Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Machine learning.”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,801 records · Page 100Linked to original sources
PubMed2004

Overview and fundamentals of urologic robot-integrated systems.

Advances in technology have revolutionized urology. Minimally invasive tools now form the core of the urologist's armamentarium. Laparoscopic surgery has become the favored approach for treating many complicated urologic ailments. Surgical robots represent the next evolutionary step in the fruitful man-machine partnership. The introduction of robotic technology in urology changes how urologists learn, teach, plan, and operate. As technology evolves, robots not only will improve performance in minimally invasive procedures, but also enhance other procedures or enable new kinds of operations.

Equipment Design↗
PubMed2003

Reliable classification of two-class cancer data using evolutionary algorithms.

In the area of bioinformatics, the identification of gene subsets responsible for classifying available disease samples to two or more of its variants is an important task. Such problems have been solved in the past by means of unsupervised learning methods (hierarchical clustering, self-organizing maps, k-mean clustering, etc.) and supervised learning methods (weighted voting approach, k-nearest neighbor method, support vector machine method, etc.). Such problems can also be posed as optimization problems of minimizing gene subset size to achieve reliable and accurate classification. The main difficulties in solving the resulting optimization problem are the availability of only a few samples compared to the number of genes in the samples and the exorbitantly large search space of solutions. Although there exist a few applications of evolutionary algorithms (EAs) for this task, here we treat the problem as a multiobjective optimization problem of minimizing the gene subset size and minimizing the number of misclassified samples. Moreover, for a more reliable classification, we consider multiple training sets in evaluating a classifier. Contrary to the past studies, the use of a multiobjective EA (NSGA-II) has enabled us to discover a smaller gene subset size (such as four or five) to correctly classify 100% or near 100% samples for three cancer samples (Leukemia, Lymphoma, and Colon). We have also extended the NSGA-II to obtain multiple non-dominated solutions discovering as much as 352 different three-gene combinations providing a 100% correct classification to the Leukemia data. In order to have further confidence in the identification task, we have also introduced a prediction strength threshold for determining a sample's belonging to one class or the other. All simulation results show consistent gene subset identifications on three disease samples and exhibit the flexibilities and efficacies in using a multiobjective EA for the gene subset identification task.

Algorithms↗
PubMed2000

Instabilities and oscillation in the deterministic Boltzmann machine.

Simulations indicate that the deterministic Boltzmann machine, unlike the stochastic Boltzmann machine from which it is derived, exhibits unstable behavior during contrastive Hebbian learning of nonlinear problems, including oscillation in the learning algorithm and extreme sensitivity to small weight perturbations. Although careful choice of the initial weight magnitudes, the learning rate, and the annealing schedule will produce convergence in most cases, the stability of the resulting solution depends on the parameters in a complex and generally indiscernible way. We show that this unstable behavior is the result of over parameterization (excessive freedom in the weights), which leads to continuous rather than isolated optimal weight solution sets. This allows the weights to drift without correction by the learning algorithm until the free energy landscape changes in such a way that the settling procedure employed finds a different minimum of the free energy function than it did previously and a gross output error occurs. Because all the weight sets in a continuous optimal solution set produce exactly the same network outputs, we define reliability, a measure of the robustness of the network, as a new performance criterion.

Algorithms↗
PubMedDate not supplied

The prospects for analogue neural VLSI.

In recent years, the efforts of analogue, neural-hardware designers have shifted from generic analogue neurocomputers to "niche" markets in sensor fusion and robotics, and we explain why this is so. We describe the main differences between digital and analogue computation, and consider the advantages of pure analogue and pulsed methods of design. We then investigate some important issues in analogue design of neural machines, namely weight storage (volatile and non-volatile), on-chip learning, and arithmetic accuracy and its relationship to noise. Finally, we outline those areas in which analogue techniques are likely to prove most useful, and speculate as to their likely long-term utility.

Computers, Analog↗
PubMed2006

Impact of implant length and diameter on survival rates.

INTRODUCTION: Despite the high success rates of endosseous oral implants, restrictions have been advocated to their placement with regard to the bone available in height and volume. The use of short or nonstandard-diameter implants could be one way to overcome this limitation. MATERIAL AND METHODS: In order to explore the relationship between implant survival rates and their length and diameter, a Medline and a hand search was conducted covering the period 1990-2005. Papers were included which reported: (1) relevant data on implant length and diameter, (2) implant survival rates; either clearly indicated or calculable from data in the paper, (3) clearly defined criteria for implant failure, and in which (4) implants were placed in healed sites and (5) studies were in human subjects. RESULTS: A total of 53 human studies fulfilled the inclusion criteria. Concerning implant length, a relatively high number of published studies (12) indicated an increased failure rate with short implants which was associated with operators' learning curves, a routine surgical preparation (independent of the bone density), the use of machined-surfaced implants, and the placement in sites with poor bone density. Recent publications (22) reporting an adapted surgical preparation and the use of textured-surfaced implants have indicated survival rates of short implants comparable with those obtained with longer ones. Considering implant diameter, a few publications on wide-diameter implants have reported an increased failure rate, which was mainly associated with the operators' learning curves, poor bone density, implant design and site preparation, and the use of a wide implant when primary stability had not been achieved with a standard-diameter implant. More recent publications with an adapted surgical preparation, new implant designs and adequate indications have demonstrated that implant survival rate and diameter have no relationship. DISCUSSION: When surgical preparation is related to bone density, textured-surfaced implants are employed, operators' surgical skills are developed, and indications for implant treatment duly considered, the survival rates for short and for wide-diameter implants has been found to be comparable with those obtained with longer implants and those of a standard diameter. The use of a short or wide implant may be considered in sites thought unfavourable for implant success, such as those associated with bone resorption or previous injury and trauma. While in these situations implant failure rates may be increased, outcomes should be compared with those associated with advanced surgical procedure such as bone grafting, sinus lifting, and the transposition of the alveolar nerve.

