Our team has been developing and applying methods for the full and partial synthesis of longitudinal synthetic data for some time now. Recently a paper describing the principles behind the approach that we took has been published. Because of the time it takes to publish, usually these articles appear a year or more after the work is done, but it is out now.
Also, my team at the EHIL lab at the University of Ottawa has been working on some machine learning and synthetic data generation problems. You will start seeing some of this work in the next few months as the projects deliver their?final results.
We have been delivering our popular?Practical Methods of De-identification and Anonymization for Health Data?Training Course in different locations including New York, San Francisco, Dubai, Toronto, and Barcelona.?If you would like to join one of our future training courses on this topic, please?let us know.
- ?“A multi-disciplinary approach on a complex topic, interspersed with real-life scenarios. The instructors are very experienced and authority figures in the field.”?- Data Architect, IT company
- ?“As an executive it is very important to know about these methodologies and best practices in the field of data”?– Senior Executive,?AI?and Data Authority
- ?“It's effortless to follow whatever the instructor explains, even for someone less familiar with this field. The practical, real-world examples woven into each explanation make it plain that they have years of experience. I got a lot out of the course!”?- Senior Data Scientist, Global Pharmaceutical company
- “An excellent 1.5-day course with skilled instructors who have published extensively on the topic. The best part was being able to apply what we learned using sample datasets in various exercises. Highly recommend!!!”?- HEOR & RWE Director, Biotech
- “This course presents not only the concepts and methods relevant to healthcare data de-identification, but also includes multiple hands-on exercises to help the attendees to gain better understanding of the subject. The instructor is very knowledgeable and presents the materials clearly and effectively.”?- Principal Data Analyst, International Health Data Provider
- On April 20 we issued a press release,?Replica Synthesis unveils Replica Synthesis 3.0. In the press release we announced that our privacy and utility preserving synthetic data generation software has been updated with an enhanced user experience, making it easier for analysts to train generative models and evaluate their utility and privacy. We unveiled Replica Synthesis 3.0 during a Privacy Enhancing Technologies (PETs) demonstration at the Privacy Symposium in Venice, mentioned in the events section below. Read the press release here.
- In late March, BMC Medical Research Methodology published our new paper,?A Method for Generating Synthetic Longitudinal Health Data. Produced by the Replica team in collaboration with the University of Alberta and Health Cities, we?assessed the feasibility of generating and sharing synthetic administrative health data using a recurrent deep learning model. The conclusion was that attribution disclosure risk was substantially less than the typical acceptable risk threshold. Results also show the synthetic dataset was suitably?similar to?the real data. Get the paper via our Knowledgebase here.
- We are collaborating with?OneTrustDataGuidance?on a new series of introductory articles about synthetic data generation intended for an international audience. If you are a?OneTrust?subscriber, you can read the first article,?What?is Synthetic Data and How is it Generated?, at this link. We have also republished the article on our?website?and you can read it here.?We will share updates on the other articles in the series in due course.??
- This month we participated in the Privacy Symposium in Venice, Italy (April 17-21). The event brings together?data protection authorities, professionals, experts, and researchers to discuss developments in data protection regulations, compliance, and innovative technologies. On April 17,?Lucy Mosquera joined the?International Cooperation and Medical Data Sharing?panel and on April 20, she unveiled our Replica Synthesis 3.0 software and it’s enhanced user interface during a?Privacy Enhancing Technologies (PETs) Demo. Find more information on our role at the event here.
- We were?exhibiting at the?Healthcare Information Management Systems Society?(HIMSS)?conference running April 17-21 in Chicago.?HIMSS23 unites thought leaders, disruptors, and changemakers across the global health information and technology spectrum.?Replica was part of the Government of Ontario’s pavilion and you can read about it here.
- On April 24-25 we delivered our training course,?Practical Anonymization Methods for Health Data, in Barcelona, Spain. The event is now sold out.?Participants have been keen to learn multiple practical and innovative data transformation techniques, such as risk-based methods and privacy-preserving synthetic data generation, to help comply with HIPAA and GDPR requirements. The course also focuses on privacy risk assessment and management methods and exercises. See information on the Barcelona course here and the other training courses we have been organizing here.
- ?We are looking forward to the IAPP Canada Privacy Symposium in Toronto late May. On May 25 I will participate in the panel breakout discussion,?More Than a PET Project?with Fahad Diwan from EY, Teresa?Scassa?from the University of Ottawa, and Christopher Parsons from the Office of the Information and Privacy Commissioner of Ontario. On May 26 I will be onstage for the closing general session,?The Next Privacy Challenge: Optimizing Data While Protecting Fundamental Rights, alongside federal Privacy Commissioner Philippe Dufresne, Chief Statistician Anil?Anora, Ontario Information and Privacy Commissioner Patricia Kosseim, and Chantal Bernier, Co-Chair of Dentons’ Global Privacy and Cybersecurity Group. Read more about the sessions here.
- On April 11 we delivered?a?webinar titled?A?Review of the New?Standard on De-identification: ISO/IEC 27559, which offers consistent process for meeting regulatory requirements for de-identification or anonymization which can be applied globally. We had many people join, suggesting this is a topic of great interest. If you missed the webinar, you?can?watch the replay here.
- On April 3rd?I delivered an introductory?Training Session: De-identification?in collaboration with the Future of Privacy Forum. This was a pre-con organized parallel with the IAPP Global Privacy Summit in Washington, D.C., where I was pleased to attend sessions and connect with so many members of the privacy community to discuss practical, responsible solutions to data access and sharing challenges. If we did not connect there, hopefully I will see you at IAPP Canada.
- In the week leading up to the Global Privacy Summit, I participated as panelist in the International Association of Privacy Professionals (IAPP) LinkedIn Live event,?Privacy Tech in Health Care and Medical Research, as part of their?PrivTechTalks series. You can watch the March 30 event recording here.
- In mid-March we delivered a Drug Information Association (DIA) webinar,?Privacy Protective Sharing of Health Datasets using De-identification and Synthetic Data Generation. This was in the lead-up to our exhibit at the DIA Europe 2023 conference in Basel, Switzerland which was March 22-24. If you missed our DIA webinar,?you?can?now watch it at this link.
- On March 9th?we held an EHIL webinar,?Synthetic Data Use: Exploring Use Cases to Optimize Data Utility. Stef James from AstraZeneca anchored the presentation on a recent paper that describes use cases for synthetic data generation in the pharmaceutical industry. The webinar recording is now available online and you can watch it at this link.
If you would like to receive the monthly newsletter directly from Khaled via email, you can also subscribe via the Contact page on the Replica website.?https://replica-analytics.com/contact/
Inspired and Inspiring. Motivated and Motivating.
1 年Hello Khaled. Do you see any application for this in the shipping and energy sector?