John Guttag

Kathleen M. Lewis, Emily Mu, Adrian V. Dalca et al.

Recent vision-language models outperform vision-only models on many image classification tasks. However, because of the absence of paired text/image descriptions, it remains difficult to fine-tune these models for fine-grained image classification. In this work, we propose a method, GIST, for generating image-specific fine-grained text descriptions from image-only datasets, and show that these text descriptions can be used to improve classification. Key parts of our method include 1. prompting a pretrained large language model with domain-specific prompts to generate diverse fine-grained text descriptions for each class and 2. using a pretrained vision-language model to match each image to label-preserving text descriptions that capture relevant visual features in the image. We demonstrate the utility of GIST by fine-tuning vision-language models on the image-and-generated-text pairs to learn an aligned vision-language representation space for improved classification. We evaluate our learned representation space in full-shot and few-shot scenarios across four diverse fine-grained classification datasets, each from a different domain. Our method achieves an average improvement of $4.1\%$ in accuracy over CLIP linear probes and an average of $1.1\%$ improvement in accuracy over the previous state-of-the-art image-text classification method on the full-shot datasets. Our method achieves similar improvements across few-shot regimes. Code is available at https://github.com/emu1729/GIST.

Victor Ion Butoi, Jose Javier Gonzalez Ortiz, Tianyu Ma et al.

While deep learning models have become the predominant method for medical image segmentation, they are typically not capable of generalizing to unseen segmentation tasks involving new anatomies, image modalities, or labels. Given a new segmentation task, researchers generally have to train or fine-tune models, which is time-consuming and poses a substantial barrier for clinical researchers, who often lack the resources and expertise to train neural networks. We present UniverSeg, a method for solving unseen medical segmentation tasks without additional training. Given a query image and example set of image-label pairs that define a new segmentation task, UniverSeg employs a new Cross-Block mechanism to produce accurate segmentation maps without the need for additional training. To achieve generalization to new tasks, we have gathered and standardized a collection of 53 open-access medical segmentation datasets with over 22,000 scans, which we refer to as MegaMedical. We used this collection to train UniverSeg on a diverse set of anatomies and imaging modalities. We demonstrate that UniverSeg substantially outperforms several related methods on unseen tasks, and thoroughly analyze and draw insights about important aspects of the proposed system. The UniverSeg source code and model weights are freely available at https://universeg.csail.mit.edu

Rajiv Movva, Divya Shanmugam, Kaihua Hou et al.

Healthcare data in the United States often records only a patient's coarse race group: for example, both Indian and Chinese patients are typically coded as "Asian." It is unknown, however, whether this coarse coding conceals meaningful disparities in the performance of clinical risk scores across granular race groups. Here we show that it does. Using data from 418K emergency department visits, we assess clinical risk score performance disparities across 26 granular groups for three outcomes, five risk scores, and four performance metrics. Across outcomes and metrics, we show that the risk scores exhibit significant granular performance disparities within coarse race groups. In fact, variation in performance within coarse groups often *exceeds* the variation between coarse groups. We explore why these disparities arise, finding that outcome rates, feature distributions, and the relationships between features and outcomes all vary significantly across granular groups. Our results suggest that healthcare providers, hospital systems, and machine learning researchers should strive to collect, release, and use granular race data in place of coarse race data, and that existing analyses may significantly underestimate racial disparities in performance.

Aniruddh Raghu, Divya Shanmugam, Eugene Pomerantsev et al.

Neural network models have demonstrated impressive performance in predicting pathologies and outcomes from the 12-lead electrocardiogram (ECG). However, these models often need to be trained with large, labelled datasets, which are not available for many predictive tasks of interest. In this work, we perform an empirical study examining whether training time data augmentation methods can be used to improve performance on such data-scarce ECG prediction problems. We investigate how data augmentation strategies impact model performance when detecting cardiac abnormalities from the ECG. Motivated by our finding that the effectiveness of existing augmentation strategies is highly task-dependent, we introduce a new method, TaskAug, which defines a flexible augmentation policy that is optimized on a per-task basis. We outline an efficient learning algorithm to do so that leverages recent work in nested optimization and implicit differentiation. In experiments, considering three datasets and eight predictive tasks, we find that TaskAug is competitive with or improves on prior work, and the learned policies shed light on what transformations are most effective for different tasks. We distill key insights from our experimental evaluation, generating a set of best practices for applying data augmentation to ECG prediction problems.

Helen Lu, Divya Shanmugam, Harini Suresh et al.

Test-time augmentation -- the aggregation of predictions across transformed examples of test inputs -- is an established technique to improve the performance of image classification models. Importantly, TTA can be used to improve model performance post-hoc, without additional training. Although test-time augmentation (TTA) can be applied to any data modality, it has seen limited adoption in NLP due in part to the difficulty of identifying label-preserving transformations. In this paper, we present augmentation policies that yield significant accuracy improvements with language models. A key finding is that augmentation policy design -- for instance, the number of samples generated from a single, non-deterministic augmentation -- has a considerable impact on the benefit of TTA. Experiments across a binary classification task and dataset show that test-time augmentation can deliver consistent improvements over current state-of-the-art approaches.

