The KM3NeT/ORCA detector (Oscillation Research with Cosmics in the Abyss) is an array of Digital Optical Modules, spheres that host 31 photomultiplier tubes, tied together in vertical structures, the Detection Units, which are anchored on the seabed. Such an array configuration can detect neutrino events from the Cherenkov radiation emitted by the secondary particles induced by neutrino interactions in the abyssal depths of the Mediterranean Sea. The KM3NeT/ORCA
detector is being deployed at a depth of 2450 m offshore Toulon, France with the determination of the Neutrino Mass Ordering being the main physics goal of the detector. The aim of this work is the study of atmospheric neutrinos with energies at the 1-100 GeV range, in order to obtain information on the atmospheric muon neutrino flux in this energy range, in which only few measurements exist by other experiments. An analysis of data collected with the 6-Detection Unit
configuration of KM3NeT/ORCA (KM3NeT/ORCA6) is being presented in this contribution. The data analyzed corresponds to a time period of one and a half year. The procedure for the selection of a high-purity atmospheric neutrino sample, using a Machine Learning classifier (Boosted Decision Tree), is described. Subsequently, an unfolding scheme is used to obtain an estimation of the atmospheric muon neutrino flux in bins of energy in the region of interest.