Medical image harmonization aims to reduce the differences in appearance caused by scanner hardware variations to allow for consistent and reliable comparisons across devices. Harmonization based on paired images from different devices has limited applicability in real-world clinical settings. On the other hand, unpaired harmonization typically does not guarantee anatomy consistency, which is problematic because anatomical information preservation is paramount. The Schrödinger bridge framework has achieved state-of-the-art style transfer performance with natural images by matching distributions of unpaired images, but this approach can also introduce anatomy changes when applied to medical images. We show that such changes occur because the Schrödinger bridge uses the square of the Euclidean distance between images as the transport cost in an entropy-regularized optimal transport problem. Such a transport cost is not appropriate for measuring anatomical distances, as medical images with the same anatomy need not have a small Euclidean distance between them. In this paper, we propose a latent metric Schrödinger bridge (LMSB) framework to improve the anatomical consistency for the harmonization of medical images. We develop an invertible network that maps medical images into a latent Euclidean metric space where the distances among images with the same anatomy are minimized using the pullback latent metric. Within this latent space, we train a Schrödinger bridge to match distributions. We show that the proposed LMSB is superior to the direct application of a Schrödinger bridge to harmonize optical coherence tomography (OCT) images.

Linear mixed-effect models provide a natural baseline for estimating disease progression using longitudinal data. They provide interpretable models at the cost of modeling assumptions on the progression profiles and their variability across subjects. A significant improvement is to embed the data in a Riemannian manifold and learn patient-specific trajectories distributed around a central geodesic. A few interpretable parameters characterize subject trajectories at the cost of a prior choice of the metric, which determines the shape of the trajectories. We extend this approach by learning the metric from the data allowing more flexibility while keeping the interpretability. Specifically, we learn the metric as the push-forward of the Euclidean metric by a diffeomorphism. This diffeomorphism is estimated iteratively as the composition of radial basis functions belonging to a reproducible kernel Hilbert space. The metric update allows us to improve the forecasting of imaging and clinical biomarkers in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort. Our results compare favorably to the 56 methods benchmarked in the TADPOLE challenge.

We devise an algorithm using a Bayesian optimization framework in conjunction with contextual visual data for the efficient localization of objects in still images. Recent research has demonstrated substantial progress in object localization and related tasks for computer vision. However, many current state-of-the-art object localization procedures still suffer from inaccuracy and inefficiency, in addition to failing to provide a principled and interpretable system amenable to high-level vision tasks. We address these issues with the current research. Our method encompasses an active search procedure that uses contextual data to generate initial bounding-box proposals for a target object. We train a convolutional neural network to approximate an offset distance from the target object. Next, we use a Gaussian Process to model this offset response signal over the search space of the target. We then employ a Bayesian active search for accurate localization of the target. In experiments, we compare our approach to a state-of-the-art bounding-box regression method for a challenging pedestrian localization task. Our method exhibits a substantial improvement over this baseline regression method.

We consider the problem of group testing with sum observations and noiseless answers, in which we aim to locate multiple objects by querying the number of objects in each of a sequence of chosen sets. We study a probabilistic setting with entropy loss, in which we assume a joint Bayesian prior density on the locations of the objects and seek to choose the sets queried to minimize the expected entropy of the Bayesian posterior distribution after a fixed number of questions. We present a new non-adaptive policy, called the dyadic policy, show that it is optimal among non-adaptive policies, and is within a factor of two of optimal among adaptive policies. This policy is quick to compute, its nonadaptive nature makes it easy to parallelize, and our bounds show that it performs well even when compared with adaptive policies. We also study an adaptive greedy policy, which maximizes the one-step expected reduction in entropy, and show that it performs at least as well as the dyadic policy, offering greater query efficiency but reduced parallelism. Numerical experiments demonstrate that both procedures outperform a divide-and-conquer benchmark policy from the literature, called sequential bifurcation, and show how these procedures may be applied in a stylized computer vision problem.

We consider the problem of quickly localizing multiple targets by asking questions of the form “How many targets are within this set" while obtaining noisy answers. This setting is a generalization to multiple targets of the game of 20 questions in which only a single target is queried. We assume that the targets are points on the real line, or in a two dimensional plane for the experiments, drawn independently from a known distribution. We evaluate the performance of a policy using the expected entropy of the posterior distribution after a fixed number of questions with noisy answers. We derive a lower bound for the value of this problem and study a specific policy, named the dyadic policy. We show that this policy achieves a value which is no more than twice this lower bound when answers are noise-free, and show a more general constant factor approximation guarantee for the noisy setting. We present an empirical evaluation of this policy on simulated data for the problem of detecting multiple instances of the same object in an image. Finally, we present experiments on localizing multiple faces simultaneously on real images.

The problem of target localization with noise is addressed. The target is a sample from a continuous random variable with known distribution and the goal is to locate it with minimum mean squared error distortion. The localization scheme or policy proceeds by queries, or questions, weather or not the target belongs to some subset as it is addressed in the 20-question framework. These subsets are not constrained to be intervals and the answers to the queries are noisy. While this situation is well studied for adaptive querying, this paper is focused on the non adaptive querying policies based on dyadic questions. The asymptotic minimum achievable distortion under such policies is derived. Furthermore, a policy named the Aurelian1is exhibited which achieves asymptotically this distortion.

This paper considers the task of finding a target location by making a limited number of sequential observations. Each observation results from evaluating an imperfect classifier of a chosen cost and accuracy on an interval of chosen length and position. Within a Bayesian framework, we study the problem of minimizing an objective that combines the entropy of the posterior distribution with the cost of the questions asked. In this problem, we show that the one-step lookahead policy is Bayes-optimal for any arbitrary time horizon. Moreover, this one-step lookahead policy is easy to compute and implement. We then use this policy in the context of localizing mitochondria in electron microscope images, and experimentally show that significant speed ups in acquisition can be gained, while maintaining near equal image quality at target locations, when compared to current policies.

