Armando Orlandi, Medical Director at the Agostino Gemelli University Hospital Foundation IRCCS, shared a post on LinkedIn about a recent article he and his colleagues co-authored, adding:
“Congratulations to Dr. Luca Mastrantoni for this important contribution to oncology research!
Proud to share the results of our recent study published in ESMO Real World Data and Digital Oncology, comparing machine learning and deep learning models for survival prediction in HR-positive/HER2-negative breast cancer.
Key study findings
- 572 patients with early-stage breast cancer analyzed
- DeepSurv demonstrated the best performance (C-index 0.70 for DFS)
- Key predictors identified: nodal status, hormone receptors, Ki-67
- Traditional ML models achieved comparable performance to deep learning approaches
Clinical implications
This work demonstrates that predictive models based on routine clinicopathological features can provide reliable survival predictions, even with limited dataset sizes. This supports the applicability of accessible predictive tools in everyday clinical practice.
Special thanks to Luca for conceiving and structuring this project with methodological rigor and innovative vision.
The study suggests that simple models can perform as well as complex deep learning architectures in small datasets, with the marginally higher discrimination of DL models needing to be weighed against computational demands and lower interpretability.
‘Sometimes, less is more’ – This research perfectly embodies the principle that rigorous application of well-established clinical factors can be just as powerful as sophisticated algorithms. Why use a sledgehammer when a scalpel will do?
Title: Comparison of machine learning and deep learning models for survival prediction in early-stage hormone receptor-positive/HER2-negative breast cancer receiving neoadjuvant chemotherapy
Authors: L. Mastrantoni, G. Garufi, G. Giordano, N. Maliziola, E. Di Monte, G. Arcuri, V. Frescura, A. Rotondi, A. Orlandi, L. Carbognin, A. Palazzo, L. Pontolillo, A. Fabi, S. Pannunzio, I. Paris, F. Marazzi, A. Franco, G. Franceschini, G. Scambia, D. Giannarelli, G. Tortora, E. Bria
Read the Full Article on ESMO Real World Data and Digital Oncology
More posts featuring Armando Orlandi.