Giorgio Berardini Email and Phone Number

Banking Risk Engineering team hire #1

As part of the Center For Neurological Imaging (CNI) at Brigham and Women's Hospital (teaching affiliate of the Harvard Medical School), I worked on my master's thesis project by designing and developing a Machine Learning pipeline capable of detecting particular brain structures in MRI scans. Abstract:"Most interpretations of medical images have been performed by expert raters such as radiologists. However, the lack of a common agreement between various raters and the intrinsic complexity of medical images creates big challenges, making it difficult to give an accurate diagnosis in a short amount of time. Deep Learning has been widely used for many medical analysis applications allowing great progression in computer-aided diagnosis in recent years; nevertheless, the set-up of a Deep Learning environment often requires a wide range of skillsets, which may not be possessed by a single or a group of scientists. A medical image platform called SPINE aims to help scientists to run statistical experiments, even though it has never been used for Deep Learning applications. The aim of this work, is to design and deploy the first Deep Learning experiment in SPINE, allowing scientists and radiologists to run their experiments without prior Computer Science knowledge. In particular, the focus will be on detecting enlarged perivascular spaces, a type of brain structure often associated with diseases such as Alzheimer's and Multiple Schelorsis. "Technologies: Docker, Python, Git (Continuous Integration), CouchDB (NO-SQL)

Taught Python programming language to first-year students

- Successfully integrated and deployed text extraction tool to compare what the user ordered (i.e., a set of database records) with what the user received (i.e., a delivery document PDF); the information are extracted from a PDF and matched against the database records to ensure that the order was successful- Developed a Content Management System for Raffaele SPA (manifarcture company)- Added features on the frontend of FatturaPA.com - a platform that manages over five million invoices per yearTechnologies: Angular, WPF, C#, Azure Cloud, Git

- Performed quantitative analysis in a range of 5 years to draw key conclusions for the sales department - Collected client's business requirements and processes through documents analysis- Developed ETL proceduresTools used: Pentaho Data Integration (ETL), Pentaho Server, Pentaho Aggregator Designer, Mondrian, Saiku Analytics, PostgreSQL, MS Server, Mondrian, ApicotDB

Awards:- 3rd place Italian Cadet Karate Championship - Ostia 2012- 3rd place Italian Cadet Karate Championship per region - Ostia 2014- 3rd place Mediterranean Trophy – Siracusa 2012- Two times Athlete of the year – Cosenza 2013 and Cosenza 2015- Four Consecutive times Region Champion – January 2012(cadet), October 2013(cadet), October 2014(junior and cadet)