PICTURE

AI-driven prediction of pathological complete response after neoadjuvant therapies in non-small cell lung cancer through multimodal data fusion.

PICTURE

Clinical Application

PICTURE

AI-driven prediction of pathological complete response after neoadjuvant therapies in non-small cell lung cancer through multimodal data fusion.

Pathological Response Prediction Multimodal Learning Explainable AI

Overview

PICTURE addresses one of the central questions in contemporary lung oncology: whether pathological complete response after neoadjuvant treatment can be predicted before surgical resection. The project assumes that a multimodal view of the patient offers a more faithful representation of treatment response than any single data source taken in isolation.

The project develops AI systems that integrate radiology imaging, histology, cytology, molecular data, and electronic health records to support pre-surgical decision-making in NSCLC. Alongside predictive performance, PICTURE emphasizes robustness, explainability, and the possibility of transferring learned models toward chemoimmunotherapy scenarios.

Research Directions

  • Fusion of radiology, histology, cytology, molecular data, and electronic health records.
  • Prediction of pathological complete response before surgery in NSCLC.
  • Robust and explainable multimodal deep learning pipelines, including transfer toward chemoimmunotherapy settings.

Related Publications

  1. A Systematic Benchmark of GAN Architectures for MRI-to-CT Synthesis
    Alessandro Pesci, Valerio Guarrasi, Marco Alì, Isabella Castiglioni, and Paolo Soda
    arXiv preprint arXiv:2603.13520, 2026
  2. SPARSE data, rich results: Few-shot semi-supervised learning via class-conditioned image translation
    Manni, Guido, Clemente Lauretti, Loredana Zollo, and Paolo Soda
    2026
    oncology generative AI clinical prediction
  3. Not another imputation method: A transformer-based model for missing values in tabular datasets
    Camillo Maria Caruso, Paolo Soda, and Valerio Guarrasi
    2026
    multimodal learning clinical prediction foundation models
  4. Learning from Limited and Incomplete Data: A Multimodal Framework for Predicting Pathological Response in NSCLC
    Alice Natalina Caragliano, Giulia Farina, Fatih Aksu, Camillo Maria Caruso, Claudia Tacconi, Carlo Greco, Lorenzo Nibid, Edy Ippolito, Michele Fiore, Giuseppe Perrone, and 3 more authors
    arXiv preprint arXiv:2603.15100, 2026
  5. Pathologic Complete Response Prediction with Machine Learning Using Hierarchical Attention Feature Extraction
    Domenico Paolo, Ciro Russo, G.I. Russo, Carlo Greco, Alessio Cortellini, Marco Russano, Sara Ramella, Paolo Soda, Claudio Marrocco, Alessandro Bria, and 1 more author
    2025
    radiomics oncology clinical prediction
  6. Multi-scale texture loss for CT denoising with GANs
    Francesco Di Feola, Lorenzo Tronchin, Valerio Guarrasi, and Paolo Soda
    2025
    generative AI radiomics
  7. MARIA: A multimodal transformer model for incomplete healthcare data
    Camillo Maria Caruso, Paolo Soda, and Valerio Guarrasi
    2025
    multimodal learning foundation models clinical prediction
  8. Multimodal Doctor-in-the-Loop: A Clinically-Guided Explainable Framework for Predicting Pathological Response in Non-Small Cell Lung Cancer
    Alice Natalina Caragliano, Claudia Tacconi, Carlo Greco, Lorenzo Nibid, Edy Ippolito, Michele Fiore, Giuseppe Perrone, Sara Ramella, Paolo Soda, and Valerio Guarrasi
    2025
    radiomics oncology multimodal learning
  9. Doctor-in-the-Loop: An explainable, multi-view deep learning framework for predicting pathological response in non-small cell lung cancer
    Alice Natalina Caragliano, Filippo Ruffini, Carlo Greco, Edy Ippolito, Michele Fiore, Claudia Tacconi, Lorenzo Nibid, Giuseppe Perrone, Sara Ramella, Paolo Soda, and 1 more author
    2025
    explainability oncology medical imaging
  10. Can Foundation Models Really Segment Tumors? A Benchmarking Odyssey in Lung CT Imaging
    Elena Mulero Ayllón, Massimiliano Mantegna, Linlin Shen, Paolo Soda, Valerio Guarrasi, and Matteo Tortora
    2025
    foundation models medical imaging
  11. Multi-stage intermediate fusion for multimodal learning to classify non-small cell lung cancer subtypes from CT and PET
    Fatih Aksu, Fabrizia Gelardi, Arturo Chiti, and Paolo Soda
    2025
    radiomics oncology multimodal learning
  12. Enhancing NSCLC Histological Subtype Classification: A Federated Learning Approach Using Triplet Loss
    Fatih Aksu, Ermanno Cordelli, Fabrizia Gelardi, Arturo Chiti, and Paolo Soda
    2025
    radiomics oncology medical imaging
  13. Toward a Multimodal Deep Learning Approach for Histological Subtype Classification in NSCLC
    Fatih Aksu, Fabrizia Gelardi, Arturo Chiti, and Paolo Soda
    2024
    oncology multimodal learning medical imaging