We propose a deep convolutional neural network architecture codenamed Inception that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. By a carefully crafted design, we increased the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the Hebbian principle and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC14 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.

/pdf/going-deeper-with-convolutions-1yobw2o2ds.pdf

Going deeper with convolutions

A survey on deep learning in medical image analysis

Convolutional Neural Networks (CNNs) have been recently employed to solve problems from both the computer vision and medical image analysis fields. Despite their popularity, most approaches are only able to process 2D images while most medical data used in clinical practice consists of 3D volumes. In this work we propose an approach to 3D image segmentation based on a volumetric, fully convolutional, neural network. Our CNN is trained end-to-end on MRI volumes depicting prostate, and learns to predict segmentation for the whole volume at once. We introduce a novel objective function, that we optimise during training, based on Dice coefficient. In this way we can deal with situations where there is a strong imbalance between the number of foreground and background voxels. To cope with the limited number of annotated volumes available for training, we augment the data applying random non-linear transformations and histogram matching. We show in our experimental evaluation that our approach achieves good performances on challenging test data while requiring only a fraction of the processing time needed by other previous methods.

V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation

Remarkable progress has been made in image recognition, primarily due to the availability of large-scale annotated datasets and deep convolutional neural networks (CNNs). CNNs enable learning data-driven, highly representative, hierarchical image features from sufficient training data. However, obtaining datasets as comprehensively annotated as ImageNet in the medical imaging domain remains a challenge. There are currently three major techniques that successfully employ CNNs to medical image classification: training the CNN from scratch, using off-the-shelf pre-trained CNN features, and conducting unsupervised CNN pre-training with supervised fine-tuning. Another effective method is transfer learning, i.e., fine-tuning CNN models pre-trained from natural image dataset to medical image tasks. In this paper, we exploit three important, but previously understudied factors of employing deep convolutional neural networks to computer-aided detection problems. We first explore and evaluate different CNN architectures. The studied models contain 5 thousand to 160 million parameters, and vary in numbers of layers. We then evaluate the influence of dataset scale and spatial image context on performance. Finally, we examine when and why transfer learning from pre-trained ImageNet (via fine-tuning) can be useful. We study two specific computer-aided detection (CADe) problems, namely thoraco-abdominal lymph node (LN) detection and interstitial lung disease (ILD) classification. We achieve the state-of-the-art performance on the mediastinal LN detection, and report the first five-fold cross-validation classification results on predicting axial CT slices with ILD categories. Our extensive empirical evaluation, CNN model analysis and valuable insights can be extended to the design of high performance CAD systems for other medical imaging tasks.

Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning

Humans perceive the three-dimensional structure of the world with apparent ease. However, despite all of the recent advances in computer vision research, the dream of having a computer interpret an image at the same level as a two-year old remains elusive. Why is computer vision such a challenging problem and what is the current state of the art? Computer Vision: Algorithms and Applications explores the variety of techniques commonly used to analyze and interpret images. It also describes challenging real-world applications where vision is being successfully used, both for specialized applications such as medical imaging, and for fun, consumer-level tasks such as image editing and stitching, which students can apply to their own personal photos and videos. More than just a source of recipes, this exceptionally authoritative and comprehensive textbook/reference also takes a scientific approach to basic vision problems, formulating physical models of the imaging process before inverting them to produce descriptions of a scene. These problems are also analyzed using statistical models and solved using rigorous engineering techniques Topics and features: structured to support active curricula and project-oriented courses, with tips in the Introduction for using the book in a variety of customized courses; presents exercises at the end of each chapter with a heavy emphasis on testing algorithms and containing numerous suggestions for small mid-term projects; provides additional material and more detailed mathematical topics in the Appendices, which cover linear algebra, numerical techniques, and Bayesian estimation theory; suggests additional reading at the end of each chapter, including the latest research in each sub-field, in addition to a full Bibliography at the end of the book; supplies supplementary course material for students at the associated website, http://szeliski.org/Book/. Suitable for an upper-level undergraduate or graduate-level course in computer science or engineering, this textbook focuses on basic techniques that work under real-world conditions and encourages students to push their creative boundaries. Its design and exposition also make it eminently suitable as a unique reference to the fundamental techniques and current research literature in computer vision.

