There is, I think, something ethereal about i —the square root of minus one. I remember first hearing about it at school. It seemed an odd beast at that time—an intruder hovering on the edge of reality.

Usually familiarity dulls this sense of the bizarre, but in the case of i it was the reverse: over the years the sense of its surreal nature intensified. It seemed that it was impossible to write mathematics that described the real world in …

I and i

[신간의 별자리x] 우리/미술, 그리고 ‘슬픔의 박물관’

From the Publisher: 
The accessible presentation of this book gives both a general view of the entire computer vision enterprise and also offers sufficient detail to be able to build useful applications. Users learn techniques that have proven to be useful by first-hand experience and a wide range of mathematical methods. A CD-ROM with every copy of the text contains source code for programming practice, color images, and illustrative movies. Comprehensive and up-to-date, this book includes essential topics that either reflect practical significance or are of theoretical importance. Topics are discussed in substantial and increasing depth. Application surveys describe numerous important application areas such as image based rendering and digital libraries. Many important algorithms broken down and illustrated in pseudo code. Appropriate for use by engineers as a comprehensive reference to the computer vision enterprise.

Computer vision : a modern approach = 计算机视觉 : 一种现代的方法

Biomedical imaging is a driver of scientific discovery and a core component of medical care and is being stimulated by the field of deep learning. While semantic segmentation algorithms enable image analysis and quantification in many applications, the design of respective specialized solutions is non-trivial and highly dependent on dataset properties and hardware conditions. We developed nnU-Net, a deep learning-based segmentation method that automatically configures itself, including preprocessing, network architecture, training and post-processing for any new task. The key design choices in this process are modeled as a set of fixed parameters, interdependent rules and empirical decisions. Without manual intervention, nnU-Net surpasses most existing approaches, including highly specialized solutions on 23 public datasets used in international biomedical segmentation competitions. We make nnU-Net publicly available as an out-of-the-box tool, rendering state-of-the-art segmentation accessible to a broad audience by requiring neither expert knowledge nor computing resources beyond standard network training.

nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation

The volume and complexity of diagnostic imaging is increasing at a pace faster than the availability of human expertise to interpret it. Artificial intelligence has shown great promise in classifying two-dimensional photographs of some common diseases and typically relies on databases of millions of annotated images. Until now, the challenge of reaching the performance of expert clinicians in a real-world clinical pathway with three-dimensional diagnostic scans has remained unsolved. Here, we apply a novel deep learning architecture to a clinically heterogeneous set of three-dimensional optical coherence tomography scans from patients referred to a major eye hospital. We demonstrate performance in making a referral recommendation that reaches or exceeds that of experts on a range of sight-threatening retinal diseases after training on only 14,884 scans. Moreover, we demonstrate that the tissue segmentations produced by our architecture act as a device-independent representation; referral accuracy is maintained when using tissue segmentations from a different type of device. Our work removes previous barriers to wider clinical use without prohibitive training data requirements across multiple pathologies in a real-world setting.

/pdf/clinically-applicable-deep-learning-for-diagnosis-and-56r182cw0z.pdf

Clinically applicable deep learning for diagnosis and referral in retinal disease

Automated delineation of organs play a vital role during the treatment planning for clinical practitioner. With magnetic resonance (MR) images being widely used in radio-therapy, the organ delineation from multi-sequence MR images has become more desirable. Segmenting abdominal organs from MR images is critical in surgical planning. However, the organs segmentation from MR volume is a challenging task, as there are different visual appearances for the same organ in the different sequences of MR images. Recently, convolutional neural network (CNN) methods have achieved promising results in automating organ delineation. However, the performance of deep learning (DL) to emerging approaches such as adversarial learning in the field of segmentation are rarely explored. In this paper, we have trained an encoder-decoder like architecture as a segmentation network, using segmentation loss minimization along with training it to adversarial attack another CNN implementing the visual Turing test discriminator. This way of adversarial training has shown significant improvement in performance. We have used the CHAOS-19 dataset of abdominal MRI for segmentation of the liver, left kidney, right kidney, and spleen. We have used Dice coefficient (DC) to measure the performance, and achieved an average DC of 0.97 with the faster convergence.

Adversarial Training of Deep Convolutional Neural Network for Multi-organ Segmentation from Multi-sequence MRI of the Abdomen

Motor imagery (MI) based brain-computer interface (BCI) plays a crucial role in various scenarios ranging from post-traumatic rehabilitation to control prosthetics. Computer-aided interpretation of MI has augmented prior mentioned scenarios since decades but failed to address interpersonal variability. Such variability further escalates in case of multi-class MI, which is currently a common practice. The failures due to interpersonal variability can be attributed to handcrafted features as they failed to extract more generalized features. The proposed approach employs convolution neural network (CNN) based model with both filtering (through axis shuffling) and feature extraction to avail end-to-end training. Axis shuffling is performed adopted in initial blocks of the model for 1D preprocessing and reduce the parameters required. Such practice has avoided the overfitting which resulted in an improved generalized model. Publicly available BCI Competition-IV 2a dataset is considered to evaluate the proposed model. The proposed model has demonstrated the capability to identify subject-specific frequency band with an average and highest accuracy of 70.5% and S3.6% respectively. Proposed CNN model can classify in real time without relying on accelerated computing device like GPU.

A Deep Convolutional Neural Network Based Classification Of Multi-Class Motor Imagery With Improved Generalization

In this work, we cast a multi-modal immunofluorescence to histology registration problem is cast to a mono-modal one by iteratively propagating tissue-specific features from one modality to generate intensities in the other modality via an implicitly learnt random-forest regression framework. The proposed method iterates between modality propagation and image registration in a unified formulation. Evaluations on real and simulated tissue deformations establish superiority of the proposed work over comparative methods in handling highly complex intermodal intensity relationships. This framework will aid in quantitative analysis of the tissue structure, functionality thus providing for co-located molecular validations and bringing in spatial fidelity to histological assessment.

Deformable registration of immunofluorescence and histology using iterative cross-modal propagation

Feature extraction using linguistic abstracts described by field experts, and their pragmatic behavior when tested
with an inference engine is of interest to computer vision researchers. Advances in image processing have added
to the complexity involved with selecting an appropriate feature extraction method for describing a linguistic
feature. In this work, we propose the usage of a set of statistical tools for evaluating the efficacy of a feature
extraction technique suitable for expressing a linguistic feature. This set of tools are based on expression of class
discrimination strength of features, overlap in their expression, and the density of outliers present in them. The
feature extraction techniques are ranked based on the scores obtained by them when tested with these tools. An
experimental study for validating these claims, based on classification of two different visual texture, expressed
using six different texture quantification techniques is also presented.

Statistical tools for evaluating classification efficacy of feature extraction techniques

/pdf/distill-dsm-computationally-efficient-method-for-2au5g70kgj.pdf

Debdoot Sheet

Papers

Adversarial Training of Deep Convolutional Neural Network for Multi-organ Segmentation from Multi-sequence MRI of the Abdomen

A Deep Convolutional Neural Network Based Classification Of Multi-Class Motor Imagery With Improved Generalization

Deformable registration of immunofluorescence and histology using iterative cross-modal propagation

Statistical tools for evaluating classification efficacy of feature extraction techniques

Distill DSM: Computationally efficient method for segmentation of medical imaging volumes