Development of a low-cost robotized 3D-prototype for automated optical microscopy diagnosis : an open-source system
Dantas de Oliveira, Allisson (Universitat Politècnica de Catalunya. Departament de Física)
Rubio Maturana, Carles (Universitat Autònoma de Barcelona. Departament de Genètica i de Microbiologia)
Zarzuela Serrat, Francesc (Hospital Universitari Vall d'Hebron. Institut de Recerca)
Carvalho, Bruno Motta de (Universidade Federal do Rio Grande do Norte. Departamento de Informática e Matemàtica Aplicada)
Sulleiro Igual, Elena (Hospital Universitari Vall d'Hebron. Institut de Recerca)
Prats, Clara (Universitat Politècnica de Catalunya. Departament de Física)
Veiga, Anna (Probitas Foundation)
Bosch, Mercedes (Probitas Foundation)
Zulueta, Javier (Probitas Foundation)
Abelló, Alberto (Universitat Politècnica de Catalunya. Departament d'Enginyeria de Serveis i Sistemes d'Informació)
Sayrol, Elisa (Universitat Pompeu Fabra)
Joseph-Munné, Joan (Hospital Universitari Vall d'Hebron. Institut de Recerca)
López i Codina, Daniel (Universitat Politècnica de Catalunya. Departament de Física)

Data:	2024
Resum:	In a clinical context, conventional optical microscopy is commonly used for the visualization of biological samples for diagnosis. However, the availability of molecular techniques and rapid diagnostic tests are reducing the use of conventional microscopy, and consequently the number of experienced professionals starts to decrease. Moreover, the continuous visualization during long periods of time through an optical microscope could affect the final diagnosis results due to induced human errors and fatigue. Therefore, microscopy automation is a challenge to be achieved and address this problem. The aim of the study is to develop a low-cost automated system for the visualization of microbiological/parasitological samples by using a conventional optical microscope, and specially designed for its implementation in resource-poor settings laboratories. A 3D-prototype to automate the majority of conventional optical microscopes was designed. Pieces were built with 3D-printing technology and polylactic acid biodegradable material with Tinkercad/Ultimaker Cura 5. 1 slicing softwares. The system's components were divided into three subgroups: microscope stage pieces, storage/autofocus-pieces, and smartphone pieces. The prototype is based on servo motors, controlled by Arduino open-source electronic platform, to emulate the X-Y and auto-focus (Z) movements of the microscope. An average time of 27. 00 ± 2. 58 seconds is required to auto-focus a single FoV. Auto-focus evaluation demonstrates a mean average maximum Laplacian value of 11. 83 with tested images. The whole automation process is controlled by a smartphone device, which is responsible for acquiring images for further diagnosis via convolutional neural networks. The prototype is specially designed for resource-poor settings, where microscopy diagnosis is still a routine process. The coalescence between convolutional neural network predictive models and the automation of the movements of a conventional optical microscope confer the system a wide range of image-based diagnosis applications. The accessibility of the system could help improve diagnostics and provide new tools to laboratories worldwide.
Drets:	Aquest document està subjecte a una llicència d'ús Creative Commons. Es permet la reproducció total o parcial, la distribució, la comunicació pública de l'obra i la creació d'obres derivades, fins i tot amb finalitats comercials, sempre i quan es reconegui l'autoria de l'obra original.
Llengua:	Anglès
Document:	Article ; recerca ; Versió publicada
Publicat a:	PloS one, Vol. 19, issue 6 (June 2024) , art. e0304085, ISSN 1932-6203

DOI: 10.1371/journal.pone.0304085
PMID: 38905190

21 p, 3.7 MB

El registre apareix a les col·leccions:
Articles > Articles de recerca
Articles > Articles publicats