MiDATA’s supporting the development of AI solutions through the launch of the MiStats+ML working group. The MiStats+ML working group represents the ideal blend of insight and expertise from the domains of statistics, medical imaging, and bioinformatics (machine learning). The MiStats+ML working group will support all department medical image-based research in all relevant domains. The team is composed of three aspects representing medical imaging, machine learning and statistical analysis:
Medical Imaging: Develops and tests experimental planning and image analysis tools as related to the investigation of developmental disorders and disease prevention.
Machine Learning: Assesses measurement reliability, conducts large-scale medical imaging analyses for the development of clinical diagnostic tests. Designs and trains image classification and regression algorithms. MiStats+ML is partnered with 16Bit for education, consultation, and commercialization.
Research Methodology and Statistical Analysis: Investigates finite mixture models, clustering and classification. Tests dynamic systems, applies simulations, numerical, topological and functional data analyses.
Weekly MiStats+ML Discussion Sessions
“Join us for our weekly open mic sessions every Wednesday evenings at 6PM where you will have the opportunity to ask questions and participate in discussions about biostatistics and machine learning for medical imaging”
Previous Discussion Sessions
July 29 2020 – “Dealing with non-inferiority and bioequivalence in Mi research”.
August 5 2020 – “Intro to dimension reduction”
August 12 2020 – Open session
August 19 2020 – “Using Cluster Analysis to Assess the Impact of Dataset Heterogeneity on Deep Convolutional Network Accuracy”
August 26 2020 – Open session
September 2 2020 – On Vacation!