Ahmed Hosny

GROW School for Oncology and Developmental Biology, Maastricht University ↗︎

2016 - (2021)

Maastricht, The Netherlands

Position based at Dana-Farber Cancer Institute. Deep learning applications in quantifying tumor phenotypes for survival, recurrence, and treatment response prediction of lung cancer patients from imaging data. Identifying molecular insights and genomic associations of captured radiographic phenotypes. Benchmarking against clinical gold standards as well as machine learning models built on clinical parameters and engineered radiomic features.

Graduate School of Design, Harvard University ↗︎

2013 - 2015

Cambridge, MA

Thesis "Material Analytics: Knowledge from Mechanical Behavior Data" studied the behavior of sensor-embedded cloud-connected physical objects as a function of human interaction, through tracking and analyzing their mechanical features. Coursework in Computational Design, Data Science, Digital Media, Digital Interfaces, Computational Solid & Structural Mechanics. Teaching Assistant @ Computational Design @ Material Distributions. Research Assistant @ Geometry Lab.

College of Architecture, Art and Design, American University of Sharjah ↗︎

2004 - 2009

Sharjah, UAE

5 year professional program. US National Architectural Accrediting Board (NAAB) accredited. Sheikh Khalifa merit-based scholarship recipient. Hong Kong study course.

Dana-Farber Cancer Institute, Harvard Medical School ↗︎

09.2016 -

Boston, MA

Position concurrent with PhD course. Collaborate with medical oncology researchers and professionals to identify clinically relevant questions, and seek to train machine learning models to answer them. Gather, curate, wrangle, and clean medical imaging datasets for lung (CT) and brain (MR) cancer studies, as well as associated genomic profiles. Perform literature review and identify state-of-the-art methods and practices. Select appropriate deep learning strategies, architectures, and perform hyper parameter optimization. Author and publish scientific manuscripts and opinion pieces. Contribute to and document open source software projects: PyRadiomics: a python package for extracting radiomic features from medical images, and Modelhub: a deployment infrastructure providing a containerized standard template for robust and ultra-fast reproducibility of published deep learning models. Develop and maintain web-based applications and pages: radiomics.io and medicalindex.ai: a platform for tracking, curating, and visualizing data on AI research in health care. Aid in procuring and maintaining GPU clusters. Assist in lab branding as well as outreach through visuals, instructional videos, and explainer animations.

2018 Assembly Cohort, MIT Media Lab & Berkman Klein Center, Harvard University ↗︎

01.2018 - 05.2018

Cambridge, MA

Selected as one of 20 participants in this program. Researched current debates in machine bias. Identified, from ethical and public interest perspectives, problematic instances in the machine learning model development pipeline. Developed dataset "nutrition label" prototypes: a diagnostic framework for dataset interrogation providing an overview of "ingredients" to data scientists, practitioners, and the public.

Mediated Matter Group, MIT Media Lab ↗︎

06.2016 - 04.2017

Cambridge, MA

Preformed 3d printing material characterization: optical transparency, bulk modulus, tensile strength, and contact angle measurements.

Wyss Institute for Biologically Inspired Engineering, Harvard University ↗︎

08.2014 - 09.2016

Cambridge, MA

Studied geometry-function relationships in biological composites (chiton scales & stingray jaws and teeth) and designed synthetic analogs through parametric geometry modeling. Developed multi-material bitmap 3d printing workflows for various medical imaging modalities. Designed and fabricated patient aortic phantoms from cardiac CT as well as hardware for sizing heart valve replacements allowing for preoperative In vitro simulations. Modeled skull base defects for endoscopic endonasal surgeries and fabricating patient-specific prosthesis.

Surgical Planning Laboratory, Brigham & Women’s Hospital ↗︎

10.2015 - 07.2016

Boston, MA

Designed and fabricated camera mounts for VR headsets. Tested mixed reality navigation environments for image-guided laparoscopy surgery.

Altis Labs ↗︎

06.2019 -

Boston, MA

Help develop studies as well as data annotation and modeling strategies. Advise executive, engineering, data science, and clinical teams. Participate in business development activities and meetings with clinical and pharmaceutical partners. Assist in recruiting candidates.

Foster + Partners ↗︎

03.2011 - 07.2013

Beijing, China

Designed and managed the 36000m² cladding and 4000 ton steel roof truss packages. Led fabrication optimization and computational design efforts. Developed roof panelization system and associated details with cladding subcontractor and suppliers. Ran steel structure vs cladding clash detection simulations through Building Information Management (BIM). Provided architectural support in engineering a hydraulic track system that slid the steel structure 180 meters over onto its plinths. Resolved on-site design issues with Local Design Institute (LDI).

Playze ↗︎

01.2010 - 01.2011

Shanghai, China

Designed and developed the conversion of 350 freight containers into a 70 room hotel, reception lobby, office building within a packaging warehouse, and organic vegetable market. Coordinated with container fabricators for bespoke structural elements, openings, and stairs. Assigned tasks and led project team. Organized daily plans and communication. Reviewed structural, mechanical, and electrical drawings.

Internships

2006 - 2014

2014: Case (Acquired by WeWork), New York, NY - 2009: James Toomey Architects, Abu Dhabi - 2008: Aedas, Dubai - 2007: Besix construct + Road and Transport Authority (RTA), Dubai & Burthill (Acquired by Stantec), Dubai - 2006: Naga Architects, Dubai