TorchEsegeta: Framework for Interpretability and Explainability of Image-based Deep Learning Models
Otto von Guericke University Magdeburg(马格德堡奥托·冯·格里克大学);German Center for Neurodegenerative Disease(德国神经退行性疾病中心);Center for Behavioral Brain Sciences(行为脑科学中心);Leibniz Institute for Neurobiology(莱布尼茨神经生物学研究所)
Clinicians are often very sceptical about applying automatic image processing approaches, especially deep learning based methods, in practice. One main reason for this is the black-box nature of these approaches and the inherent problem of missing insights of the automatically derived decisions. In order to increase trust in these methods, this paper presents approaches that help to interpret and explain the results of deep learning algorithms by depicting the anatomical areas which influence the decision of the algorithm most. Moreover, this research presents a unified framework, TorchEsegeta, for applying various interpretability and explainability techniques for deep learning models and generate visual interpretations and explanations for clinicians to corroborate their clinical findings. In addition, this will aid in gaining confidence in such methods. The framework builds on existing interpretability and explainability techniques that are currently focusing on classification models, extending them to segmentation tasks. In addition, these methods have been adapted to 3D models for volumetric analysis. The proposed framework provides methods to quantitatively compare visual explanations using infidelity and sensitivity metrics. This framework can be used by data scientists to perform post-hoc interpretations and explanations of their models, develop more explainable tools and present the findings to clinicians to increase their faith in such models. The proposed framework was evaluated based on a use case scenario of vessel segmentation models trained on Time-of-fight (TOF) Magnetic Resonance Angiogram (MRA) images of the human brain. Quantitative and qualitative results of a comparative study of different models and interpretability methods are presented. Furthermore, this paper provides an extensive overview of several existing interpretability and explainability methods.
Exploring Visual Interpretability for Contrastive Language-Image Pre-training
The Hong Kong University of Science and Technology(香港科技大学);University of Technology Sydney(悉尼科技大学)
对比图像预训练(CLIP)通过对自然语言的现成监督来学习丰富的表示。它提高了下游视觉任务的性能,包括但不限于zero-shot、长尾、分割、检索、字幕和视频。然而,CLIP的视觉可解释性研究很少,特别是对于原始特征图。为了提供其预测的视觉解释,我们提出了图像-文本相似性图(ITSM)。在此基础上,我们惊讶地发现,CLIP喜欢的背景区域比前景,并显示错误的可视化结果对人类的理解。这种现象在视觉Transformer和卷积网络中都是普遍存在的,这表明这个问题是独特的,而不是由于特定的网络。通过实验,我们发现问题出在池化部分,不适当的池化方法会导致一种称为语义转换的现象。针对这个问题,我们提出了可解释的对比图像预训练(ECLIP),它通过Masked Max Pooling来纠正可解释性。具体来说,为了避免语义转移,我们用max pooling代替了原来的注意力池,专注于自信的前景,并在训练过程中得到自由注意力的指导。在三个数据集上的实验表明,ECLIP大大提高了CLIP的可解释性,并在很大程度上超过了以前的可解释性方法。代码将在稍后发布。
Exploring the Role of Explainability in AI-Assisted Embryo Selection
Barcelona Supercomputing Center(巴塞罗那超级计算中心);Hospital Clínic de Barcelona(巴塞罗那诊所医院);Universitat Politécnica de Catalunya - Barcelona TECH(加泰罗尼亚理工大学-巴塞罗那TECH)
A Review on Explainability in Multimodal Deep Neural Nets
Symbiosis Institute of Technology (SIT)(共生工程技术学院(SIT));Symbiosis International (Deemed University)(共生国际大学);Symbiosis Centre for Applied Artificial Intelligence(共生应用人工智能中心)
ContrastDiagnosis: Enhancing Interpretability in Lung Nodule Diagnosis Using Contrastive Learning
East China Normal University(华东师范大学);NVIDIA Corp.(英伟达公司);Nagoya University(名古屋大学);National Institute of Informatics(信息学研究所);The First Affiliated Hospital of Nanjing Medical University(南京医科大学第一附属医院)
Interpretability of a Deep Learning Model in the Application of Cardiac MRI Segmentation with an ACDC Challenge Dataset
School of Medicine, University College Dublin(都柏林大学学院医学院);School of Computer Science, University College Dublin(都柏林大学学院计算机科学学院);Department of Statistics, University of Wisconsin–Madison(威斯康星大学麦迪逊分校统计系)
Deep Learning Model Explainability for Inspection Accuracy Improvement in the Automotive Industry
Faurecia Clean Mobility(佛吉亚清洁出行);Ecole Centrale de Lyon(里昂中央理工学院);FEMTO-ST Institute, CNRS(法国国家科学研究中心FEMTO-ST研究所);Univ. Bourgogne Franche-Comte (UBFC)(勃艮第弗朗什-孔泰大学)
Explainability in mulimodal deep transformation models for stroke outcome prediction
University Hospital Zurich, Department of Neurology(苏黎世大学医院神经内科);Zurich University of Applied Sciences, Institute of Data Analysis and Process Design (IDP)(苏黎世应用科学大学数据分析与流程设计研究所);University of Zurich, Epidemiology, Biostatistics and Prevention Institute (EBPI)(苏黎世大学流行病学、生物统计学与预防研究所);Thurgau Institute for Digital Transformation (TIDIT)(图尔高数字转型研究所)
Explaining First Impressions: Modeling, Recognizing, and Explaining Apparent Personality from Videos
University of Barcelona(巴塞罗那大学);Computer Vision Center(计算机视觉中心);UPSud/INRIA, Université Paris-Saclay(巴黎萨克雷大学巴黎南大学/法国国家信息与自动化研究所);ChaLearn(ChaLearn挑战学习组织);Aix Marseille Univ(艾克斯-马赛大学);CNRS(法国国家科学研究中心);Microsoft Research(微软研究院);Boğaziçi University(博阿齐奇大学);INAOE(墨西哥国家天体物理光学与电子研究所);Namık Kemal University(纳默克·凯马尔大学);Nagoya University(名古屋大学);Radboud University(拉德堡德大学);Universitat Oberta de Catalunya(加泰罗尼亚开放大学)