Multiple Instance Learning (MIL) is a powerful framework for weakly supervised learning, particularly useful when fine-grained annotations are unavailable. Despite growing interest in deep MIL methods, the field lacks standardized tools for model development, evaluation, and comparison, which hinders reproducibility and accessibility. To address this, we present torchmil , an open-source Python library built on PyTorch. torchmil offers a unified, modular, and extensible framework, featuring basic building blocks for MIL models, a standardized data format, and a curated collection of benchmark datasets and models. The library includes comprehensive documentation and tutorials to support both practitioners and researchers. torchmil aims to accelerate progress in MIL and lower the entry barrier for new users. Available at https://torchmil.readthedocs.io .

The Multiple Instance Learning (MIL) paradigm is attracting plenty of attention in medical imaging classification, where labeled data is scarce. MIL methods cast medical images as bags of instances (e.g. patches in whole slide images, or slices in CT scans), and only bag labels are required for training. Deep MIL approaches have obtained promising results by aggregating instance-level representations via an attention mechanism to compute the bag-level prediction. These methods typically capture both local interactions among adjacent instances and global, long-range dependencies through various mechanisms. However, they treat attention values deterministically, potentially overlooking uncertainty in the contribution of individual instances. In this work we propose a novel probabilistic framework that estimates a probability distribution over the attention values, and accounts for both global and local interactions. In a comprehensive evaluation involving eleven state-of-the-art baselines and three medical datasets, we show that our approach achieves top predictive performance in different metrics. Moreover, the probabilistic treatment of the attention provides uncertainty maps that are interpretable in terms of illness localization.

In this paper we propose the use of the Hyperbolic Secant (HS) distribution as a prior for the Blind Image Deconvolution (BID) problem. It is well-known that when high-pass filters are applied to natural images, the resulting coefficients are sparse. We leverage this property using the HS distribution, a seldom explored Super Gaussian distribution with suitable properties for this problem. Using the Pólya-Gamma distribution, we derive an explicit Gaussian Scale Mixture representation. This representation is then used to propose a novel variational Bayesian algorithm that outperforms state-of-the-art BID methods.

Multiple Instance Learning (MIL) is widely used in medical imaging classification to reduce the labeling effort. While only bag labels are available for training, one typically seeks predictions at both bag and instance levels (classification and localization tasks, respectively). Early MIL methods treated the instances in a bag independently. Recent methods account for global and local dependencies among instances. Although they have yielded excellent results in classification, their performance in terms of localization is comparatively limited. We argue that these models have been designed to target the classification task, while implications at the instance level have not been deeply investigated. Motivated by a simple observation -- that neighboring instances are likely to have the same label -- we propose a novel, principled, and flexible mechanism to model local dependencies. It can be used alone or combined with any mechanism to model global dependencies (e.g., transformers). A thorough empirical validation shows that our module leads to state-of-the-art performance in localization while being competitive or superior in classification. Our code is at https://github.com/Franblueee/SmMIL.

Multiple Instance Learning (MIL) is a weakly supervised paradigm that has been successfully applied to many different scientific areas and is particularly well suited to medical imaging. Probabilistic MIL methods, and more specifically Gaussian Processes (GPs), have achieved excellent results due to their high expressiveness and uncertainty quantification capabilities. One of the most successful GP-based MIL methods, VGPMIL, resorts to a variational bound to handle the intractability of the logistic function. Here, we formulate VGPMIL using Pólya-Gamma random variables. This approach yields the same variational posterior approximations as the original VGPMIL, which is a consequence of the two representations that the Hyperbolic Secant distribution admits. This leads us to propose a general GP-based MIL method that takes different forms by simply leveraging distributions other than the Hyperbolic Secant one. Using the Gamma distribution we arrive at a new approach that obtains competitive or superior predictive performance and efficiency. This is validated in a comprehensive experimental study including one synthetic MIL dataset, two well-known MIL benchmarks, and a real-world medical problem. We expect that this work provides useful ideas beyond MIL that can foster further research in the field.

Histological images are often tainted with two or more stains to reveal their underlying structures and conditions. Blind Color Deconvolution (BCD) techniques separate colors (stains) and structural information (concentrations), which is useful for the processing, data augmentation, and classification of such images. Classical BCD methods rely on a complicated optimization procedure that has to be carried out on each image independently, i.e., they are not amortized methods. In contrast, once they have been trained, deep neural networks can be used in a fast, amortized manner on unseen inputs. Unfortunately, the lack of large databases of ground truth color and concentrations has limited the development of deep models for BCD. In this work, we propose a deep variational Bayesian BCD neural network (BCD-Net) for stain separation and concentration estimation. BCD-Net is trained by maximizing the evidence lower bound of the observed images, which does not require the use of ground truth examples of stains and concentrations. Results obtained using two multicenter databases (Camelyon-17 and a stain separation benchmark) demonstrate the effectiveness of BCD-Net in the stain separation tasks, while drastically reducing the computation time compared to classical non-amortized methods.