human brain and its dozens of mobile sorts, and the human immune method and the dozens of biological processes that it integrates across cognition, pathogen response, diet program-exercise-being overweight, and synaptic pruning. Sunshine gained her BS in everyday living sciences from Linyi University in China and an MS in developmental biology from the University of Chinese Academy of Sciences in China.

Jacqueline Valeri

Jacqueline Valeri is a PhD scholar in organic engineering who makes use of neural networks for antibiotics discovery. Her attempts include the recycling of compounds from current compound libraries and the computationally assisted layout of novel therapeutics. She is also energized by broader purposes of equipment understanding and artificial intelligence in the fields of well being treatment and biomedicine. Valeri attained her BSE and MSE in bioengineering from the College of Pennsylvania.

Clinton Wang

A PhD university student in EECS, Clinton Wang SM ’20 has produced a new style of conditional generative adversarial community based on spatial-depth transforms. It achieves superior impression fidelity, is sturdy to artifacts in instruction information, and generalizes to held-out scientific sites. Wang now aims to increase his design to even a lot more tough purposes, like visualizing transformations of focal pathologies, these types of as lesions, wherever it could serve as a effective instrument for characterizing biomarkers of malignancy and therapy reaction. Wang earned a BS in biomedical engineering from Yale College and an SM in electrical engineering and personal computer science from MIT.

Next Post

How Omicron escapes from antibodies | MIT News

A new analyze from MIT implies that the dozens of mutations in the spike protein of the Omicron variant support it to evade all 4 of the classes of antibodies that can concentrate on the SARS-CoV-2 virus that causes Covid-19. This consists of antibodies produced by vaccinated or formerly contaminated […]