publications

2025

1. MICCAI 2025 (LMID)

   Spatio-Temporal Conditional Diffusion Model for Forecasting Future Multiple Sclerosis Lesion Masks Conditioned on Treatments

   Gian Mario Favero, Ge Ya Luo, Nima Fathi, and 5 more authors

   2025
2. MIDL 2025 (Oral)

   Conditional Diffusion Models are Medical Image Classifiers that Provide Explainability and Uncertainty for Free

   Gian Mario Favero^*, Parham Saremi^*, Emily Kaczmarek, and 2 more authors

   2025

   Abs arXiv Code Website

   Discriminative classifiers have become a foundational tool in deep learning for medical imaging, excelling at learning separable features of complex data distributions. However, these models often need careful design, augmentation, and training techniques to ensure safe and reliable deployment. Recently, diffusion models have become synonymous with generative modeling in 2D. These models showcase robustness across a range of tasks including natural image classification, where classification is performed by comparing reconstruction errors across images generated for each possible conditioning input. This work presents the first exploration of the potential of class conditional diffusion models for 2D medical image classification. First, we develop a novel majority voting scheme shown to improve the performance of medical diffusion classifiers. Next, extensive experiments on the CheXpert and ISIC Melanoma skin cancer datasets demonstrate that foundation and trained-from-scratch diffusion models achieve competitive performance against SOTA discriminative classifiers without the need for explicit supervision. In addition, we show that diffusion classifiers are intrinsically explainable, and can be used to quantify the uncertainty of their predictions, increasing their trustworthiness and reliability in safety-critical, clinical contexts. Further information is available on our project page.

2024

1. ICLR 2025 (Poster)

   Beyond FVD: Enhanced Evaluation Metrics for Video Generation Quality

   Gian Mario Favero^*, Ge Ya Luo^*, Zhi Hao Luo, and 2 more authors

   2024

   Abs arXiv Code Website

   The Frechet Video Distance (FVD) is a widely adopted metric for evaluating video generation distribution quality. However, its effectiveness relies on critical assumptions. Our analysis reveals three significant limitations: (1) the non-Gaussianity of the Inflated 3D Convnet (I3D) feature space; (2) the insensitivity of I3D features to temporal distortions; (3) the impractical sample sizes required for reliable estimation. These findings undermine FVD’s reliability and show that FVD falls short as a standalone metric for video generation evaluation. After extensive analysis of a wide range of metrics and backbone architectures, we propose JEDi, the JEPA Embedding Distance, based on features derived from a Joint Embedding Predictive Architecture, measured using Maximum Mean Discrepancy with polynomial kernel. Our experiments on multiple open-source datasets show clear evidence that it is a superior alternative to the widely used FVD metric, requiring only 16% of the samples to reach its steady value, while increasing alignment with human evaluation by 34%, on average.

2021

1. SAE 2021

   Electrochemical Analysis of High Capacity Li-Ion Pouch Cell for Automotive Applications

   L. Sacchetti, O. Jianu, and G. Favero

   SAE Technical Paper, 2021

   Abs DOI

   Major original equipment manufacturers (OEMs) have already marketed electric vehicles in large scale but apart from business strategies and policies, the real engineering problems must be addressed. Lithium-ion batteries are a promising technology for energy storage; however, their low energy density and complex electro-chemical nature, compared to fossil fuels, presents additional challenges. Their complex nature and strong temperature dependence during operation must be studied with additional accuracy, capable to predict their behavior. In this research, a pseudo two dimensional (P2D) electro-chemical model, for a recent high capacity NMC pouch cell for automotive applications is developed. The electrochemical model with its temperature dependent parameters is validated at high, low, and reference temperature within 10°C to 50°C temperature range. For each temperature various discharge C-rates to accurately replicate the battery cell operational conditions. The overall goodness of the model is proven with limited RMS errors in all the cases. Low temperatures and high C-rates are discovered to limit sensibly the battery performances. The complete analysis provides valuable design considerations for the battery thermal management system (BTMS) to enhance performance, cycle life and safety of future electrified vehicle energy storage systems.