Jose Javier Gonzalez Ortiz, John Guttag, Adrian Dalca

Hypernetworks, neural networks that predict the parameters of another neural network, are powerful models that have been successfully used in diverse applications from image generation to multi-task learning. Unfortunately, existing hypernetworks are often challenging to train. Training typically converges far more slowly than for non-hypernetwork models, and the rate of convergence can be very sensitive to hyperparameter choices. In this work, we identify a fundamental and previously unidentified problem that contributes to the challenge of training hypernetworks: a magnitude proportionality between the inputs and outputs of the hypernetwork. We demonstrate both analytically and empirically that this can lead to unstable optimization, thereby slowing down convergence, and sometimes even preventing any learning. We present a simple solution to this problem using a revised hypernetwork formulation that we call Magnitude Invariant Parametrizations (MIP). We demonstrate the proposed solution on several hypernetwork tasks, where it consistently stabilizes training and achieves faster convergence. Furthermore, we perform a comprehensive ablation study including choices of activation function, normalization strategies, input dimensionality, and hypernetwork architecture; and find that MIP improves training in all scenarios. We provide easy-to-use code that can turn existing networks into MIP-based hypernetworks.

Kathleen M. Lewis, John Guttag

Online clothing catalogs lack diversity in body shape and garment size. Brands commonly display their garments on models of one or two sizes, rarely including plus-size models. To our knowledge, our paper presents the first method for generating images of garments and models in a new target size to tackle the size under-representation problem. Our primary technical contribution is a conditional generative adversarial network that learns deformation fields at multiple resolutions to realistically change the size of models and garments. Results from our two user studies show SizeGAN outperforms alternative methods along three dimensions -- realism, garment faithfulness, and size -- which are all important for real world use.

Aniruddh Raghu, Payal Chandak, Ridwan Alam et al.

Self-supervised learning (SSL) for clinical time series data has received significant attention in recent literature, since these data are highly rich and provide important information about a patient's physiological state. However, most existing SSL methods for clinical time series are limited in that they are designed for unimodal time series, such as a sequence of structured features (e.g., lab values and vitals signs) or an individual high-dimensional physiological signal (e.g., an electrocardiogram). These existing methods cannot be readily extended to model time series that exhibit multimodality, with structured features and high-dimensional data being recorded at each timestep in the sequence. In this work, we address this gap and propose a new SSL method -- Sequential Multi-Dimensional SSL -- where a SSL loss is applied both at the level of the entire sequence and at the level of the individual high-dimensional data points in the sequence in order to better capture information at both scales. Our strategy is agnostic to the specific form of loss function used at each level -- it can be contrastive, as in SimCLR, or non-contrastive, as in VICReg. We evaluate our method on two real-world clinical datasets, where the time series contains sequences of (1) high-frequency electrocardiograms and (2) structured data from lab values and vitals signs. Our experimental results indicate that pre-training with our method and then fine-tuning on downstream tasks improves performance over baselines on both datasets, and in several settings, can lead to improvements across different self-supervised loss functions.

Godfrey Tan, Massimiliano Poletto, John Guttag et al.

Role classification involves grouping hosts into related roles. It exposes the logical structure of a network, simplifies network management tasks such as policy checking and network segmentation, and can be used to improve the accuracy of network monitoring and analysis algorithms such as intrusion detection. This paper defines the role classification problem and introduces two practical algorithms that group hosts based on observed connection patterns while dealing with changes in these patterns over time. The algorithms have been implemented in a commercial network monitoring and analysis product for enterprise networks. Results from grouping two enterprise networks show that the number of groups identified by our algorithms can be two orders of magnitude smaller than the number of hosts and that the way our algorithms group hosts highly reflects the logical structure of the networks.

Godfrey Tan, John Guttag

Wireless local area networks (WLANs) based on the family of 802.11 technologies are becoming ubiquitous. These technologies support multiple data transmission rates. Transmitting at a lower data rate (by using a more resilient modulation scheme) increases the frame transmission time but reduces the bit error rate. In non-cooperative environments such as public hot-spots or WLANs operated by different enterprises that are physically close to each other, individual nodes attempt to maximize their achieved throughput by adjusting the data rate or frame size used, irrespective of the impact of this on overall system performance. In this paper, we show both analytically using a game theoretic model and through simulation that the existing 802.11 distributed MAC protocol, DCF (for Distributed Coordination Function), as well as its enhanced version, which is being standardized as part of 802.11e, can lead non-cooperative nodes to undesirable Nash equilibriums in which the wireless channel is inefficiently used. We show that by establishing independence between the allocation of the shared channel resource and the transmission strategies used by individual nodes, an ideal MAC protocol can lead rational nodes to arrive at equilibriums in which all competing nodes achieve higher throughputs than with DCF.

Jose Javier Gonzalez Ortiz, John Guttag, Adrian Dalca

Convolutional Neural Networks (CNNs) are the predominant model used for a variety of medical image analysis tasks. At inference time, these models are computationally intensive, especially with volumetric data. In principle, it is possible to trade accuracy for computational efficiency by manipulating the rescaling factor in the downsample and upsample layers of CNN architectures. However, properly exploring the accuracy-efficiency trade-off is prohibitively expensive with existing models. To address this, we introduce Scale-Space HyperNetworks (SSHN), a method that learns a spectrum of CNNs with varying internal rescaling factors. A single SSHN characterizes an entire Pareto accuracy-efficiency curve of models that match, and occasionally surpass, the outcomes of training many separate networks with fixed rescaling factors. We demonstrate the proposed approach in several medical image analysis applications, comparing SSHN against strategies with both fixed and dynamic rescaling factors. We find that SSHN consistently provides a better accuracy-efficiency trade-off at a fraction of the training cost. Trained SSHNs enable the user to quickly choose a rescaling factor that appropriately balances accuracy and computational efficiency for their particular needs at inference.

