Magnetic　particle　imaging　(MPI)　technology　can　generate　a　real-time　magnetic　nanoparticle　(MNP)　distribution　image　for　biological　tissues,　and　its　use　can　overcome　the　limitations　imposed　in　magnetic　hyperthermia　treatments　by　the　unpredictable　MNP　distribution　after　the　intratumoral　injection　of　nanofluid.　However,　the　MNP　concentration　distribution　is　generally　difficult　to　be　extracted　from　MPI　images.　This　study　proposes　an　approach　to　extract　the　corresponding　concentration　value　of　each　pixel　from　an　MPI　image　by　a　least　squares　method　(LSM),　which　is　then　translated　as　MNP　concentration　distribution　by　an　interpolation　function.　The　resulting　MPI-based　concentration　distribution　is　used　to　evaluate　the　treatment　effect　and　the　results　are　compared　with　the　ones　of　two　baseline　cases　under　the　same　dose:　uniform　distribution　and　MPI-based　distribution　considering　diffusion.　Additionally,　the　treatment　effect　for　all　these　cases　is　affected　by　the　blood　perfusion　rate,　which　is　also　investigated　deeply　in　this　study.　The　results　demonstrate　that　the　proposed　method　can　be　used　to　effectively　reconstruct　the　concentration　distribution　from　MPI　images,　and　that　the　weighted　LSM　considering　a　quartic　polynomial　for　interpolation　provides　the　best　results　with　respect　to　other　cases　considered.　Furthermore,　the　results　show　that　the　uniformity　of　MNP　distribution　has　a　positive　correlation　with　both　therapeutic　temperature　distribution　and　thermal　damage　degree　for　the　same　dose　and　a　critical　power　dissipation　value　in　the　MNPs.　The　MNPs　uniformity　inside　biological　tissue　can　be　improved　by　the　diffusion　behavior　after　the　nanofluid　injection,　which　can　ultimately　reflect　as　an　improvement　of　treatment　effect.　In　addition,　the　blood　perfusion　rate　considering　local　temperature　can　have　a　positive　effect　on　the　treatment　compared　to　the　case　which　considers　a　constant　value　during　magnetic　hyperthermia.　©　2023　Chinese　Physical　Society　and　IOP　Publishing　Ltd.