arXiv subjects

Alison Pouch

Publications and source records attributed to Alison Pouch.

9 recordsLinked to original sources

arXiv2026

Bayesian Posterior Sampling for Synthetic Shape Generation of Heart Valves

Statistical shape models (SSMs) for heart valves commonly rely on principal component analysis (PCA). They are used to support downstream tasks, including \textit{in silico} modeling, morphological analysis, and interventional planning. However, PCA-based SSMs lack a mechanism for conditional shape generation, i.e., they can create non-physical shapes and perform poorly in low-data regimes (<20 shapes). To overcome these problems, we propose instead a Bayesian posterior sampling framework to generate valve shapes from a posterior estimate. The prior relies on a Gaussian mixture model with data-driven mixture modes. The likelihood estimate is obtained through a classifier trained to distinguish valid from invalid regions in the compact proper orthogonal decomposition (POD) coefficient space. We verify the framework on a model problem and validate it on parametrically constructed aortic valve datasets. Thereby, we demonstrate that our method captures multiple modes, respects decision boundaries in shape space, and outperforms PCA-based SSMs in low-data regimes. We also characterize the framework's performance as a function of dataset size, identifying where diminishing returns arise for the proposed generative shape model. Finally, we apply the framework to a cohort of ten three-dimensional transesophageal echocardiography images of adult human tricuspid valves. We first segment images to extract shapes, then generate a set of physiologically plausible new shapes. We demonstrate downstream applications for both valves, including \textit{in silico} modeling of valve mechanics and synthetic image-mask creation to augment limited datasets. The proposed approach bootstraps building image-mask datasets more efficiently than PCA-based SSMs. Although demonstrated only for the aortic and tricuspid valves, the methodology is broadly applicable to all valves.

cs.CE↗

arXiv2026

EchoVQA: Enabling Conversational Assistance for Point-of-Care Cardiac Ultrasound

Point-of-care transthoracic echocardiography (TTE) enables cardiac assessment in virtually any clinical setting, yet its diagnostic utility remains constrained by the expertise required for image acquisition and interpretation. Visual question answering (VQA) offers a promising paradigm for bridging this expertise gap through interactive clinical assistance, but existing echocardiography VQA datasets are limited in scale, restricted to high-quality images, and only cover a few views. We introduce EchoVQA, the first large-scale VQA dataset for echocardiography, comprising 14,299 images and 74,819 question-answer pairs. The dataset integrates public sources (EchoNet-Dynamic, CAMUS) with our own point-of-care acquisitions from two handheld probes (Lumify, Clarius), spanning diverse views and including both high-quality and suboptimal images. Uniquely, EchoVQA includes acquisition guidance questions to help users optimize transducer positioning toward a diagnostic apical 4-chamber view for left ventricular ejection fraction estimation -- a challenging task for novice operators in point-of-care settings. We further develop a parameter-efficient method based on multimodal learnable prompts achieving state-of-the-art performance on most benchmarks, including EchoVQA, with significantly less trainable parameters than existing state-of-the-art approaches.

cs.CV↗

arXiv2026

Biomechanically Informed Image Registration for Patient-Specific Aortic Valve Strain Analysis

Aortic valve (AV) biomechanics play a critical role in maintaining normal cardiac function. Pathological variations, particularly in bicuspid aortic valves, alter leaflet loading, increase strain, and accelerate disease progression. Accurate patient-specific characterization of valve geometry and deformation is therefore essential for predicting disease progression and guiding durable repair. Current imaging and computational methods often fail to capture rapid valve motion and complex patient-specific features, limiting precise biomechanical assessment. To address these limitations, we developed an image registration framework coupled with the finite element method (FEM) to improve AV tracking and biomechanical evaluation. The valve geometries derived from 4D echocardiography and CT were used to simulate AV closure and generate intermediate deformation states. These FEM-generated states facilitated leaflet tracking, while image registration corrected misalignment between simulations and imaging data. In 20 patients, FEM-augmented registration improved accuracy by 40% compared with direct registration. This improvement enabled more reliable strain estimation by measuring leaflet deformation directly from imaging and reducing uncertainties associated with boundary conditions and material assumptions. Areal, Green-Lagrange, and deviatoric strains were quantified in adult trileaflet/bicuspid valves, as well as pediatric patients, revealing distinct deformation patterns across valve groups. Convergence in mean deviatoric strain between adult trileaflet and pediatric valves suggests volumetric deformation underlies age- and size-related differences in AV mechanics. The FEM-augmented registration enhances tracking and biomechanical evaluation accuracy, providing clinically relevant insights into patient-specific AV deformation to support individualized medical and intervention planning.

q-bio.QM↗

arXiv2025

Increasing LLM response trustworthiness using voting ensembles

Despite huge advances, LLMs still lack convenient and reliable methods to quantify the uncertainty in their responses, making them difficult to trust in high-stakes applications. One of the simplest approaches to eliciting more accurate answers is to select the mode of many responses, a technique known as ensembling. In this work, we expand on typical ensembling approaches by looking at ensembles with a variable voting threshold. We introduce a theoretical framework for question answering and show that, by permitting ensembles to "abstain" from providing an answer when the dominant response falls short of the threshold, it is possible to dramatically increase the trustworthiness of the remaining answers. From this framework, we derive theoretical results as well as report experimental results on two problem domains: arithmetic problem solving and clinical-note question-answering. In both domains, we observe that large gains in answer trustworthiness can be achieved using highly restrictive voting ensembles, while incurring relatively modest reductions in response yield and accuracy. Due to this quality, voting ensembles may be particularly useful in applications - such as healthcare and data annotation - that require a high degree of certainty but which may not require that every question receive an automated answer.