S. Mazdak Abulnaga, Andrew Hoopes, Malte Hoffmann et al.

Accurate registration of brain MRI scans is fundamental for cross-subject analysis in neuroscientific studies. This involves aligning both the cortical surface of the brain and the interior volume. Traditional methods treat volumetric and surface-based registration separately, which often leads to inconsistencies that limit downstream analyses. We propose a deep learning framework, NeurAlign, that registers $3$D brain MRI images by jointly aligning both cortical and subcortical regions through a unified volume-and-surface-based representation. Our approach leverages an intermediate spherical coordinate space to bridge anatomical surface topology with volumetric anatomy, enabling consistent and anatomically accurate alignment. By integrating spherical registration into the learning, our method ensures geometric coherence between volume and surface domains. In a series of experiments on both in-domain and out-of-domain datasets, our method consistently outperforms both classical and machine learning-based registration methods -- improving the Dice score by up to 7 points while maintaining regular deformation fields. Additionally, it is orders of magnitude faster than the standard method for this task, and is simpler to use because it requires no additional inputs beyond an MRI scan. With its superior accuracy, fast inference, and ease of use, NeurAlign sets a new standard for joint cortical and subcortical registration.

Katie Matton, Purvaja Balaji, Hamzeh Ghasemzadeh et al.

Phonotrauma refers to vocal fold tissue damage resulting from exposure to forces during voicing. It occurs on a continuum from mild to severe, and treatment options can vary based on severity. Assessment of severity involves a clinician's expert judgment, which is costly and can vary widely in reliability. In this work, we present the first method for automatically classifying phonotrauma severity from vocal fold images. To account for the ordinal nature of the labels, we adopt a widely used ordinal regression framework. To account for label uncertainty, we propose a novel modification to ordinal regression loss functions that enables them to operate on soft labels reflecting annotator rating distributions. Our proposed soft ordinal regression method achieves predictive performance approaching that of clinical experts, while producing well-calibrated uncertainty estimates. By providing an automated tool for phonotrauma severity assessment, our work can enable large-scale studies of phonotrauma, ultimately leading to improved clinical understanding and patient care.

Emily Mu, John Guttag, Maggie Makar

Given a similarity metric, contrastive methods learn a representation in which examples that are similar are pushed together and examples that are dissimilar are pulled apart. Contrastive learning techniques have been utilized extensively to learn representations for tasks ranging from image classification to caption generation. However, existing contrastive learning approaches can fail to generalize because they do not take into account the possibility of different similarity relations. In this paper, we propose a novel multi-similarity contrastive loss (MSCon), that learns generalizable embeddings by jointly utilizing supervision from multiple metrics of similarity. Our method automatically learns contrastive similarity weightings based on the uncertainty in the corresponding similarity, down-weighting uncertain tasks and leading to better out-of-domain generalization to new tasks. We show empirically that networks trained with MSCon outperform state-of-the-art baselines on in-domain and out-of-domain settings.

Robert Lewis, Katie Matton, Rosalind W. Picard et al.

Self-supervised pre-training with contrastive learning is a powerful method for learning from sparsely labeled data. However, performance can drop considerably when there is a shift in the distribution of data from training to test time. We study this phenomenon in a setting in which the training data come from multiple domains, and the test data come from a domain not seen at training that is subject to significant covariate shift. We present a new method for contrastive learning that incorporates domain labels to increase the domain invariance of learned representations, leading to improved out-of-distribution generalization. Our method adjusts the temperature parameter in the InfoNCE loss -- which controls the relative weighting of negative pairs -- using the probability that a negative sample comes from the same domain as the anchor. This upweights pairs from more similar domains, encouraging the model to discriminate samples based on domain-invariant attributes. Through experiments on a variant of the MNIST dataset, we demonstrate that our method yields better out-of-distribution performance than domain generalization baselines. Furthermore, our method maintains strong in-distribution task performance, substantially outperforming baselines on this measure.

Davis Blalock, Jose Javier Gonzalez Ortiz, Jonathan Frankle et al.

Neural network pruning---the task of reducing the size of a network by removing parameters---has been the subject of a great deal of work in recent years. We provide a meta-analysis of the literature, including an overview of approaches to pruning and consistent findings in the literature. After aggregating results across 81 papers and pruning hundreds of models in controlled conditions, our clearest finding is that the community suffers from a lack of standardized benchmarks and metrics. This deficiency is substantial enough that it is hard to compare pruning techniques to one another or determine how much progress the field has made over the past three decades. To address this situation, we identify issues with current practices, suggest concrete remedies, and introduce ShrinkBench, an open-source framework to facilitate standardized evaluations of pruning methods. We use ShrinkBench to compare various pruning techniques and show that its comprehensive evaluation can prevent common pitfalls when comparing pruning methods.

Adrian V. Dalca, Guha Balakrishnan, John Guttag et al.

Classical deformable registration techniques achieve impressive results and offer a rigorous theoretical treatment, but are computationally intensive since they solve an optimization problem for each image pair. Recently, learning-based methods have facilitated fast registration by learning spatial deformation functions. However, these approaches use restricted deformation models, require supervised labels, or do not guarantee a diffeomorphic (topology-preserving) registration. Furthermore, learning-based registration tools have not been derived from a probabilistic framework that can offer uncertainty estimates. In this paper, we build a connection between classical and learning-based methods. We present a probabilistic generative model and derive an unsupervised learning-based inference algorithm that uses insights from classical registration methods and makes use of recent developments in convolutional neural networks (CNNs). We demonstrate our method on a 3D brain registration task for both images and anatomical surfaces, and provide extensive empirical analyses. Our principled approach results in state of the art accuracy and very fast runtimes, while providing diffeomorphic guarantees. Our implementation is available at http://voxelmorph.csail.mit.edu.

