Machine learning for help unveiling neural correlates of Parkinson's Disease

• Grant number: 18/25902-0
• Support Opportunities: Scholarships abroad - Research Internship - Doctorate
• Start date: May 15, 2019
• End date: March 22, 2020
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Roseli Aparecida Francelin Romero
• Grantee: Caetano Mazzoni Ranieri
• Supervisor: Patricia A Vargas
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Institution abroad: Heriot-Watt University, Edinburgh, Scotland
• Associated to the scholarship: 17/02377-5 - Machine Learning and Applications for Robotics in Smart Environments, BP.DR

Abstract

Parkinson's disease is a neurodegenerative disease characterized by progressive impairment of the patient's cognitive and motor functions, which may lead to depression or dementia in later stages. There are no techniques for early diagnosis of this disease, making it recognizable only after significant neural impairment. In addition, it is common that therapies may only be developed with the use of animal models. In this project, a recent database of marmosets monkeys' neural data with and without the disease will be adopted. This data will be used to train models of deep neural networks to extract relevant characteristics and generate an automated diagnosis of the disease. In addition, the characteristics obtained will be used as input to a robotic environment using the iCub robot, which should generate a simple set of behaviors corresponding to the neural signals, serving as the initial robotic model to simulate the symptoms. With this, it is expected to contribute to the elaboration of techniques for earlier diagnosis of the disease and to participate in the construction of a robotic model alternative to the use of animals in the development of therapies.

Scientific publications

(6)

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

RANIERI, CAETANO M.; PIMENTEL, JHIELSON M.; ROMANO, MARCELO R.; ELIAS, LEONARDO A.; ROMERO, ROSELI A. F.; LONES, MICHAEL A.; ARAUJO, MARIANA F. P.; VARGAS, PATRICIA A.; MOIOLI, RENAN C.. A Data-Driven Biophysical Computational Model of Parkinson's Disease Based on Marmoset Monkeys. IEEE ACCESS, v. 9, p. 122548-122567, 2021. (13/07375-0, 17/02377-5, 18/25902-0, 18/11075-5)

PIMENTEL, JHIELSON M.; MOIOLI, RENAN C.; DE ARAUJO, MARIANA F. P.; RANIERI, CAETANO M.; ROMERO, ROSELI A. F.; BROZ, FRANK; VARGAS, PATRICIA A.. Neuro4PD: An Initial Neurorobotics Model of Parkinson's Disease. FRONTIERS IN NEUROROBOTICS, v. 15, JUL 1 2021. (17/02377-5, 18/25902-0)

RANIERI, CAETANO MAZZONI; MACLEOD, SCOTT; DRAGONE, MAURO; VARGAS, PATRICIA AMANCIO; ROMERO, ROSELI APARECIDA FRANCELIN. Activity Recognition for Ambient Assisted Living with Videos, Inertial Units and Ambient Sensors. SENSORS, v. 21, n. 3, FEB 2021. (17/02377-5, 18/25902-0, 17/01687-0, 13/07375-0)

RANIERI, CAETANO M.; MOIOLI, RENAN C.; ROMERO, ROSELI A. F.; DE ARAUJO, MARIANA F. P.; DE SANTANA, MAXWELL BARBOSA; PIMENTEL, JHIELSON M.; VARGAS, PATRICIA A.; IEEE. Unveiling Parkinson's Disease Features from a Primate Model with Deep Neural Networks. 2020 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), v. N/A, p. 8-pg., 2020-01-01. (18/25902-0, 17/02377-5, 13/07375-0)

RANIERI, CAETANO M.; MOIOLI, RENAN C.; VARGAS, PATRICIA A.; ROMERO, ROSELI A. F.. A neurorobotics approach to behaviour selection based on human activity recognition. COGNITIVE NEURODYNAMIC, v. N/A, p. 20-pg., 2022-09-27. (18/25902-0, 21/10921-2, 13/07375-0, 17/02377-5, 17/01687-0)

RANIERI, CAETANO M.; VARGAS, PATRICIA A.; ROMERO, ROSELI A. F.; IEEE. Uncovering Human Multimodal Activity Recognition with a Deep Learning Approach. 2020 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), v. N/A, p. 8-pg., 2020-01-01. (17/02377-5, 13/07375-0, 17/01687-0, 18/25902-0)