Aerial Mycelium Process Development
Ecovative Design
2023–2026
Aerial mycelium grows readily in a lab and unpredictably at production scale.
I developed the cultivation process end to end: the growth media across both phases, the pre-colonization and aerial growth stages, and the environmental controls that drive morphology and yield. I built the instrumentation underneath it as well, with custom sensors and microcontrollers feeding real-time data acquisition and closed-loop feedback holding the setpoints. I also contributed to a platform tracking more than 25 real-time conditions, where models read that data and recommended setpoints for an operator to accept or override; the control loops then held whatever setpoint was chosen. The process became repeatable enough to run at manufacturing scale. The work led to co-inventorship on a patent.
Mycelium Cultivation Process Development Closed-loop Control Instrumentation Real-time Sensing Data Acquisition Patent
Partner Technology Transfer & Field Process Engineering
Ecovative Design
2023–2026
A process only its authors can run is not yet a process. Moving one onto someone else's floor, staff and equipment is where most of them break.
I transferred the aerial mycelium process to partner facilities in Ontario, in Kennett Square, Pennsylvania, and in Ammerzoden in the Netherlands, and stood up production at each. That covered the recipes, the environmental control programs and the sensing and control systems, plus operations training and hands-on troubleshooting, and often taking direct control of partner lines through start-up. I trained operators in their own language and wrote the documentation the sites kept using after each engagement. I facilitated one Pennsylvania site's conversion from Agaricus mushroom production to full mycelium production inside a year, on site through the switch, and ran the root-cause analysis on a novel contamination event, tracing it to operator practice, inputs and environmental conditions. Data pipelines fed results from partner lines back into our own models, so what happened on their floor informed the next experiment on ours.
Technology Transfer Field Process Engineering Operations Training Data Pipelines International Partners Scale-up
Quantitative MRI for Tumor Treatment Response
Weill Cornell Medical College / NYU
2020–2023
Telling whether a tumor is responding to treatment usually means taking a biopsy.
I built and validated MRI methods to measure cellular water exchange, and with it treatment response, in glioma models without one. As first author in Scientific Reports, I showed that a two-flip-angle DCE-MRI approach can quantify intracellular water lifetime (τᵢ), and that the measure correlates negatively with FDG-PET uptake.
DCE-MRI Pharmacokinetic Modeling Cancer Biology Quantitative Imaging Scientific Reports
Image Texture Analysis & Reproducibility in Preclinical MRI
Weill Cornell Medical College
2021–2023
Radiomics features are only as trustworthy as the images under them, and reproducibility is a known problem in the field.
I quantified how image resolution affects the reliability of texture features derived from pharmacokinetic parameter maps. As first author in Tomography, I demonstrated that 3D isotropic resolution is critical for reproducible radiomics analysis at 7T, which informs how preclinical imaging protocols are standardized and makes quantitative biomarkers more robust across studies.
Radiomics Image Analysis Reproducibility 7T MRI Preclinical Imaging Tomography
Active Contrast Encoding (ACE) MRI Development
NYU / Weill Cornell
2020–2022
Characterizing a tumor microenvironment means separating contrast-agent kinetics from tissue relaxation, which conventional DCE-MRI does inefficiently.
I helped develop and validate ACE-MRI, which estimates both at once. I presented this work at multiple ISMRM annual meetings, and it was published in a collaborative journal article.
ACE-MRI Method Development Tumor Microenvironment ISMRM Collaboration
