Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Co-Planar Stereotaxic Atlas of the Human Brain—3-Dimensional Proportional System: An Approach to Cerebral Imaging, J. Talairach, P. Tournoux. Georg Thieme Verlag, New York (1988), 122 pp., 130 figs. DM 268

/pdf/convex-analysisnoer-san-nojin-zhan-nituite-5dl2rbmoyr.pdf

Convex Analysisの二,三の進展について

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

AACR Centennial Conference: Translational Cancer Medicine-- Nov 4-8, 2007; Singapore

PL02-05 

All cancers are due to abnormalities in DNA. The availability of the human genome sequence has led to the proposal that resequencing of cancer genomes will reveal the full complement of somatic mutations and hence all the cancer genes. To explore the nature of the information that will be derived from cancer genome sequencing we have sequenced the coding exons of the family of 518 protein kinases, ~1.3Mb DNA per cancer sample, in 210 cancers of diverse histological types. Despite the screen being directed toward the coding regions of a gene family that has previously been strongly implicated in oncogenesis, the results indicate that the majority of somatic mutations detected are “passengers”. There is considerable variation in the number and pattern of these mutations between individual cancers, indicating substantial diversity of processes of molecular evolution between cancers. The imprints of exogenous mutagenic exposures, mutagenic treatment regimes and DNA repair defects can all be seen in the distinctive mutational signatures of individual cancers. This systematic mutation screen and others have previously yielded a number of cancer genes that are frequently mutated in one or more cancer types and which are now anticancer drug targets (for example BRAF , PIK3CA , and EGFR ). However, detailed analyses of the data from our screen additionally suggest that there exist a large number of additional “driver” mutations which are distributed across a substantial number of genes. It therefore appears that cells may be able to utilise mutations in a large repertoire of potential cancer genes to acquire the neoplastic phenotype. However, many of these genes are employed only infrequently. These findings may have implications for future anticancer drug development.

Patterns of Somatic Mutation in Human Cancer Genomes

Image patch classification is an important task in many different medical imaging applications. In this work, we have designed a customized Convolutional Neural Networks (CNN) with shallow convolution layer to classify lung image patches with interstitial lung disease (ILD). While many feature descriptors have been proposed over the past years, they can be quite complicated and domain-specific. Our customized CNN framework can, on the other hand, automatically and efficiently learn the intrinsic image features from lung image patches that are most suitable for the classification purpose. The same architecture can be generalized to perform other medical image or texture classification tasks.

Medical image classification with convolutional neural network

We introduce in this chapter some fundamental theories for content-based image retrieval. Section 1.1 looks at the development of content-based image retrieval techniques. Then, as the emphasis of this chapter, we introduce in detail in Section 1.2 some widely used methods for visual content descriptions. After that, we briefly address similarity/distance measures between visual features, the indexing schemes, query formation, relevance feedback, and system performance evaluation in Sections 1.3, 1.4 and 1.5. Details of these techniques are discussed in subsequent chapters. Finally, we draw a conclusion in Section 1.6.

Fundamentals of Content-Based Image Retrieval

The availability of medical imaging data from clinical archives, research literature, and clinical manuals, coupled with recent advances in computer vision offer the opportunity for image-based diagnosis, teaching, and biomedical research. However, the content and semantics of an image can vary depending on its modality and as such the identification of image modality is an important preliminary step. The key challenge for automatically classifying the modality of a medical image is due to the visual characteristics of different modalities: some are visually distinct while others may have only subtle differences. This challenge is compounded by variations in the appearance of images based on the diseases depicted and a lack of sufficient training data for some modalities. In this paper, we introduce a new method for classifying medical images that uses an ensemble of different convolutional neural network (CNN) architectures. CNNs are a state-of-the-art image classification technique that learns the optimal image features for a given classification task. We hypothesise that different CNN architectures learn different levels of semantic image representation and thus an ensemble of CNNs will enable higher quality features to be extracted. Our method develops a new feature extractor by fine-tuning CNNs that have been initialized on a large dataset of natural images. The fine-tuning process leverages the generic image features from natural images that are fundamental for all images and optimizes them for the variety of medical imaging modalities. These features are used to train numerous multiclass classifiers whose posterior probabilities are fused to predict the modalities of unseen images. Our experiments on the ImageCLEF 2016 medical image public dataset (30 modalities; 6776 training images, and 4166 test images) show that our ensemble of fine-tuned CNNs achieves a higher accuracy than established CNNs. Our ensemble also achieves a higher accuracy than methods in the literature evaluated on the same benchmark dataset and is only overtaken by those methods that source additional training data.

/pdf/an-ensemble-of-fine-tuned-convolutional-neural-networks-for-4oc7swcxuk.pdf

An Ensemble of Fine-Tuned Convolutional Neural Networks for Medical Image Classification.

Spatial and temporal plantar pressure distributions are important and useful measures in footwear evaluation, athletic training, clinical gait analysis, and pathology foot diagnosis. However, present plantar pressure measurement and analysis systems are more or less uncomfortable to wear and expensive. This paper presents an in-shoe plantar pressure measurement and analysis system based on a textile fabric sensor array, which is soft, light, and has a high-pressure sensitivity and a long service life. The sensors are connected with a soft polymeric board through conductive yarns and integrated into an insole. A stable data acquisition system interfaces with the insole, wirelessly transmits the acquired data to remote receiver through Bluetooth path. Three configuration modes are incorporated to gain connection with desktop, laptop, or smart phone, which can be configured to comfortably work in research laboratories, clinics, sport ground, and other outdoor environments. A real-time display and analysis software is presented to calculate parameters such as mean pressure, peak pressure, center of pressure (COP), and shift speed of COP. Experimental results show that this system has stable performance in both static and dynamic measurements.

/pdf/in-shoe-plantar-pressure-measurement-and-analysis-system-3hvir6oioe.pdf

In-Shoe Plantar Pressure Measurement and Analysis System Based on Fabric Pressure Sensing Array

In the field of saliency detection, many graph-based algorithms heavily depend on the accuracy of the pre-processed superpixel segmentation, which leads to significant sacrifice of detail information from the input image. In this paper, we propose a novel bottom-up saliency detection approach that takes advantage of both region-based features and image details. To provide more accurate saliency estimations, we first optimize the image boundary selection by the proposed erroneous boundary removal. By taking the image details and region-based estimations into account, we then propose the regularized random walks ranking to formulate pixel-wised saliency maps from the superpixel-based background and foreground saliency estimations. Experiment results on two public datasets indicate the significantly improved accuracy and robustness of the proposed algorithm in comparison with 12 state-of-the-art saliency detection approaches.

/pdf/robust-saliency-detection-via-regularized-random-walks-2nzdzc4isv.pdf

David Dagan Feng

Papers

Medical image classification with convolutional neural network

Fundamentals of Content-Based Image Retrieval

An Ensemble of Fine-Tuned Convolutional Neural Networks for Medical Image Classification.

In-Shoe Plantar Pressure Measurement and Analysis System Based on Fabric Pressure Sensing Array

Robust saliency detection via regularized random walks ranking