Computer aid system for the diagnosis of psychiatric disorders based on facial anthropometric measurements

• Grant number: 20/01992-0
• Support Opportunities: Research Grants - eScience and Data Science Program - Regular Program Grants
• Start date: October 01, 2020
• End date: March 31, 2024
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Ariane Machado Lima
• Grantee: Ariane Machado Lima
• Host Institution: Escola de Artes, Ciências e Humanidades (EACH). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated researchers: Fátima de Lourdes dos Santos Nunes Marques ; Helena Paula Brentani ; Hélio Pedrini
• Associated scholarship(s): 22/08065-3 - Desenvolvimento de aplicativo para coleta de imagens faciais para o auxílio ao diagnóstico de transtornos do neurodesenvolvimento, BP.IC

Abstract

Neurodevelopmental disorders are psychiatric disorders, usually multifactorial, that are triggered during the neurodevelopment. Evidence shows that this fact also causes facial alterations in relation to a control group. Such alterations can be used to build a system to aid in the diagnosis of such disorders based on facial anthropometric measurements. Such a system would have the advantage of assisting early diagnosis, since the traditional diagnosis is based on information about the child's behavior, which generally postpones the diagnosis to school age. However, due to the characteristics of cerebral neuroplasticity, the earlier the diagnosis and the beginning of interventions, the greater the effectiveness of the treatment for improving the child's quality of life. Preliminary results from our pilot project indicated 80% accuracy in identifying autism versus control. In this pilot, the images were captured by a semi-professional camera and using a reference object to normalize the images. A user of such system should follow a detailed protocol for capturing the photograph, download the image file on a computer and run the classification program. Such a procedure is very tiresome in the clinical practice. In this project we propose the evolution of the system in two aspects: 1) inclusion of other disorders in order to detect comorbidities (more than one disorder in the same individual) 2) simplification of the system usage. For this second aspect we intend to develop a smartphone application where the user can scan the child's face and, when the application identifies a suitable image in terms of image quality and pose, captures the photograph, performs the processing and shows the result. (AU)

Scientific publications

(4)

(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)

FEITOSA, RAYSSA M. M. W.; PRIETO-OLIVEIRA, PAULA; BRENTANI, HELENA; MACHADO-LIMA, ARIANE. MicroRNA target prediction tools for animals: Where we are at and where we are going to-A systematic review. COMPUTATIONAL BIOLOGY AND CHEMISTRY, v. 100, p. 14-pg., 2022-07-31. (18/18560-6, 20/01992-0)

DE ARAUJO, HERMON FARIA; NUNES, FATIMA L. S.; MACHADO-LIMA, ARIANE; ACM. The impact of different facial expression intensities on the performance of pre-trained emotion recognition models. 37TH ANNUAL ACM SYMPOSIUM ON APPLIED COMPUTING, v. N/A, p. 8-pg., 2022-01-01. (20/01992-0, 14/50889-7)

CARDOSO, THIAGO, V; MICHELASSI, GABRIEL C.; FRANCO, FELIPE O.; SUMIYA, FERNANDO M.; PORTOLESE, JOANA; BRENTANI, HELENA; MACHADO-LIMA, ARIANE; NUNES, FATIMA L. S.; ALMEIDA, JR; GONZALEZ, AR; et al. Autism Spectrum Disorder diagnosis based on trajectories of eye tracking data. 2021 IEEE 34TH INTERNATIONAL SYMPOSIUM ON COMPUTER-BASED MEDICAL SYSTEMS (CBMS), v. N/A, p. 6-pg., 2021-01-01. (20/01992-0, 14/50889-7)

OLIVEIRA, JESSICA S.; FRANCO, FELIPE O.; REVERS, MIRIAN C.; SILVA, ANDREIA F.; PORTOLESE, JOANA; BRENTANI, HELENA; MACHADO-LIMA, ARIANE; NUNES, FATIMA L. S.. Computer-aided autism diagnosis based on visual attention models using eye tracking. SCIENTIFIC REPORTS, v. 11, n. 1, MAY 12 2021. (11/50761-2, 14/50889-7, 20/01992-0)