Our first and current research project, the Peter Moss Acute Myeloid & Lymphoblastic Leukemia AI Research Project, was created in 2018 by Adam Milton-Barker after his grandfather, Peter Moss, was diagnosed with Acute Myeloid Leukemia (AML), given a few weeks to live.
The immediate challenge for Peter's family was finding information and advice that could help them understand how to live with the disease. With AML, there are no known warning signs, so early detection is very hard, if not impossible. In Peter's case, the disease was completely missed in a standard blood test 1 month before being diagnosed with AML.
Adam has a number of years experience developing computer vision and natural language understanding applications ocused on real world applications, and had previously created a Computer Vision (Artificial Intelligence) project to detect a form of breast cancer known as Invasive Ductal Carcinoma. BreastCancerAI was where Adam's experience began with using modern technologies for medical purposes. He wanted to see if he could use that experience to build a classifier to detect AML.
Unlike finding open data for breast cancer, computer vision datasets for Acute Myeloid Leukemia are hard to come across. A post made in Artificial Intelligence & Deep Learning (AIDL) group on Facebook directed the team to the Acute Lymphoblastic Leukemia Image Database for Image Processing dataset by Fabio Scotti - Associate Professor Dipartimento di Informatica @ Università degli Studi di Milano. This dataset became the foundation of our research since 2018.
Thanks to donators and sharers of couple of fundraisers created by Adam in honor of Peter, close to a thousand euros was raised for Moffitt Cancer Center in Florida, one of the facilities that advised Peter through his battle with AML. We will ontinue to raise funds each year for Moffitt and The Hematology and Oncology Center in Port St Lucie who treated Peter on a regular basis.
During 2019, the project attracted researchers and developers from throughout the globe and are we started a free student program that allows students to work on real world medical problems in the medical industry.
Through the Intel Software Innovators Program, we were given the opportunity to talk about and present our technology at several events throughout Europe, including Embedded World 2019, Nuremberg, Germany, Intel Dev Affinity Day 2019, Munich, Germany and Codemotion 2019, Madrid, Spain.
Estela Cabezas joined our team in 2019 as a student on our student program and presented our project at Embedded World 2019, Intel Dev Affinity Day 2019 and Codemotion 2019. Estela was able to use our project as case study in her thesis: Applied Analytics for clinical decision support and not long after joined our team as a permanent team member.
In August 2019, after battling AML for one day short of a year since being terminally diagnosed, Peter Moss sadly passed away. In his memory the team committed to the fight against leukemia & other cancers in his name.
Our goal remains to continue providing researchers, students, developers and medical professionals a way to contribute and collaborate towards finding ways to battle Leukemia with technology, as well as giving the public a platform to find open information about leukemia, and speak with our team and professionals.
In January 2020, Peter Moss Leukemia AI Research Association website was created in the anticipation of forming a non-profit. For just over a year the team had volunteered their free time to the Peter Moss Acute Myeloid & Lymphoblastic Leukemia AI Research Project, and we have made great progress in that time. We now have a talented, global team of volunteers, but to really make things happen, we need full-time dedicated members. The next natural step in our journey is to register as a non profit and allow some of our team to work full & part time.
In March 2020 the world was plunged into lockdown due to the COVID-19 (SARS-COV-2) pandemic. Our team quickly began to focus our time on COVID-19 to help understand and fight the pandemic, our second official research project was born: Peter Moss COVID-19 AI Research Project.
On March 18th, Dr Amita Kapoor published an article, COVID 2020 — A data scientist perspective, based on her research on the progression of COVID-19. Soon after this work was referenced in the peer reviewed paper: Covid-19 spread: Reproduction of data and prediction using a SIR model on Euclidean network by Kathakali Biswas, Abdul Khaleque, and Parongama Sen.
Our team continued to work on projects for early detection and data analysis and attracted a number of volunteers that are now part of our permanent team.
On March 28th, President, Adam Milton-Barker was admitted to Parc Tauli Hospital, Sabadell as a potential COVID-19 patient. During his stay in the hospital Adam was confined in the patients due to the lack of beds available in the hospital. Adam was concerned with the amount of time that hospital staff were being exposed to the virus, and an idea began to form for using robots to reduce the risk for medical staff in hospitals that were dealing with the pandemic. From this idea, HIAS - Hospital Intelligent Automation System, EMAR - Emergency Assistance Robot & EMAR Mini were born, a local intelligent network for hospitals and medical centers, and a tele-operated robot designed to reduce medical staff's exposure to contagious diseases such as COVID-19 and other dangerous situations we may face in the future.
Through the Peter Moss COVID-19 AI Research Project we made two very important partnerships.
In April we teamed up with Simeon Pieterkosky CVO and founder at Aquaai, Simeon joined us as Robotics Product Designer to help us take EMAR - Emergency Assistance Robot through the design and manufacturing stages.
In May we teamed up with Plamenlancaster: Professor Plamen Angelov from Lancaster University/ Centre Director @ Lira, & his researcher, Eduardo Soares PhD. PlamenLancaster pioneered eXplainable Deep Learning and created the SARS-COV-2 Ct-Scan Dataset, a large dataset of CT scans for SARS-CoV-2 (COVID-19) identification. Our team are working on open-source, real-world diagnosis systems based on xDNN and the SARS-COV-2 Ct-Scan Dataset to help fight the COVID-19 pandemic.