/pdf/computer-vision-algorithms-and-applications-25dn6wu83j.pdf

Computer Vision: Algorithms and Applications

http://www.thelancet.com/article/S2589750022001868/pdf

Recommendations for the development and use of imaging test sets to investigate the test performance of artificial intelligence in health screening.

Foundation models are considered a breakthrough in all applications of AI, promising robust and reusable mechanisms for feature extraction, alleviating the need for large amounts of high quality annotated training data for task-specific prediction models. However, foundation models may potentially encode and even reinforce existing biases present in historic datasets. Given the limited ability to scrutinize foundation models, it remains unclear whether the opportunities outweigh the risks in safety critical applications such as clinical decision making. In our statistical bias analysis of a recently published, and publicly accessible chest X-ray foundation model, we found reasons for concern as the model seems to encode protected characteristics including biological sex and racial identity. When used for the downstream application of disease detection, we observed substantial degradation of performance of the foundation model compared to a standard model with specific disparities in protected subgroups. While research into foundation models for healthcare applications is in an early stage, we hope to raise awareness of the risks by highlighting the importance of conducting thorough bias and subgroup performance analyses.

Risk of Bias in Chest X-ray Foundation Models

Low signal-to-noise ratio, imaging artifacts and bone boundaries appearing several millimeters in thickness have hampered the success of ultrasound (US) guided computer assisted orthopedic surgery procedures. In this paper we propose a robust and accurate bone localization method. The proposed approach is based on the enhancement of bone surfaces using the combination of three different local image phase features. The extracted local phase image features are used as an input to an L1 norm-based contextual regularization method for the enhancement of bone shadow regions. During the final stage the enhanced bone features and shadow region information is combined into a dynamic programming solution for the localization of the bone surface data. Qualitative and quantitative validation was performed on 150 in vivo US scans obtained from seven subjects by scanning femur, knee, distal radius and vertebrae bones. Validation against expert segmentation achieved a mean surface localization error of 0.26 mm a 67% improvement over state of

/pdf/computational-methods-and-clinical-applications-in-3y0an7404l.pdf

Computational Methods and Clinical Applications in Musculoskeletal Imaging

We explore whether recent advances in generative adversarial networks (GANs) enable synthesis of realistic medical images that are hard to distinguish from real ones, even by domain experts. High-quality synthetic images can be useful for data augmentation, domain transfer, and out-of-distribution detection. However, generating realistic images is challenging, particularly for Full Field Digital Mammograms (FFDM), due to the highresolution, textural heterogeneity, fine structural details and specific tissue properties. We employ progressive GANs to synthesize mammograms at a resolution of 1280x1024 pixels, the highest reported so far. In order to assess the perceptual realism, we designed a user study where experts are asked to distinguish real and generated images with exciting results.

/pdf/mammogan-high-resolution-synthesis-of-realistic-mammograms-53pijv7pbs.pdf

MammoGAN: High-Resolution Synthesis of Realistic Mammograms

This article discusses how the language of causality can shed new light on the major challenges in machine learning for medical imaging: 1) data scarcity, which is the limited availability of high-quality annotations, and 2) data mismatch, whereby a trained algorithm may fail to generalize in clinical practice. Looking at these challenges through the lens of causality allows decisions about data collection, annotation procedures, and learning strategies to be made (and scrutinized) more transparently. We discuss how causal relationships between images and annotations can not only have profound effects on the performance of predictive models, but may even dictate which learning strategies should be considered in the first place. For example, we conclude that semi-supervision may be unsuitable for image segmentation---one of the possibly surprising insights from our causal analysis, which is illustrated with representative real-world examples of computer-aided diagnosis (skin lesion classification in dermatology) and radiotherapy (automated contouring of tumours). We highlight that being aware of and accounting for the causal relationships in medical imaging data is important for the safe development of machine learning and essential for regulation and responsible reporting. To facilitate this we provide step-by-step recommendations for future studies.

/pdf/causality-matters-in-medical-imaging-3euetu5yno.pdf

Ben Glocker

Papers

Recommendations for the development and use of imaging test sets to investigate the test performance of artificial intelligence in health screening.

Risk of Bias in Chest X-ray Foundation Models

Computational Methods and Clinical Applications in Musculoskeletal Imaging

MammoGAN: High-Resolution Synthesis of Realistic Mammograms

Causality matters in medical imaging.