Fig. 7: Machine learning models predict presence minimal residual disease in canine osteosarcoma. | Laboratory Investigation

Fig. 7: Machine learning models predict presence minimal residual disease in canine osteosarcoma.

From: Development of an exosomal gene signature to detect residual disease in dogs with osteosarcoma using a novel xenograft platform and machine learning

Fig. 7

Post-treatment samples (test set) from dogs with osteosarcoma (n = 24) were classified as “osteosarcoma=detectable” or “osteosarcoma-not detectable” based on predictions from A the four best-performing machine learning models (KNN, BAG, RF, and EXT), or B the three most sensitive machine learning models (LR, LDA, RDG). Kaplan–Meier survival curves demonstrating time to relapse for subset of dogs with osteosarcoma with available survival data, comparing those whose post-treatment samples were classified as “osteosarcoma-detectable” with those whose post-treatment samples were classified as “osteosarcoma-not detectable”, A p = 0.1001; B p = 0.0398.

Back to article page