Adrian V. Dalca, John Guttag, Mert R. Sabuncu

A wide range of systems exhibit high dimensional incomplete data. Accurate estimation of the missing data is often desired, and is crucial for many downstream analyses. Many state-of-the-art recovery methods involve supervised learning using datasets containing full observations. In contrast, we focus on unsupervised estimation of missing image data, where no full observations are available - a common situation in practice. Unsupervised imputation methods for images often employ a simple linear subspace to capture correlations between data dimensions, omitting more complex relationships. In this work, we introduce a general probabilistic model that describes sparse high dimensional imaging data as being generated by a deep non-linear embedding. We derive a learning algorithm using a variational approximation based on convolutional neural networks and discuss its relationship to linear imputation models, the variational auto encoder, and deep image priors. We introduce sparsity-aware network building blocks that explicitly model observed and missing data. We analyze proposed sparsity-aware network building blocks, evaluate our method on public domain imaging datasets, and conclude by showing that our method enables imputation in an important real-world problem involving medical images. The code is freely available as part of the \verb|neuron| library at http://github.com/adalca/neuron.

Adrian V. Dalca, John Guttag, Mert R. Sabuncu

We consider the problem of segmenting a biomedical image into anatomical regions of interest. We specifically address the frequent scenario where we have no paired training data that contains images and their manual segmentations. Instead, we employ unpaired segmentation images to build an anatomical prior. Critically these segmentations can be derived from imaging data from a different dataset and imaging modality than the current task. We introduce a generative probabilistic model that employs the learned prior through a convolutional neural network to compute segmentations in an unsupervised setting. We conducted an empirical analysis of the proposed approach in the context of structural brain MRI segmentation, using a multi-study dataset of more than 14,000 scans. Our results show that an anatomical prior can enable fast unsupervised segmentation which is typically not possible using standard convolutional networks. The integration of anatomical priors can facilitate CNN-based anatomical segmentation in a range of novel clinical problems, where few or no annotations are available and thus standard networks are not trainable. The code is freely available at http://github.com/adalca/neuron.

Kathleen M. Lewis, Natalia S. Rost, John Guttag et al.

We introduce SparseVM, a method that registers clinical-quality 3D MR scans both faster and more accurately than previously possible. Deformable alignment, or registration, of clinical scans is a fundamental task for many clinical neuroscience studies. However, most registration algorithms are designed for high-resolution research-quality scans. In contrast to research-quality scans, clinical scans are often sparse, missing up to 86% of the slices available in research-quality scans. Existing methods for registering these sparse images are either inaccurate or extremely slow. We present a learning-based registration method, SparseVM, that is more accurate and orders of magnitude faster than the most accurate clinical registration methods. To our knowledge, it is the first method to use deep learning specifically tailored to registering clinical images. We demonstrate our method on a clinically-acquired MRI dataset of stroke patients and on a simulated sparse MRI dataset. Our code is available as part of the VoxelMorph package at http://voxelmorph.mit.edu/.

Adrian V. Dalca, Guha Balakrishnan, John Guttag et al.

Traditional deformable registration techniques achieve impressive results and offer a rigorous theoretical treatment, but are computationally intensive since they solve an optimization problem for each image pair. Recently, learning-based methods have facilitated fast registration by learning spatial deformation functions. However, these approaches use restricted deformation models, require supervised labels, or do not guarantee a diffeomorphic (topology-preserving) registration. Furthermore, learning-based registration tools have not been derived from a probabilistic framework that can offer uncertainty estimates. In this paper, we present a probabilistic generative model and derive an unsupervised learning-based inference algorithm that makes use of recent developments in convolutional neural networks (CNNs). We demonstrate our method on a 3D brain registration task, and provide an empirical analysis of the algorithm. Our approach results in state of the art accuracy and very fast runtimes, while providing diffeomorphic guarantees and uncertainty estimates. Our implementation is available online at http://voxelmorph.csail.mit.edu .

Guha Balakrishnan, Amy Zhao, Mert R. Sabuncu et al.

We present a fast learning-based algorithm for deformable, pairwise 3D medical image registration. Current registration methods optimize an objective function independently for each pair of images, which can be time-consuming for large data. We define registration as a parametric function, and optimize its parameters given a set of images from a collection of interest. Given a new pair of scans, we can quickly compute a registration field by directly evaluating the function using the learned parameters. We model this function using a convolutional neural network (CNN), and use a spatial transform layer to reconstruct one image from another while imposing smoothness constraints on the registration field. The proposed method does not require supervised information such as ground truth registration fields or anatomical landmarks. We demonstrate registration accuracy comparable to state-of-the-art 3D image registration, while operating orders of magnitude faster in practice. Our method promises to significantly speed up medical image analysis and processing pipelines, while facilitating novel directions in learning-based registration and its applications. Our code is available at https://github.com/balakg/voxelmorph .

Hallee E. Wong, Marianne Rakic, John Guttag et al.

