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

Training a deep convolutional neural network (CNN) from scratch is difficult because it requires a large amount of labeled training data and a great deal of expertise to ensure proper convergence. A promising alternative is to fine-tune a CNN that has been pre-trained using, for instance, a large set of labeled natural images. However, the substantial differences between natural and medical images may advise against such knowledge transfer. In this paper, we seek to answer the following central question in the context of medical image analysis: Can the use of pre-trained deep CNNs with sufficient fine-tuning eliminate the need for training a deep CNN from scratch? To address this question, we considered four distinct medical imaging applications in three specialties (radiology, cardiology, and gastroenterology) involving classification, detection, and segmentation from three different imaging modalities, and investigated how the performance of deep CNNs trained from scratch compared with the pre-trained CNNs fine-tuned in a layer-wise manner. Our experiments consistently demonstrated that 1) the use of a pre-trained CNN with adequate fine-tuning outperformed or, in the worst case, performed as well as a CNN trained from scratch; 2) fine-tuned CNNs were more robust to the size of training sets than CNNs trained from scratch; 3) neither shallow tuning nor deep tuning was the optimal choice for a particular application; and 4) our layer-wise fine-tuning scheme could offer a practical way to reach the best performance for the application at hand based on the amount of available data.

Convolutional Neural Networks for Medical Image Analysis: Full Training or Fine Tuning?

MUCKE aims to mine a large volume of images, to structure them conceptually and to use this conceptual structuring in order to improve large-scale image retrieval. The last decade witnessed important progress concerning low-level image representations. However, there are a number problems which need to be solved in order to unleash the full potential of image mining in applications. The central problem with low-level representations is the mismatch between them and the human interpretation of image content. This problem can be instantiated, for instance, by the incapability of existing descriptors to capture spatial relationships between the concepts represented or by their incapability to convey an explanation of why two images are similar in a content-based image retrieval framework. We start by assessing existing local descriptors for image classification and by proposing to use co-occurrence matrices to better capture spatial relationships in images. The main focus in MUCKE is on cleaning large scale Web image corpora and on proposing image representations which are closer to the human interpretation of images. Consequently, we introduce methods which tackle these two problems and compare results to state of the art methods. Note: some aspects of this deliverable are withheld at this time as they are pending review. Please contact the authors for a preview.

http://ieeexplore.ieee.org/iel2/4524/12820/00592460.pdf

Image Processing

IEEE Transactions on Pattern Analysis and Machine Intelligence

The WAVES investigation on the WIND spacecraft will provide comprehensive measurements of the radio and plasma wave phenomena which occur in Geospace. Analyses of these measurements, in coordination with the other onboard plasma, energetic particles, and field measurements will help us understand the kinetic processes that are important in the solar wind and in key boundary regions of the Geospace. These processes are then to be interpreted in conjunction with results from the other ISTP spacecraft in order to discern the measurements and parameters for mass, momentum, and energy flow throughout geospace. This investigation will also contribute to observations of radio waves emitted in regions where the solar wind is accelerated. The WAVES investigation comprises several innovations in this kind of instrumentation: among which the first use, to our knowledge, of neural networks in real-time on board a scientific spacecraft to analyze data and command observation modes, and the first use of a wavelet transform-like analysis in real time to perform a spectral analysis of a broad band signal.

/pdf/waves-the-radio-and-plasma-wave-investigation-on-the-wind-ngfb2pv1t7.pdf

Waves: The Radio and Plasma Wave Investigation on the Wind Spacecraft

Three-dimensional imaging applications require high resolution images that finally result in high data volumes. Due to bandwidth and storage restrictions, an efficient and robust compression scheme must be developed in order to overcome these limitations. This work presents a hardware implementation of a real-time disparity estimation scheme targeted but not limited to integral photography (IP) 3D imaging applications. The proposed system demonstrates an efficient architecture which copes with the increased bandwidth demands that 3D imaging technology requires. Moreover, the system can successfully process high resolution IP video sequences in real-time.

/pdf/fpga-based-architecture-for-real-time-ip-video-and-image-4zirwytnc6.pdf

FPGA-based architecture for real-time IP video and image compression

This work addresses the detection of protein spots in 2D gel electrophoresis images. A novel morphology-based approach is proposed, which utilizes the dilation operator for the localization of regional intensity maxima associated with protein spots. A disk-shaped structuring element (SE) is selected in agreement with the prevalent roundish shape of the majority of protein spots. Thus, spots within rectangular-shaped streaks are correctly detected. SE size is set considering that a certain radius value allows the discrimination of individual spots located in complex spot regions as well as small-sized spots. Moreover, spurious intensity maxima associated with noise are ignored by applying regional intensity constraints. The results of the experimental evaluation lead to the conclusion that the proposed approach detects more actual protein spots and less false spots than a renowned 2D-GE image analysis software package. Furthermore, it does not require user intervention.

/pdf/protein-spot-detection-in-2d-ge-images-using-morphological-40r9b1jo3e.pdf

Protein spot detection in 2D-GE images using morphological operators

Analyzing 2D-gel electrophoresis images remains a challenging task. Amongst the respective stages of the analysis, detection and segmentation of each individual spot are the most crucial. The commercial availability of several software programs and techniques notwithstanding, spot detection and segmentation are extremely resource intensive and frustratingly time-consuming; without human intervention, they fail to detect several protein spots, such as low intensity spots or overlapping protein, while they detect various spurious spots. This paper presents an original approach to detecting and segmenting spots in 2D-gel electrophoresis images. The conducted experiments in a set of 16-bit images demonstrate that the proposed approach is very effective and it outperforms existing techniques even when it is applied to images containing several overlapping spots as well as to images containing spots of various intensities, sizes and shapes.

/pdf/detection-and-segmentation-in-2d-gel-electrophoresis-images-591pj551qv.pdf

Detection and segmentation in 2D gel electrophoresis images

Most processing tasks in three-dimensional (3D) applications share common ground in terms of algorithm design and implementation. Furthermore, time consuming tasks like 3D reconstruction of objects from generic multiview images and compression of massive amounts of 3D image data, require enormous data processing power, and in many cases computations should be performed in real-time. In this paper we present a hardware architecture that jointly addresses both tasks. The proposed design features processing elements and memory modules that form a common compression and reconstruction datapath. Extensive pipelining minimizes throughput delays and memory access operations. Our implementation achieves real-time performance for both tasks, efficiently addressing demanding 3D imaging and video applications.

/pdf/real-time-processing-pipeline-for-3d-imaging-applications-4gz8xilgsz.pdf

Real-time processing pipeline for 3D imaging applications

In this paper we propose a methodology for intelligent analysis of genomic measurements It is based on a sequential scheme of Support Vector Machines and it can be used for class prediction of multiclass genomic samples The proposed methodology was evaluated using two lung cancer datasets The results are comparable and in many cases higher to the accuracy of relevant methodologies that have been proposed in the literature

/pdf/intelligent-analysis-of-genomic-measurements-4ljh6vv1h8.pdf

Dimitris Maroulis

Papers

FPGA-based architecture for real-time IP video and image compression

Protein spot detection in 2D-GE images using morphological operators

Detection and segmentation in 2D gel electrophoresis images

Real-time processing pipeline for 3D imaging applications

Intelligent Analysis of Genomic Measurements