- Translational Tensors
- Posts
- The most impactful translational tensor to date
The most impactful translational tensor to date
Applying Neural Networks from Canine Osteosarcoma to Human Osteosarcoma
Deep Domain Adversarial Learning for Species-Agnostic Classification of Histologic Subtypes of Osteosarcoma
Sushant Patkar, Jessica Beck, Stephanie Harmon, Christina Mazcko, Baris Turkbey, Peter Choyke, G Thomas Brown, Amy LeBlanc
![](https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/bef9bf28-c64c-46ef-8419-149cc280fbcf/fabian-gieske-cbIKeuURaq8-unsplash.jpg)
Photo by Fabian Gieske on Unsplash
Use of Machine Learning and Computational Pathology for Osteosarcoma Classification
The article discusses the use of machine learning and computational pathology to classify distinct histologic patterns of osteosarcoma, a rare but aggressive pediatric malignancy affecting bone tissue. Osteosarcoma is unique in its high heterogeneity and scarcity of samples, which makes histology-based prognosis challenging. To overcome this limitation, the study proposes the use of dogs with spontaneous osteosarcoma as a model species due to their strong biological and histologic similarities to humans. The convolutional neural network was trained with canine data to classify distinct histologic patterns of osteosarcoma for humans. The research team achieved an average multiclass F1 score of 0.77 and 0.80 when compared with the ground truth in canines and humans, respectively. They found that their trained model, when used to characterize the histologic landscape of 306 canine osteosarcomas, uncovered distinct clusters with markedly different clinical responses to standard therapy.
![](https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/f4655477-adfc-474f-8acf-09e0e9f001bb/image.png)
Potential Biomarkers and Canine Models for Osteosarcoma Research
The results of this study can serve as a potential biomarker in dogs with osteosarcoma due to its clinical and molecular similarities with human osteosarcoma. The research also sheds light on the potential of AI-derived biomarkers in osteosarcoma histology, supporting the development of further studies in the field of AI and computational pathology. Additionally, naturally occurring canine osteosarcoma has strong biological, molecular, and histologic similarities to human osteosarcoma, making it a useful model species for advancing disease research and treatment. In conclusion, this study highlights the potential of using canines with osteosarcoma as a powerful translational model for histologic subtype classification and potential biomarker identification, contributing to better treatment options and prognosis in both dogs and humans.
![](https://media.beehiiv.com/cdn-cgi/image/fit=scale-down,format=auto,onerror=redirect,quality=80/uploads/asset/file/4d1f62e9-67a3-4080-b387-a72c58d29df4/image.png)
Link to Article: Deep Domain Adversarial Learning for Species-Agnostic Classification of Histologic Subtypes of Osteosarcoma - PubMed (nih.gov)
This article was summarized by an AI tool that uses natural language processing. The tool is not perfect and may make mistakes or produce inaccurate or irrelevant information, but is reviewed by the post’s author prior to publishing. If you want to learn more about the article, please refer to the original source that is cited at the end of the article.