1. To  receive  a  copy  of  the  database,  Researcher  must sign  this  document,   observing the restrictions listed below.  The signed document should be digitized and sent through email to: Este endereço de email está sendo protegido de spambots. Você precisa do JavaScript ativado para vê-lo. (Marly Costa, Federal University of Amazonas – UFAM, Manaus – Amazonas, Brazil)

  2. Once received a request, we will forward a link to access and download of database.



All the image databases were built by the Research Group in Pattern Recognition and Optimization at the Federal University of Amazonas (henceforth, “RG_PRO-UFAM”). Our research group owns copyright of the images and serves as the unique source for the databases distributed exclusively to research purposes. 

They are available for other research groups to evaluate their algorithms and compare with other proposals developed with the aim of improving the development of systems, in the shortest possible time, to support the automatic diagnosis of tuberculosis.

The digital field images acquired are conventional microscopy images of sputum smears from patients with suspected pulmonary tuberculosis. The sputum smear slides from which the images were acquired were prepared and provided by the project "Evaluation of alternative dyes for the diagnosis of tuberculosis" carried out from 2008 to 2010, approved under protocol number 186/08 by the Ethics Committee for Research with Human Beings - CEP/INPA (National Institute for Research in the Amazon).


Researcher agrees to the following terms and conditions: 

1.              The database, in whole or in part, will not be distributed, published or disseminated in any way. 

2.              The databases may not be  used  to  any  other  purposes  than  academic  research. 

3.              All documents and papers that report  on  research  that  uses  all or portions  of  image databases must include an appropriate citation:  

Serrão, M. K. M., Costa, M.G. F., Fujimoto, L. B. M. , Ogusku, M. M. Costa Filh, C. F. F. Automatic Bright-field Smear Microscopy for Diagnosis of Pulmonary Tuberculosis, 2024 Computers in Medicine and Biology,  accepted for publication feb. 15, 2024

 4.              A copy  of  all  reports  and  papers  that  use  the database and are for public or general release must be forwarded immediately upon release  or publication to  email address: Este endereço de email está sendo protegido de spambots. Você precisa do JavaScript ativado para vê-lo.  


Download the document to request