Development and Validation of a New Laboratory Tool for the Remote Computerized Image-Based Diagnosis of Skin Lesions Caused by Leishmania spp.

• Grant number: 18/19245-7
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: December 01, 2018
• End date: November 30, 2021
• Field of knowledge: Health Sciences - Collective Health - Public Health
• Principal Investigator: Jancarlo Ferreira Gomes
• Grantee: Aline do Nascimento Benitez
• Host Institution: Faculdade de Ciências Médicas (FCM). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil
• Associated research grant: 14/12236-1 - AnImaLS: Annotation of Images in Large Scale: what can machines and specialists learn from interaction?, AP.TEM

Abstract

Teledermatology is the distance study and practice of dermatological affections by use of descriptive and audio-visual information. In practice, such information are gathered on field and sent to a reference center for discussion and assistance to professionals on diagnosis and patient management. However, this possibility has effectively focused on commonly affections of tumoral characteristics, for example melanomas and carcinomas. In such scenario, the main objective of the present project will be to develop and validate by unprecedent manner in laboratory a computational teledermatological tool directed to the diagnostic of human cutaneous lesions caused by parasite Leishmania spp., by employment of dermatoscope equipment for the obtention of distance images of such lesions and posterior diagnosis by imagens of this affection by computer in laboratory. Initially, the creation of a specific colored solution, preferentially on fluorescence basis, for application on cutaneous region of this pathology, will be elementary on this study to allow viable detections of these lesions by contrast to attend the computerized reading of images. Following, cutaneous lesions will be investigated in different hosts with clinical suspicious of leishmaniasis, which will be characterized as determinants to this human pathology, with auxiliary serological and molecular confirmations. In this view, an image bank with 400 images, of which 200 images of different cutaneous lesions by Leishmania spp., and 200 images of other different human cutaneous lesions (other correlated anomalies) will be developed with the purpose of making feasible and develop the recognition module of computer patterns. Thus, a validation in laboratory of this new diagnostic tool in different human hosts will be established on distance with clinical suspicion for leishmaniasis. Therefore, 200 images from different hosts will utilized for a qualitative statistical comparative evaluation between this new computational diagnostic tool and conventional laboratory techniques, used as reference in this study.