cs.AI↗

arXiv2024

SynStitch: a Self-Supervised Learning Network for Ultrasound Image Stitching Using Synthetic Training Pairs and Indirect Supervision

Ultrasound (US) image stitching can expand the field-of-view (FOV) by combining multiple US images from varied probe positions. However, registering US images with only partially overlapping anatomical contents is a challenging task. In this work, we introduce SynStitch, a self-supervised framework designed for 2DUS stitching. SynStitch consists of a synthetic stitching pair generation module (SSPGM) and an image stitching module (ISM). SSPGM utilizes a patch-conditioned ControlNet to generate realistic 2DUS stitching pairs with known affine matrix from a single input image. ISM then utilizes this synthetic paired data to learn 2DUS stitching in a supervised manner. Our framework was evaluated against multiple leading methods on a kidney ultrasound dataset, demonstrating superior 2DUS stitching performance through both qualitative and quantitative analyses. The code will be made public upon acceptance of the paper.

eess.IV↗

arXiv2024

PRISM Lite: A lightweight model for interactive 3D placenta segmentation in ultrasound

Placenta volume measured from 3D ultrasound (3DUS) images is an important tool for tracking the growth trajectory and is associated with pregnancy outcomes. Manual segmentation is the gold standard, but it is time-consuming and subjective. Although fully automated deep learning algorithms perform well, they do not always yield high-quality results for each case. Interactive segmentation models could address this issue. However, there is limited work on interactive segmentation models for the placenta. Despite their segmentation accuracy, these methods may not be feasible for clinical use as they require relatively large computational power which may be especially prohibitive in low-resource environments, or on mobile devices. In this paper, we propose a lightweight interactive segmentation model aiming for clinical use to interactively segment the placenta from 3DUS images in real-time. The proposed model adopts the segmentation from our fully automated model for initialization and is designed in a human-in-the-loop manner to achieve iterative improvements. The Dice score and normalized surface Dice are used as evaluation metrics. The results show that our model can achieve superior performance in segmentation compared to state-of-the-art models while using significantly fewer parameters. Additionally, the proposed model is much faster for inference and robust to poor initial masks. The code is available at https://github.com/MedICL-VU/PRISM-placenta.

eess.IV↗

arXiv2024

Interactive Segmentation Model for Placenta Segmentation from 3D Ultrasound images

Placenta volume measurement from 3D ultrasound images is critical for predicting pregnancy outcomes, and manual annotation is the gold standard. However, such manual annotation is expensive and time-consuming. Automated segmentation algorithms can often successfully segment the placenta, but these methods may not consistently produce robust segmentations suitable for practical use. Recently, inspired by the Segment Anything Model (SAM), deep learning-based interactive segmentation models have been widely applied in the medical imaging domain. These models produce a segmentation from visual prompts provided to indicate the target region, which may offer a feasible solution for practical use. However, none of these models are specifically designed for interactively segmenting 3D ultrasound images, which remain challenging due to the inherent noise of this modality. In this paper, we evaluate publicly available state-of-the-art 3D interactive segmentation models in contrast to a human-in-the-loop approach for the placenta segmentation task. The Dice score, normalized surface Dice, averaged symmetric surface distance, and 95-percent Hausdorff distance are used as evaluation metrics. We consider a Dice score of 0.95 a successful segmentation. Our results indicate that the human-in-the-loop segmentation model reaches this standard. Moreover, we assess the efficiency of the human-in-the-loop model as a function of the amount of prompts. Our results demonstrate that the human-in-the-loop model is both effective and efficient for interactive placenta segmentation. The code is available at \url{https://github.com/MedICL-VU/PRISM-placenta}.

cs.CV↗

arXiv2024

SAM3D: Zero-Shot Semi-Automatic Segmentation in 3D Medical Images with the Segment Anything Model

We introduce SAM3D, a new approach to semi-automatic zero-shot segmentation of 3D images building on the existing Segment Anything Model. We achieve fast and accurate segmentations in 3D images with a four-step strategy involving: user prompting with 3D polylines, volume slicing along multiple axes, slice-wide inference with a pretrained model, and recomposition and refinement in 3D. We evaluated SAM3D performance qualitatively on an array of imaging modalities and anatomical structures and quantify performance for specific structures in abdominal pelvic CT and brain MRI. Notably, our method achieves good performance with zero model training or finetuning, making it particularly useful for tasks with a scarcity of preexisting labeled data. By enabling users to create 3D segmentations of unseen data quickly and with dramatically reduced manual input, these methods have the potential to aid surgical planning and education, diagnostic imaging, and scientific research.

eess.IV↗

arXiv2019

Diffeomorphic Medial Modeling

Deformable shape modeling approaches that describe objects in terms of their medial axis geometry (e.g., m-reps [Pizer et al., 2003]) yield rich geometrical features that can be useful for analyzing the shape of sheet-like biological structures, such as the myocardium. We present a novel shape analysis approach that combines the benefits of medial shape modeling and diffeomorphometry. Our algorithm is formulated as a problem of matching shapes using diffeomorphic flows under constraints that approximately preserve medial axis geometry during deformation. As the result, correspondence between the medial axes of similar shapes is maintained. The approach is evaluated in the context of modeling the shape of the left ventricular wall from 3D echocardiography images.

cs.GR↗