Short abstract of the presentation
Mycetoma is a chronic, neglected infectious disease endemic in tropical and
subtropical areas that may lead to severe disability. By considering the
causative agents, mycetoma is classified into eumycetoma (fungus) and
actinomycetoma (bacteria). The diagnosis strategy relies on the clinical
presentation and the identification of mycetoma causative agents. Accurate
identification of the causative agents is a priority for mycetoma diagnosis.
The current identification tools include molecular techniques, cytology,
histology, and grain culture. Although histopathology is the optimal tool to be
used in endemic areas, it requires expert pathologists for conclusive
identification, which are lacking in endemic rural areas.
With the advent of digital pathology, automated image analysis algorithms can
be used to solve this issue. The main aim of this presentation is to introduce
the novel computational diagnostic model for mycetoma diagnostic using
histopathological microscopic images. The presented model can play a
fundamental role in the non-specialised clinical centres because it reaches an
accuracy comparable to expert pathologists.
Mycetoma diagnosis, Digital Histopathology, Image Analysis, Artificial