Marcelo Cicconet, Ph.D
Dr. Cicconet's early contributions -- while at the Instituto de Matemática Pura e Aplicada and the Georgia Institute of Technology -- are in Human-Computer Interaction with applications in music. There he developed computer vision and machine learning approaches to analyze audio and video signals, independently and simultaneously, and worked on new interfaces for music production, performance, and education.
At New York University -- first at the Center for Genomics and Systems Biology, and later at the Courant Institute of Mathematical Sciences -- he worked on computational analysis of time-lapse and high-throughput image data, and developed low-level image analysis methods based on wavelets, complex-number voting schemes, and symmetry. He helped releasing a mouse embryo database for cell tracking and division detection as well as a mirror-symmetry database for the development of symmetry detection algorithms.
At Harvard Medical School, Dr. Cicconet continued to apply CV and ML techniques to help biology researchers with various image analysis tasks (such as segmentation, tracking, registration, and morphology analysis), both in 2D and 3D, light and electron microscopy. He also further released software and databases to the community (e.g. for nuclei segmentation and image near-duplicate identification).
Dr. Cicconet is currently the data science lead at deliberate.ai, working on multimodal AI-based assessment of psychiatric and neurological health.