Methods for tracking an object have generally fallen into two groups: tracking by detection and tracking through local optimization. The advantage of detection-based tracking is its ability to deal with target appearance and disappearance, but it does not naturally take advantage of target motion continuity during detection. The advantage of local optimization is efficiency and accuracy, but it requires additional algorithms to initialize tracking when the target is lost. To bridge these two approaches, we propose a framework for unified detection and tracking as a time-series Bayesian estimation problem. The basis of our approach is to treat both detection and tracking as a sequential entropy minimization problem, where the goal is to determine the parameters describing a target in each frame. To do this we integrate the Active Testing (AT) paradigm with Bayesian filtering, and this results in a framework capable of both detecting and tracking robustly in situations where the target object enters and leaves the field of view regularly. We demonstrate our approach on a retinal tool tracking problem and show through extensive experiments that our method provides an efficient and robust tracking solution.

We provide a novel search technique which uses a hierarchical model and a mutual information gain heuristic to efficiently prune the search space when localizing faces in images. We show exponential gains in computation over traditional sliding window approaches, while keeping similar performance levels.

Yeast two-hybrid screens are an important method for mapping pairwise physical interactions between proteins. The fraction of interactions detected in independent screens can be very small, and an outstanding challenge is to determine the reason for the low overlap. Low overlap can arise from either a high false-discovery rate (interaction sets have low overlap because each set is contaminated by a large number of stochastic false-positive interactions) or a high false-negative rate (interaction sets have low overlap because each misses many true interactions). We extend capture-recapture theory to provide the first unified model for false-positive and false-negative rates for two-hybrid screens. Analysis of yeast, worm, and fly data indicates that 25% to 45% of the reported interactions are likely false positives. Membrane proteins have higher false-discovery rates on average, and signal transduction proteins have lower rates. The overall false-negative rate ranges from 75% for worm to 90% for fly, which arises from a roughly 50% false-negative rate due to statistical undersampling and a 55% to 85% false-negative rate due to proteins that appear to be systematically lost from the assays. Finally, statistical model selection conclusively rejects the Erdös-Rényi network model in favor of the power law model for yeast and the truncated power law for worm and fly degree distributions. Much as genome sequencing coverage estimates were essential for planning the human genome sequencing project, the coverage estimates developed here will be valuable for guiding future proteomic screens. All software and datasets are available in and , -, and -, and are also available from our Web site, http://www.baderzone.org.

We propose a new method for estimating the probability mass function (pmf) of a discrete and finite random variable from a small sample. We focus on the observed counts--the number of times each value appears in the sample--and define the maximum likelihood set (MLS) as the set of pmfs that put more mass on the observed counts than on any other set of counts possible for the same sample size. We characterize the MLS in detail in this article. We show that the MLS is a diamond-shaped subset of the probability simplex [0,1]k bounded by at most k x (k-1) hyper-planes, where k is the number of possible values of the random variable. The MLS always contains the empirical distribution, as well as a family of Bayesian estimators based on a Dirichlet prior, particularly the well-known Laplace estimator. We propose to select from the MLS the pmf that is closest to a fixed pmf that encodes prior knowledge. When using Kullback-Leibler distance for this selection, the optimization problem comprises finding the minimum of a convex function over a domain defined by linear inequalities, for which standard numerical procedures are available. We apply this estimate to language modeling using Zipf's law to encode prior knowledge and show that this method permits obtaining state-of-the-art results while being conceptually simpler than most competing methods.

We build a maximum entropy model for skin detection. This model imposes constraints on various marginal distributions. Parameter estimation as well as optimization cannot be tackled without approximations. We propose to use a tree approximation of the pixel lattice. Parameter estimation is then reduced to the estimations of color histograms for neighbor pixels. Moreover, the belief propagation algorithm permits to obtain fast solution for skin probability at pixel locations. We assess the performance on the Compaq database. 1

Visual information retrieval systems use low-level such as color, texture and shape for queries. Users usually have a more abstract notion of what will satisfy them. Using low-level to correspond to high-level abstractions is one aspect of the gap.In this paper, we introduce intermediate features. These are low-level semantic features and level image features. That is, in one hand, they can be arranged to produce high level concept and in another hand, they can be learned from a small annotated database. These can then be used in an retrieval system.We report experiments where intermediate are textures. These are learned from a small annotated database. The resulting indexing procedure is then demonstrated to be superior to a standard color histrogram indexing.

The construction of classification trees is nearly always top-down, locally optimal, and data-driven. Such recursive designs are often globally inefficient, for instance, in terms of the mean depth necessary to reach a given classification rate. We consider statistical models for which exact global optimization is feasible, and thereby demonstrate that recursive and global procedures may result in very different tree graphs and overall performance.

We present a new approach for tracking roads from satellite images, and thereby illustrate a general computational strategy ("active testing") for tracking 1D structures and other recognition tasks in computer vision. Our approach is related to recent work in active vision on "where to look next" and motivated by the "divide-and-conquer" strategy of parlour games. We choose "tests" (matched filters for short road segments) one at a time in order to remove as much uncertainty as possible about the "true hypothesis" (road position) given the results of the previous tests. The tests are chosen online based on a statistical model for the joint distribution of tests and hypotheses. The problem of minimizing uncertainty (measured by entropy) is formulated in simple and explicit analytical terms. At each iteration new image data are examined and a new entropy minimization problem is solved (exactly), resulting in a new image location to inspect, and so forth. We report experiments using panchromatic SPOT satellite imagery with a ground resolution of ten meters.