Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

A general non-parametric technique is proposed for the analysis of a complex multimodal feature space and to delineate arbitrarily shaped clusters in it. The basic computational module of the technique is an old pattern recognition procedure: the mean shift. For discrete data, we prove the convergence of a recursive mean shift procedure to the nearest stationary point of the underlying density function and, thus, its utility in detecting the modes of the density. The relation of the mean shift procedure to the Nadaraya-Watson estimator from kernel regression and the robust M-estimators; of location is also established. Algorithms for two low-level vision tasks discontinuity-preserving smoothing and image segmentation - are described as applications. In these algorithms, the only user-set parameter is the resolution of the analysis, and either gray-level or color images are accepted as input. Extensive experimental results illustrate their excellent performance.

Mean shift: a robust approach toward feature space analysis

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

A survey on deep learning in medical image analysis

Computer vision applications have come to rely increasingly on superpixels in recent years, but it is not always clear what constitutes a good superpixel algorithm. In an effort to understand the benefits and drawbacks of existing methods, we empirically compare five state-of-the-art superpixel algorithms for their ability to adhere to image boundaries, speed, memory efficiency, and their impact on segmentation performance. We then introduce a new superpixel algorithm, simple linear iterative clustering (SLIC), which adapts a k-means clustering approach to efficiently generate superpixels. Despite its simplicity, SLIC adheres to boundaries as well as or better than previous methods. At the same time, it is faster and more memory efficient, improves segmentation performance, and is straightforward to extend to supervoxel generation.

/pdf/slic-superpixels-compared-to-state-of-the-art-superpixel-5d8sq04am9.pdf

SLIC Superpixels Compared to State-of-the-Art Superpixel Methods

Since it is hard to handcraft the prior knowledge in a shape detection framework, machine learning methods are preferred to exploit the expert annotation of the target shape in a database. In the previous approaches [1, 2], an optimal similarity transformation is exhaustively searched for to maximize the response of a trained classification model. At best, these approaches only give a rough estimate of the position of a non-rigid shape. In this paper, we propose a novel machine learning based approach to achieve a refined shape detection result. We train a model that has the largest response on a reference shape and a smaller response on other shapes. During shape detection, we search for an optimal non-rigid deformation to maximize the response of the trained model on the deformed image block. Since exhaustive searching is inapplicable for a non-rigid deformation space with a high dimension, currently, example based searching is used instead. Experiments on two applications, left ventricle endocardial border detection and facial feature detection, demonstrate the robustness of our approach. It outperforms the well-known ASM and AAM approaches on challenging samples.

Example Based Non-rigid Shape Detection

Cardiac magnetic resonance imaging (MRI) has advanced to become a powerful tool in clinical practice. Extraction of morphological and functional features from cardiac MR imaging for diagnosis and disease monitoring remains a time-consuming task for clinicians. We present a fully automatic approach to extracting the structures and dynamics for both left and right ventricles. The cine short-axis stack of a cardiac MR scan is used to reconstruct a 3D volume sequence. A joint LV-RV model is introduced to delineate the boundaries of both left and right ventricles in each frame, and to combine both spatial and temporal context to track the chamber boundary motion over cardiac cycles. Both qualitative and quantitative results show promise of the proposed method.

/pdf/automatic-delineation-of-left-and-right-ventricles-in-2gaznv3n5a.pdf

Automatic Delineation of Left and Right Ventricles in Cardiac MRI Sequences Using a Joint Ventricular Model

A system (100) for computer aided detection and decision support includes an ontology of image representations (101) for injecting meaning into and adding relationships among image contents, an image understanding and parsing module (103) in communication with the ontology of image representations (101) for extracting structures from an image (104) including the image contents, and a reasoning engine (102) in communication with the ontology and the image understanding and parsing module (103) for classifying the image contents, wherein the system (100) receives the image (104) and corresponding descriptive information (105).

Computer Aided Detection and Decision Support

Building accurate and robust artificial intelligence systems for medical image assessment requires not only the research and design of advanced deep learning models but also the creation of large and curated sets of annotated training examples. Constructing such datasets, however, is often very costly – due to the complex nature of annotation tasks and the high level of expertise required for the interpretation of medical images (e.g., expert radiologists). To counter this limitation, we propose a method for self-supervised learning of rich image features based on contrastive learning and online feature clustering. For this purpose we leverage large training datasets of over 100,000,000 medical images of various modalities, including radiography, computed tomography (CT), magnetic resonance (MR) imaging and ultrasonography. We propose to use these features to guide model training in supervised and hybrid self-supervised/supervised regime on various downstream tasks. We highlight a number of advantages of this strategy on challenging image assessment problems in radiography, CT and MR: 1) Significant increase in accuracy compared to the state-of-theart (e.g., AUC boost of 3-7% for detection of abnormalities from chest radiography scans and hemorrhage detection on brain CT); 2) Acceleration of model convergence during training by up to 85% compared to using no pretraining (e.g., 83% when training a model for detection of brain metastases in MR scans); 3) Increase in robustness to various image augmentations, such as intensity variations, rotations or scaling reflective of data variation seen in the field.

Self-supervised Learning from 100 Million Medical Images

We propose a solution to the problem of object recognition given a continuous video sequence containing multiple views of an object. Initially, object models are acquired from images of the objects taken from different views. Recognition is achieved from the video sequences by employing a multiple hypothesis approach. Appearance similarity, and pose transition smoothness constraints are used to estimate the probability of the measurement being generated from a certain model hypothesis at each time instant. A smooth gradient direction feature that is quasiinvariant to illumination changes and noise is used to represent the appearance of object. The pose of the object at each time instant is modelled as a von Mises-Fisher distribution. Recognition is achieved by choosing the hypothesis set that has accumulated the maximum evidence at the end of the sequence. We have performed detailed experiments demonstrating the viability of the proposed approach.

Dorin Comaniciu

Papers

Example Based Non-rigid Shape Detection

Automatic Delineation of Left and Right Ventricles in Cardiac MRI Sequences Using a Joint Ventricular Model

Computer Aided Detection and Decision Support

Self-supervised Learning from 100 Million Medical Images

A probabilistic framework for object recognition in video