Biomedical image segmentation is a crucial part of both scientific research and clinical care. With enough labelled data, deep learning models can be trained to accurately automate specific biomedical image segmentation tasks. However, manually segmenting images to create training data is highly labor intensive and requires domain expertise. We present \emph{ScribblePrompt}, a flexible neural network based interactive segmentation tool for biomedical imaging that enables human annotators to segment previously unseen structures using scribbles, clicks, and bounding boxes. Through rigorous quantitative experiments, we demonstrate that given comparable amounts of interaction, ScribblePrompt produces more accurate segmentations than previous methods on datasets unseen during training. In a user study with domain experts, ScribblePrompt reduced annotation time by 28% while improving Dice by 15% compared to the next best method. ScribblePrompt's success rests on a set of careful design decisions. These include a training strategy that incorporates both a highly diverse set of images and tasks, novel algorithms for simulated user interactions and labels, and a network that enables fast inference. We showcase ScribblePrompt in an interactive demo, provide code, and release a dataset of scribble annotations at https://scribbleprompt.csail.mit.edu

Katie Matton, Robert Osazuwa Ness, John Guttag et al.

Large language models (LLMs) are capable of generating plausible explanations of how they arrived at an answer to a question. However, these explanations can misrepresent the model's "reasoning" process, i.e., they can be unfaithful. This, in turn, can lead to over-trust and misuse. We introduce a new approach for measuring the faithfulness of LLM explanations. First, we provide a rigorous definition of faithfulness. Since LLM explanations mimic human explanations, they often reference high-level concepts in the input question that purportedly influenced the model. We define faithfulness in terms of the difference between the set of concepts that LLM explanations imply are influential and the set that truly are. Second, we present a novel method for estimating faithfulness that is based on: (1) using an auxiliary LLM to modify the values of concepts within model inputs to create realistic counterfactuals, and (2) using a Bayesian hierarchical model to quantify the causal effects of concepts at both the example- and dataset-level. Our experiments show that our method can be used to quantify and discover interpretable patterns of unfaithfulness. On a social bias task, we uncover cases where LLM explanations hide the influence of social bias. On a medical question answering task, we uncover cases where LLM explanations provide misleading claims about which pieces of evidence influenced the model's decisions.

S. Mazdak Abulnaga, Andrew Hoopes, Neel Dey et al. · mit

We present MultiMorph, a fast and efficient method for constructing anatomical atlases on the fly. Atlases capture the canonical structure of a collection of images and are essential for quantifying anatomical variability across populations. However, current atlas construction methods often require days to weeks of computation, thereby discouraging rapid experimentation. As a result, many scientific studies rely on suboptimal, precomputed atlases from mismatched populations, negatively impacting downstream analyses. MultiMorph addresses these challenges with a feedforward model that rapidly produces high-quality, population-specific atlases in a single forward pass for any 3D brain dataset, without any fine-tuning or optimization. MultiMorph is based on a linear group-interaction layer that aggregates and shares features within the group of input images. Further, by leveraging auxiliary synthetic data, MultiMorph generalizes to new imaging modalities and population groups at test-time. Experimentally, MultiMorph outperforms state-of-the-art optimization-based and learning-based atlas construction methods in both small and large population settings, with a 100-fold reduction in time. This makes MultiMorph an accessible framework for biomedical researchers without machine learning expertise, enabling rapid, high-quality atlas generation for diverse studies.

Divya Shanmugam, Shuvom Sadhuka, Manish Raghavan et al.

It remains difficult to evaluate machine learning classifiers in the absence of a large, labeled dataset. While labeled data can be prohibitively expensive or impossible to obtain, unlabeled data is plentiful. Here, we introduce Semi-Supervised Model Evaluation (SSME), a method that uses both labeled and unlabeled data to evaluate machine learning classifiers. SSME is the first evaluation method to take advantage of the fact that: (i) there are frequently multiple classifiers for the same task, (ii) continuous classifier scores are often available for all classes, and (iii) unlabeled data is often far more plentiful than labeled data. The key idea is to use a semi-supervised mixture model to estimate the joint distribution of ground truth labels and classifier predictions. We can then use this model to estimate any metric that is a function of classifier scores and ground truth labels (e.g., accuracy or expected calibration error). We present experiments in four domains where obtaining large labeled datasets is often impractical: (1) healthcare, (2) content moderation, (3) molecular property prediction, and (4) image annotation. Our results demonstrate that SSME estimates performance more accurately than do competing methods, reducing error by 5.1x relative to using labeled data alone and 2.4x relative to the next best competing method. SSME also improves accuracy when evaluating performance across subsets of the test distribution (e.g., specific demographic subgroups) and when evaluating the performance of language models.

Divya Shanmugam, Helen Lu, Swami Sankaranarayanan et al.

A conformal classifier produces a set of predicted classes and provides a probabilistic guarantee that the set includes the true class. Unfortunately, it is often the case that conformal classifiers produce uninformatively large sets. In this work, we show that test-time augmentation (TTA)--a technique that introduces inductive biases during inference--reduces the size of the sets produced by conformal classifiers. Our approach is flexible, computationally efficient, and effective. It can be combined with any conformal score, requires no model retraining, and reduces prediction set sizes by 10%-14% on average. We conduct an evaluation of the approach spanning three datasets, three models, two established conformal scoring methods, different guarantee strengths, and several distribution shifts to show when and why test-time augmentation is a useful addition to the conformal pipeline.

Hallee E. Wong, Jose Javier Gonzalez Ortiz, John Guttag et al.

