Traditional cancer staging relies on tumor size & spread, but it misses key patient factors. DeepMerkel, an AI-powered tool, provides personalized survival predictions for Merkel Cell Carcinoma.

A hybrid machine learning approach for the personalized prognostication of aggressive skin cancers
Go to source).
‘Merkel cell carcinoma (MCC) is a rare but deadly skin cancer. A new AI-powered tool, DeepMerkel, improves survival predictions by analyzing tumor and patient-specific factors. #CancerTreatment #PrecisionMedicine’

MCC prognostication relies on tumor size and spread but this method often underestimates survival by ignoring other key factors such as patient demographics and socioeconomic status.




To address this, an international team of researchers combined machine learning with clinical expertise to develop DeepMerkel, a web-based AI tool that predicts treatment outcomes based on personal and tumor-specific features. The researchers suggest that this system could also be applied to other aggressive skin cancers improving clinical decision-making and giving patients better treatment choices.
DeepMerkel integrates deep learning and XGBoost a machine learning technique to provide personalized, time-dependent survival prediction. It is based on a broader set of factors like age, ethnicity, income level and tumor characteristics.
It demonstrates higher accuracy than traditional tumor staging systems. The model distinguished survival probabilities based on individual patient features, emphasizing the potential of AI-driven precision medicine.
They also developed a real-time web-based survival calculator tool that helps clinicians predict individual survival rates. By tailoring prognosis to each patient, this innovation could lead to better treatment planning and improved patient outcomes.
Advertisement
Reference:
- A hybrid machine learning approach for the personalized prognostication of aggressive skin cancers - (https://www.nature.com/articles/s41746-024-01329-9)
Source-Medindia