Machine learning for rhabdomyosarcoma histopathology

Emerging innovations and advancements in the treatment of extremity and truncal soft tissue sarcomas

“…8 Beyond imaging, AI models have led to the development of convolutional neuronal networks that have successfully classified sarcoma subtypes, showing high accuracy in STS among pediatric and young adults. 9 In a multicenter trial, Foersch et al utilized 506 histological slides from 291 patients with STS to develop a deep-learning model that accurately diagnosed common STS pathologies, the most common being leiomyosarcoma. 10 Machine learning was also shown to define sarcoma classifications for diagnosis by utilizing tumor methylation profiles.…”

“…Another application included utilizing algorithms to read MR images, which successfully predicted the grading of some STS 8 . Beyond imaging, AI models have led to the development of convolutional neuronal networks that have successfully classified sarcoma subtypes, showing high accuracy in STS among pediatric and young adults 9 . In a multicenter trial, Foersch et al utilized 506 histological slides from 291 patients with STS to develop a deep‐learning model that accurately diagnosed common STS pathologies, the most common being leiomyosarcoma 10 .…”

Artificial intelligence applications in pediatric oncology diagnosis

“…The AI-based techniques on radiological images have been only applied in extremity tumors and the diagnostic potential of AI-based strategies needs further exploration in other tumors, such as abdominal and chest tumors. Zhang et al [55] and Frankel et al [56] developed DL-CNN differential diagnosis system with an AUC of 0.889 for pre-pathologist screening and quantifying diagnosis likelihood of trained soft-tissue sarcoma subtypes based on whole histopathology tissue slides. Considering the limited worldwide availability of sarcoma pathology expertise, this AI-based approach suggested assistance for local pathologists to quickly narrow the differential diagnosis of the sarcoma subtype in children, adolescents, and young adults.…”

How engineers and drug developers are working to change childhood cancer’s deadly calculus

“…Berlow co-developed an automated screening tool that ranks by diagnosis likelihood all the possible cancers a person could have. This helps pathologists to decide quickly which confirmatory tests to conduct, and could prove especially useful in rural areas and developing countries that lack pathologists 2 . Trained on 424 tissue slides of sarcoma tumours, the model is more than 88% effective at detecting all tested sarcoma subtypes.…”