Hallucination Index: An Image Quality Metric for Generative Reconstruction Models

被引:0
|
作者
Tivnan, Matthew
Yoon, Siyeop
Chen, Zhennong
Li, Xiang
Wu, Dufan
Li, Quanzheng [1 ]
机构
[1] Massachusetts Gen Hosp, 55 Fruit St, Boston, MA 02114 USA
来源
MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2024, PT X | 2024年 / 15010卷
关键词
Hallucinations; Uncertainty; Diffusion Models; Generative Models; Medical Image Reconstruction;
D O I
10.1007/978-3-031-72117-5_42
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Generative image reconstruction algorithms such as measurement conditioned diffusion models are increasingly popular in the field of medical imaging. These powerful models can transform low signal-to-noise ratio (SNR) inputs into outputs with the appearance of high SNR. However, the outputs can have a new type of error called hallucinations. In medical imaging, these hallucinations may not be obvious to a Radiologist but could cause diagnostic errors. Generally, hallucination refers to error in estimation of object structure caused by a machine learning model, but there is no widely accepted method to evaluate hallucination magnitude. In this work, we propose a new image quality metric called the hallucination index. Our approach is to compute the Hellinger distance from the distribution of reconstructed images to a zero hallucination reference distribution. To evaluate our approach, we conducted a numerical experiment with electron microscopy images, simulated noisy measurements, and applied diffusion based reconstructions. We sampled the measurements and the generative reconstructions repeatedly to compute the sample mean and covariance. For the zero hallucination reference, we used the forward diffusion process applied to ground truth. Our results show that higher measurement SNR leads to lower hallucination index for the same apparent image quality. We also evaluated the impact of early stopping in the reverse diffusion process and found that more modest denoising strengths can reduce hallucination. We believe this metric could be useful for evaluation of generative image reconstructions or as a warning label to inform radiologists about the degree of hallucinations in medical images.
引用
收藏
页码:449 / 458
页数:10
相关论文
共 50 条
  • [1] Image feature index: A novel metric for quantifying chest radiographic image quality
    Liang, Zejun
    Tang, Jing
    Xu, Peng
    Zeng, Wen
    Zhang, Jiye
    Zhang, Yu
    Zeng, Lingming
    Wang, Hairong
    Xia, Chunchao
    Li, Zhenlin
    MEDICAL PHYSICS, 2023, 50 (05) : 2805 - 2815
  • [2] High Quality Image Reconstruction of Point Models
    Marroquim, Ricardo
    Oliveira, Antonio
    Cavalcanti, Paulo Roma
    SIBGRAPI 2008: XXI BRAZILIAN SYMPOSIUM ON COMPUTER GRAPHICS AND IMAGE PROCESSING, 2008, : 297 - 304
  • [3] The Infinite Index: Information Retrieval on Generative Text-To-Image Models
    Deckers, Niklas
    Froebe, Maik
    Kiesel, Johannes
    Pandolfo, Gianluca
    Schroeder, Christopher
    Stein, Benno
    Potthast, Martin
    PROCEEDINGS OF THE 2023 CONFERENCE ON HUMAN INFORMATION INTERACTION AND RETRIEVAL, CHIIR 2023, 2023, : 172 - 186
  • [4] A Metric for Evaluating Image Quality Difference Perception Ability in Blind Image Quality Assessment Models
    Zhu, Jinchi
    Li, Yuying
    Zhao, Yidan
    Lin, Qiang
    Zhang, Suiyu
    Ma, Xiaoyu
    Yu, Dingguo
    PROCEEDINGS OF THE 3RD WORKSHOP ON QUALITY OF EXPERIENCE IN VISUAL MULTIMEDIA APPLICATIONS, QOEVMA 2024, 2024, : 12 - 20
  • [5] Image quality metric based on multidimensional contrast perception models
    Pons, AM
    Malo, J
    Artigas, JM
    Capilla, P
    DISPLAYS, 1999, 20 (02) : 93 - 110
  • [6] Hallucination Detection for Generative Large Language Models by Bayesian Sequential Estimation
    Wang, Xiaohua
    Yan, Yuliang
    Huang, Longtao
    Zheng, Xiaoqing
    Huang, Xuanjing
    2023 CONFERENCE ON EMPIRICAL METHODS IN NATURAL LANGUAGE PROCESSING (EMNLP 2023), 2023, : 15361 - 15371
  • [7] Generative Image Reconstruction From Gradients
    Sotthiwat, Ekanut
    Zhen, Liangli
    Zhang, Chi
    Li, Zengxiang
    Goh, Rick Siow Mong
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2025, 36 (01) : 21 - 31
  • [8] Halftone Image Quality Evaluation Based on Reconstruction Index Model
    Zhou, Xiao
    Shi, Ruizhi
    Li, Da
    Li, Shenghui
    Wang, Yusheng
    ADVANCED GRAPHIC COMMUNICATIONS, PACKAGING TECHNOLOGY AND MATERIALS, 2016, 369 : 231 - 240
  • [9] Perceptual Quality Metric With Internal Generative Mechanism
    Wu, Jinjian
    Lin, Weisi
    Shi, Guangming
    Liu, Anmin
    IEEE TRANSACTIONS ON IMAGE PROCESSING, 2013, 22 (01) : 43 - 54
  • [10] Characterizations of image quality for CT with iterative reconstruction algorithm - development of a routine quality control metric
    Yang, Chang-Ying Joseph
    Bjarnason, Thorarin Albert
    MEDICAL PHYSICS, 2014, 41 (08) : 7 - 7