Medical researchers and clinicians often need to perform novel segmentation tasks on a set of related images. Existing methods for segmenting a new dataset are either interactive, requiring substantial human effort for each image, or require an existing set of previously labeled images. We introduce a system, MultiverSeg, that enables practitioners to rapidly segment an entire new dataset without requiring access to any existing labeled data from that task or domain. Along with the image to segment, the model takes user interactions such as clicks, bounding boxes or scribbles as input, and predicts a segmentation. As the user segments more images, those images and segmentations become additional inputs to the model, providing context. As the context set of labeled images grows, the number of interactions required to segment each new image decreases. We demonstrate that MultiverSeg enables users to interactively segment new datasets efficiently, by amortizing the number of interactions per image to achieve an accurate segmentation. Compared to using a state-of-the-art interactive segmentation method, MultiverSeg reduced the total number of clicks by 36% and scribble steps by 25% to achieve 90% Dice on sets of images from unseen tasks. We release code and model weights at https://multiverseg.csail.mit.edu

Zachary Berger, Daniel Prakah-Asante, John Guttag et al.

This position paper argues that current benchmarking practice in 12-lead ECG representation learning must be fixed to ensure progress is reliable and aligned with clinically meaningful objectives. The field has largely converged on three public multi-label benchmarks (PTB-XL, CPSC2018, CSN) dominated by arrhythmia and waveform-morphology labels, even though the ECG is known to encode substantially broader clinical information. We argue that downstream evaluation should expand to include an assessment of structural heart disease and patient-level forecasting, in addition to other evolving ECG-related endpoints, as relevant clinical targets. Next, we outline evaluation best practices for multi-label, imbalanced settings, and show that when they are applied, the literature's current conclusion about which representations perform best is altered. Furthermore, we demonstrate the surprising result that a randomly initialized encoder with linear evaluation matches state-of-the-art pre-training on many tasks. This motivates the use of a random encoder as a reasonable baseline model. We substantiate our observations with an empirical evaluation of three representative ECG pre-training approaches across six evaluation settings: the three standard benchmarks, a structural disease dataset, hemodynamic inference, and patient forecasting.

Marianne Rakic, Hallee E. Wong, Jose Javier Gonzalez Ortiz et al.

Existing learning-based solutions to medical image segmentation have two important shortcomings. First, for most new segmentation task, a new model has to be trained or fine-tuned. This requires extensive resources and machine learning expertise, and is therefore often infeasible for medical researchers and clinicians. Second, most existing segmentation methods produce a single deterministic segmentation mask for a given image. In practice however, there is often considerable uncertainty about what constitutes the correct segmentation, and different expert annotators will often segment the same image differently. We tackle both of these problems with Tyche, a model that uses a context set to generate stochastic predictions for previously unseen tasks without the need to retrain. Tyche differs from other in-context segmentation methods in two important ways. (1) We introduce a novel convolution block architecture that enables interactions among predictions. (2) We introduce in-context test-time augmentation, a new mechanism to provide prediction stochasticity. When combined with appropriate model design and loss functions, Tyche can predict a set of plausible diverse segmentation candidates for new or unseen medical images and segmentation tasks without the need to retrain.

Andrew Hoopes, Malte Hoffmann, Douglas N. Greve et al.

We introduce HyperMorph, a framework that facilitates efficient hyperparameter tuning in learning-based deformable image registration. Classical registration algorithms perform an iterative pair-wise optimization to compute a deformation field that aligns two images. Recent learning-based approaches leverage large image datasets to learn a function that rapidly estimates a deformation for a given image pair. In both strategies, the accuracy of the resulting spatial correspondences is strongly influenced by the choice of certain hyperparameter values. However, an effective hyperparameter search consumes substantial time and human effort as it often involves training multiple models for different fixed hyperparameter values and may lead to suboptimal registration. We propose an amortized hyperparameter learning strategy to alleviate this burden by learning the impact of hyperparameters on deformation fields. We design a meta network, or hypernetwork, that predicts the parameters of a registration network for input hyperparameters, thereby comprising a single model that generates the optimal deformation field corresponding to given hyperparameter values. This strategy enables fast, high-resolution hyperparameter search at test-time, reducing the inefficiency of traditional approaches while increasing flexibility. We also demonstrate additional benefits of HyperMorph, including enhanced robustness to model initialization and the ability to rapidly identify optimal hyperparameter values specific to a dataset, image contrast, task, or even anatomical region, all without the need to retrain models. We make our code publicly available at http://hypermorph.voxelmorph.net.

Davis Blalock, John Guttag

Multiplying matrices is among the most fundamental and compute-intensive operations in machine learning. Consequently, there has been significant work on efficiently approximating matrix multiplies. We introduce a learning-based algorithm for this task that greatly outperforms existing methods. Experiments using hundreds of matrices from diverse domains show that it often runs $100\times$ faster than exact matrix products and $10\times$ faster than current approximate methods. In the common case that one matrix is known ahead of time, our method also has the interesting property that it requires zero multiply-adds. These results suggest that a mixture of hashing, averaging, and byte shuffling$-$the core operations of our method$-$could be a more promising building block for machine learning than the sparsified, factorized, and/or scalar quantized matrix products that have recently been the focus of substantial research and hardware investment.

Aniruddh Raghu, John Guttag, Katherine Young et al.

The impact of machine learning models on healthcare will depend on the degree of trust that healthcare professionals place in the predictions made by these models. In this paper, we present a method to provide people with clinical expertise with domain-relevant evidence about why a prediction should be trusted. We first design a probabilistic model that relates meaningful latent concepts to prediction targets and observed data. Inference of latent variables in this model corresponds to both making a prediction and providing supporting evidence for that prediction. We present a two-step process to efficiently approximate inference: (i) estimating model parameters using variational learning, and (ii) approximating maximum a posteriori estimation of latent variables in the model using a neural network, trained with an objective derived from the probabilistic model. We demonstrate the method on the task of predicting mortality risk for patients with cardiovascular disease. Specifically, using electrocardiogram and tabular data as input, we show that our approach provides appropriate domain-relevant supporting evidence for accurate predictions.

