In the field of computational pathology, the use of decision support systems powered by state-of-the-art deep learning solutions has been hampered by the lack of large labeled datasets. Until recently, studies relied on datasets in the order of few hundreds of slides which are not enough to train a model that can work at scale in the clinic. Here, we have gathered a dataset consisting of 12,160 slides, two orders of magnitude larger than previous datasets in pathology and equivalent to 25 times the pixel count of the entire ImageNet dataset. Given the size of our dataset it is possible for us to train a deep learning model under the Multiple Instance Learning (MIL) assumption where only the overall slide diagnosis is necessary for training, avoiding all the expensive pixel-wise annotations that are usually part of supervised learning approaches. We test our framework on a complex task, that of prostate cancer diagnosis on needle biopsies. We performed a thorough evaluation of the performance of our MIL pipeline under several conditions achieving an AUC of 0.98 on a held-out test set of 1,824 slides. These results open the way for training accurate diagnosis prediction models at scale, laying the foundation for decision support system deployment in the clinic.
Making an informed, correct and quick decision can be life-saving. It's crucial for animals during an escape behaviour or for autonomous cars during driving. The decision can be complex and may involve an assessment of the amount of threats present and the nature of each threat. Thus, we should expect early sensory processing to supply classification information fast and accurately, even before relying the information to higher brain areas or more complex system components downstream. Today, advanced convolutional artificial neural networks can successfully solve visual detection and classification tasks and are commonly used to build complex decision making systems. However, in order to perform well on these tasks they require increasingly complex, "very deep" model structure, which is costly in inference run-time, energy consumption and number of training samples, only trainable on cloud-computing clusters. A single spiking neuron has been shown to be able to solve recognition tasks for homogeneous Poisson input statistics, a commonly used model for spiking activity in the neocortex. When modeled as leaky integrate and fire with gradient decent learning algorithm it was shown to posses a variety of complex computational capabilities. Here we improve its implementation. We also account for more natural stimulus generated inputs that deviate from this homogeneous Poisson spiking. The improved gradient-based local learning rule allows for significantly better and stable generalization. We also show that with its improved capabilities it can count weakly labeled concepts by applying our model to a problem of multiple instance learning (MIL) with counting where labels are only available for collections of concepts. In this counting MNIST task the neuron exploits the improved implementation and outperforms conventional ConvNet architecture under similar condtions.
This paper strives for spatio-temporal localization of human actions in videos. In the literature, the consensus is to achieve localization by training on bounding box annotations provided for each frame of each training video. As annotating boxes in video is expensive, cumbersome and error-prone, we propose to bypass box-supervision. Instead, we introduce action localization based on point-supervision. We start from unsupervised spatio-temporal proposals, which provide a set of candidate regions in videos. While normally used exclusively for inference, we show spatio-temporal proposals can also be leveraged during training when guided by a sparse set of point annotations. We introduce an overlap measure between points and spatio-temporal proposals and incorporate them all into a new objective of a Multiple Instance Learning optimization. During inference, we introduce pseudo-points, visual cues from videos, that automatically guide the selection of spatio-temporal proposals. We outline five spatial and one temporal pseudo-point, as well as a measure to best leverage pseudo-points at test time. Experimental evaluation on three action localization datasets shows our pointly-supervised approach (i) is as effective as traditional box-supervision at a fraction of the annotation cost, (ii) is robust to sparse and noisy point annotations, (iii) benefits from pseudo-points during inference, and (iv) outperforms recent weakly-supervised alternatives. This leads us to conclude that points provide a viable alternative to boxes for action localization.
Multiple instance learning (MIL) is a variation of supervised learning where a single class label is assigned to a bag of instances. In this paper, we state the MIL problem as learning the Bernoulli distribution of the bag label where the bag label probability is fully parameterized by neural networks. Furthermore, we propose a neural network-based permutation-invariant aggregation operator that corresponds to the attention mechanism. Notably, an application of the proposed attention-based operator provides insight into the contribution of each instance to the bag label. We show empirically that our approach achieves comparable performance to the best MIL methods on benchmark MIL datasets and it outperforms other methods on a MNIST-based MIL dataset and two real-life histopathology datasets without sacrificing interpretability.
The training of medical image analysis systems using machine learning approaches follows a common script: collect and annotate a large dataset, train the classifier on the training set, and test it on a hold-out test set. This process bears no direct resemblance with radiologist training, which is based on solving a series of tasks of increasing difficulty, where each task involves the use of significantly smaller datasets than those used in machine learning. In this paper, we propose a novel training approach inspired by how radiologists are trained. In particular, we explore the use of meta-training that models a classifier based on a series of tasks. Tasks are selected using teacher-student curriculum learning, where each task consists of simple classification problems containing small training sets. We hypothesize that our proposed meta-training approach can be used to pre-train medical image analysis models. This hypothesis is tested on the automatic breast screening classification from DCE-MRI trained with weakly labeled datasets. The classification performance achieved by our approach is shown to be the best in the field for that application, compared to state of art baseline approaches: DenseNet, multiple instance learning and multi-task learning.
Multiple instance (MI) learning with a convolutional neural network enables end-to-end training in the presence of weak image-level labels. We propose a new method for aggregating predictions from smaller regions of the image into an image-level classification by using the quantile function. The quantile function provides a more complete description of the heterogeneity within each image, improving image-level classification. We also adapt image augmentation to the MI framework by randomly selecting cropped regions on which to apply MI aggregation during each epoch of training. This provides a mechanism to study the importance of MI learning. We validate our method on five different classification tasks for breast tumor histology and provide a visualization method for interpreting local image classifications that could lead to future insights into tumor heterogeneity.