@article{fdi:010086665,
  title = {{W}ing {I}nterferential {P}atterns ({WIP}s) and machine learning, a step toward automatized tsetse ({G}lossina spp.) identification},
  author = {{C}annet, {A}. and {S}imon-{C}hane, {C}. and {A}khoundi, {M}. and {H}istace, {A}. and {R}omain, {O}. and {S}ouchaud, {M}. and {J}acob, {P}. and {D}elaunay, {P}. and {S}ereno, {D}. and {B}ouss{\`e}s, {P}hilippe and {G}r{\'e}baut, {P}ascal and {G}eiger, {A}nne and de {B}eer, {C}. and {K}aba, {D}. and {S}ereno, {D}enis},
  editor = {},
  language = {{ENG}},
  abstract = {{A} simple method for accurately identifying {G}lossina spp in the field is a challenge to sustain the future elimination of {H}uman {A}frican {T}rypanosomiasis ({HAT}) as a public health scourge, as well as for the sustainable management of {A}frican {A}nimal {T}rypanosomiasis ({AAT}). {C}urrent methods for {G}lossina species identification heavily rely on a few well-trained experts. {M}ethodologies that rely on molecular methodologies like {DNA} barcoding or mass spectrometry protein profiling ({MALDI} {TOFF}) haven't been thoroughly investigated for {G}lossina sp. {N}evertheless, because they are destructive, costly, time-consuming, and expensive in infrastructure and materials, they might not be well adapted for the survey of arthropod vectors involved in the transmission of pathogens responsible for {N}eglected {T}ropical {D}iseases, like {HAT}. {T}his study demonstrates a new type of methodology to classify {G}lossina species. {I}n conjunction with a deep learning architecture, a database of {W}ing {I}nterference {P}atterns ({WIP}s) representative of the {G}lossina species involved in the transmission of {HAT} and {AAT} was used. {T}his database has 1766 pictures representing 23 {G}lossina species. {T}his cost-effective methodology, which requires mounting wings on slides and using a commercially available microscope, demonstrates that {WIP}s are an excellent medium to automatically recognize {G}lossina species with very high accuracy.},
  keywords = {},
  booktitle = {},
  journal = {{S}cientific {R}eports - {N}ature},
  volume = {12},
  numero = {1},
  pages = {20086 [15 p.]},
  ISSN = {2045-2322},
  year = {2022},
  DOI = {10.1038/s41598-022-24522-w},
  URL = {https://www.documentation.ird.fr/hor/fdi:010086665},
}