Andrew Hoopes, Malte Hoffmann, Bruce Fischl et al.

We present HyperMorph, a learning-based strategy for deformable image registration that removes the need to tune important registration hyperparameters during training. Classical registration methods solve an optimization problem to find a set of spatial correspondences between two images, while learning-based methods leverage a training dataset to learn a function that generates these correspondences. The quality of the results for both types of techniques depends greatly on the choice of hyperparameters. Unfortunately, hyperparameter tuning is time-consuming and typically involves training many separate models with various hyperparameter values, potentially leading to suboptimal results. To address this inefficiency, we introduce amortized hyperparameter learning for image registration, a novel strategy to learn the effects of hyperparameters on deformation fields. The proposed framework learns a hypernetwork that takes in an input hyperparameter and modulates a registration network to produce the optimal deformation field for that hyperparameter value. In effect, this strategy trains a single, rich model that enables rapid, fine-grained discovery of hyperparameter values from a continuous interval at test-time. We demonstrate that this approach can be used to optimize multiple hyperparameters considerably faster than existing search strategies, leading to a reduced computational and human burden as well as increased flexibility. We also show several important benefits, including increased robustness to initialization and the ability to rapidly identify optimal hyperparameter values specific to a registration task, dataset, or even a single anatomical region, all without retraining the HyperMorph model. Our code is publicly available at http://voxelmorph.mit.edu.

Divya Shanmugam, Davis Blalock, Guha Balakrishnan et al.

Test-time augmentation -- the aggregation of predictions across transformed versions of a test input -- is a common practice in image classification. Traditionally, predictions are combined using a simple average. In this paper, we present 1) experimental analyses that shed light on cases in which the simple average is suboptimal and 2) a method to address these shortcomings. A key finding is that even when test-time augmentation produces a net improvement in accuracy, it can change many correct predictions into incorrect predictions. We delve into when and why test-time augmentation changes a prediction from being correct to incorrect and vice versa. Building on these insights, we present a learning-based method for aggregating test-time augmentations. Experiments across a diverse set of models, datasets, and augmentations show that our method delivers consistent improvements over existing approaches.

Roshni Sahoo, Divya Shanmugam, John Guttag

Current unsupervised domain adaptation methods can address many types of distribution shift, but they assume data from the source domain is freely available. As the use of pre-trained models becomes more prevalent, it is reasonable to assume that source data is unavailable. We propose an unsupervised method for adapting a source classifier to a target domain that varies from the source domain along natural axes, such as brightness and contrast. Our method only requires access to unlabeled target instances and the source classifier. We validate our method in scenarios where the distribution shift involves brightness, contrast, and rotation and show that it outperforms fine-tuning baselines in scenarios with limited labeled data.

Marianne Rakic, John Guttag, Adrian V. Dalca

Changes over time in brain anatomy can provide important insight for treatment design or scientific analyses. We present a method that predicts how a brain MRI for an individual will change over time. We model changes using a diffeomorphic deformation field that we predict using function using convolutional neural networks. Given a predicted deformation field, a baseline scan can be warped to give a prediction of the brain scan at a future time. We demonstrate the method using the ADNI cohort, and analyze how performance is affected by model variants and the subject-specific information provided. We show that the model provides good predictions and that external clinical data can improve predictions.

Ava P. Soleimany, Harini Suresh, Jose Javier Gonzalez Ortiz et al.

Global eradication of malaria depends on the development of drugs effective against the silent, yet obligate liver stage of the disease. The gold standard in drug development remains microscopic imaging of liver stage parasites in in vitro cell culture models. Image analysis presents a major bottleneck in this pipeline since the parasite has significant variability in size, shape, and density in these models. As with other highly variable datasets, traditional segmentation models have poor generalizability as they rely on hand-crafted features; thus, manual annotation of liver stage malaria images remains standard. To address this need, we develop a convolutional neural network architecture that utilizes spatial dropout sampling for parasite segmentation and epistemic uncertainty estimation in images of liver stage malaria. Our pipeline produces high-precision segmentations nearly identical to expert annotations, generalizes well on a diverse dataset of liver stage malaria parasites, and promotes independence between learned feature maps to model the uncertainty of generated predictions.

Maggie Makar, Fredrik D. Johansson, John Guttag et al.

Estimation of individual treatment effects is commonly used as the basis for contextual decision making in fields such as healthcare, education, and economics. However, it is often sufficient for the decision maker to have estimates of upper and lower bounds on the potential outcomes of decision alternatives to assess risks and benefits. We show that, in such cases, we can improve sample efficiency by estimating simple functions that bound these outcomes instead of estimating their conditional expectations, which may be complex and hard to estimate. Our analysis highlights a trade-off between the complexity of the learning task and the confidence with which the learned bounds hold. Guided by these findings, we develop an algorithm for learning upper and lower bounds on potential outcomes which optimize an objective function defined by the decision maker, subject to the probability that bounds are violated being small. Using a clinical dataset and a well-known causality benchmark, we demonstrate that our algorithm outperforms baselines, providing tighter, more reliable bounds.

Adrian V. Dalca, Marianne Rakic, John Guttag et al.

