lazyslide.models.tile_prediction.PathProfilerQC

class PathProfilerQC(model_path=None, token=None)

Bases: TilePredictionModel

GitHub Paper Params: 11.2M GPL-3.0 [Haghighat et al., 2022] Quality assessment of histology images

predict(image)

Predict the class of the input image using the PathProfiler model. The model expects a tensor of shape [B, C, H, W].

Suggested Name | Description |

———————— | —————————————————————— |

diagnostic_quality | Whether the image is usable for diagnosis (1 = good quality) |

visual_cleanliness | Whether the image appears normal and free of artefacts (1 = clean) |

focus_issue | Degree of focus issue (1 = severe, 0.5 = slight, 0 = none) |

staining_issue | Degree of staining issue (1 = severe, 0.5 = slight, 0 = none) |

tissue_folding_present | Whether tissue folding is present (1 = present) |

misc_artifacts_present | Whether other artefacts are present (1 = present) |