We develop a learning framework for building deformable templates, which play a fundamental role in many image analysis and computational anatomy tasks. Conventional methods for template creation and image alignment to the template have undergone decades of rich technical development. In these frameworks, templates are constructed using an iterative process of template estimation and alignment, which is often computationally very expensive. Due in part to this shortcoming, most methods compute a single template for the entire population of images, or a few templates for specific sub-groups of the data. In this work, we present a probabilistic model and efficient learning strategy that yields either universal or conditional templates, jointly with a neural network that provides efficient alignment of the images to these templates. We demonstrate the usefulness of this method on a variety of domains, with a special focus on neuroimaging. This is particularly useful for clinical applications where a pre-existing template does not exist, or creating a new one with traditional methods can be prohibitively expensive. Our code and atlases are available online as part of the VoxelMorph library at http://voxelmorph.csail.mit.edu.

Divya Shanmugam, Davis Blalock, John Guttag

Patients who suffer an acute coronary syndrome are at elevated risk for adverse cardiovascular events such as myocardial infarction and cardiovascular death. Accurate assessment of this risk is crucial to their course of care. We focus on estimating a patient's risk of cardiovascular death after an acute coronary syndrome based on a patient's raw electrocardiogram (ECG) signal. Learning from this signal is challenging for two reasons: 1) positive examples signifying a downstream cardiovascular event are scarce, causing drastic class imbalance, and 2) each patient's ECG signal consists of thousands of heartbeats, accompanied by a single label for the downstream outcome. Machine learning has been previously applied to this task, but most approaches rely on hand-crafted features and domain knowledge. We propose a method that learns a representation from the raw ECG signal by using a multiple instance learning framework. We present a learned risk score for cardiovascular death that outperforms existing risk metrics in predicting cardiovascular death within 30, 60, 90, and 365 days on a dataset of 5000 patients.

Guha Balakrishnan, Amy Zhao, Mert R. Sabuncu et al.

We present VoxelMorph, a fast learning-based framework for deformable, pairwise medical image registration. Traditional registration methods optimize an objective function for each pair of images, which can be time-consuming for large datasets or rich deformation models. In contrast to this approach, and building on recent learning-based methods, we formulate registration as a function that maps an input image pair to a deformation field that aligns these images. We parameterize the function via a convolutional neural network (CNN), and optimize the parameters of the neural network on a set of images. Given a new pair of scans, VoxelMorph rapidly computes a deformation field by directly evaluating the function. In this work, we explore two different training strategies. In the first (unsupervised) setting, we train the model to maximize standard image matching objective functions that are based on the image intensities. In the second setting, we leverage auxiliary segmentations available in the training data. We demonstrate that the unsupervised model's accuracy is comparable to state-of-the-art methods, while operating orders of magnitude faster. We also show that VoxelMorph trained with auxiliary data improves registration accuracy at test time, and evaluate the effect of training set size on registration. Our method promises to speed up medical image analysis and processing pipelines, while facilitating novel directions in learning-based registration and its applications. Our code is freely available at voxelmorph.csail.mit.edu.

Harini Suresh, Jen J. Gong, John Guttag

Machine learning approaches have been effective in predicting adverse outcomes in different clinical settings. These models are often developed and evaluated on datasets with heterogeneous patient populations. However, good predictive performance on the aggregate population does not imply good performance for specific groups. In this work, we present a two-step framework to 1) learn relevant patient subgroups, and 2) predict an outcome for separate patient populations in a multi-task framework, where each population is a separate task. We demonstrate how to discover relevant groups in an unsupervised way with a sequence-to-sequence autoencoder. We show that using these groups in a multi-task framework leads to better predictive performance of in-hospital mortality both across groups and overall. We also highlight the need for more granular evaluation of performance when dealing with heterogeneous populations.

Guha Balakrishnan, Amy Zhao, Adrian V. Dalca et al.

We address the computational problem of novel human pose synthesis. Given an image of a person and a desired pose, we produce a depiction of that person in that pose, retaining the appearance of both the person and background. We present a modular generative neural network that synthesizes unseen poses using training pairs of images and poses taken from human action videos. Our network separates a scene into different body part and background layers, moves body parts to new locations and refines their appearances, and composites the new foreground with a hole-filled background. These subtasks, implemented with separate modules, are trained jointly using only a single target image as a supervised label. We use an adversarial discriminator to force our network to synthesize realistic details conditioned on pose. We demonstrate image synthesis results on three action classes: golf, yoga/workouts and tennis, and show that our method produces accurate results within action classes as well as across action classes. Given a sequence of desired poses, we also produce coherent videos of actions.

Ronnachai Jaroensri, Amy Zhao, Guha Balakrishnan et al.

For many movement disorders, such as Parkinson's disease and ataxia, disease progression is visually assessed by a clinician using a numerical disease rating scale. These tests are subjective, time-consuming, and must be administered by a professional. This can be problematic where specialists are not available, or when a patient is not consistently evaluated by the same clinician. We present an automated method for quantifying the severity of motion impairment in patients with ataxia, using only video recordings. We consider videos of the finger-to-nose test, a common movement task used as part of the assessment of ataxia progression during the course of routine clinical checkups. Our method uses neural network-based pose estimation and optical flow techniques to track the motion of the patient's hand in a video recording. We extract features that describe qualities of the motion such as speed and variation in performance. Using labels provided by an expert clinician, we train a supervised learning model that predicts severity according to the Brief Ataxia Rating Scale (BARS). The performance of our system is comparable to that of a group of ataxia specialists in terms of mean error and correlation, and our system's predictions were consistently within the range of inter-rater variability. This work demonstrates the feasibility of using computer vision and machine learning to produce consistent and clinically useful measures of motor impairment.