Bone Density↗
PubMed1995

The Helmholtz machine.

Discovering the structure inherent in a set of patterns is a fundamental aim of statistical inference or learning. One fruitful approach is to build a parameterized stochastic generative model, independent draws from which are likely to produce the patterns. For all but the simplest generative models, each pattern can be generated in exponentially many ways. It is thus intractable to adjust the parameters to maximize the probability of the observed patterns. We describe a way of finessing this combinatorial explosion by maximizing an easily computed lower bound on the probability of the observations. Our method can be viewed as a form of hierarchical self-supervised learning that may relate to the function of bottom-up and top-down cortical processing pathways.

Algorithms↗
PubMed1989

Of mulattos, mischlinge, and Minsky's machines.

Hegel once remarked that "what experience and history teach us is that people and governments never have learned anything from history or acted on principles deduced from it." Historically, efforts to define a human being sufficiently equipped biologically or politically to meet a set of factitious standards for inclusion in the community of mankind have invariably resulted in unspeakable injustices. The continuing exclusion of the human fetus from this community is another (and the latest) tragic example of the historical myopia of which Hegel spoke.

Abortion, Legal↗
PubMed1995

Sampling strategies in a statistical approach to clinical classification.

This paper studies the sampling strategies for the Expert Network (EexNet), a statistical learning system used for patient record classification at the Mayo Clinic. The goal is to achieve high accuracy classification at an affordable computational cost in very large applications. The learning curves of ExpNet were observed with respect to the choice of training resources, the size, vocabulary coverage and category coverage of a training set, and the category distribution over training instances. A method combining advantages of different sampling strategies is proposed and evaluated using a large training corpus. As a result, Expert Network has achieved its nearly-optimal classification accuracy (measured by average precision) using a relatively small training set, with a fast real-time response which satisfies the needs of human-machine interaction.

Expert Systems↗
PubMed2002

The brain-machine disanalogy revisited.

Michael Conrad was a pioneer in investigating biological information processing. He believed that there are fundamental lessons to be learned from the structure and behavior of biological brains that we are far from understanding or have implemented in our computers. Accumulation of advances in several fields have confirmed his views in broad outline but not necessarily in some of the strong forms he had tried to establish. For example, his assertion that programmable computers are intrinsically incapable of the brain's efficient and adaptive behavior has not received much examination. Yet, this is clearly a direction that could afford much insight into fundamental differences between brain and machine. In this paper, we pay tribute to Michael, by examining his pioneering thoughts on the brain-machine disanalogy in some depth and from the hindsight of a decade later. We argue that as long as we stay within the frame of reference of classical computation, it is not possible to confirm that programmability places a fundamental limitation on computing power, although the resources required to implement a programmable interface leave fewer resources for actual problem-solving work. However, if we abandon the classical computational frame and adopt one in which the user interacts with the system (artificial or natural) in real time, it becomes easier to examine the key attributes that Michael believed place biological brains on a higher plane of capability than artificial ones. While we then see some of these positive distinctions confirmed (e.g. the limitations of symbol manipulation systems in addressing real-world perception problems), we also see attributes in which the implementation in bioware constrains the behavior of real brains. We conclude by discussing how new insights are emerging, that look at the time-bound problem-solving constraints under which organisms have had to survive and how their so-called 'fast and frugal' faculties are tuned to the environments that coevolved with them. These directions open new paths for a multifaceted understanding of what biological brains do and what we can learn from them. We close by suggesting how the discrete event modeling and simulation paradigm offers a suitable medium for exploring these paths.

Behavior↗
PubMed1990

Initiating ethanol drinking in a simian social group in a naturalistic setting.

We examined in nine group-living, male Japanese Snow monkeys the initiation of alcohol-drinking behavior in an enriched environment where both social and nonsocial stimuli could influence drinking. The monkeys usually could move freely between an indoor shelter and a large outdoor corral, which contained three drinkometers. During daily 2-hr sessions in pre- and post-training periods (with food and water available ad libitum), the drinkometers held (on different days) Koolaid-saccharin, water, water with 5% ethanol, Koolaid-saccharin with 5% ethanol, or Koolaid-sucrose with acetic acid (matched in calories and palatability to Koolaid-saccharin with 5% ethanol). Between pre- and post-training periods was a long training period, in which Koolaid-saccharin with various ethanol concentrations was presented with the daily food ration in 2-hr drinking sessions. On other selected training days, water with 5% ethanol was presented. During training sessions some monkeys drank high doses; others did not. During the period of peak drinking, the daily mean ethanol consumption ranged among animals from 0.54-1.99 ml/kg. Blood ethanol concentrations then sometimes exceeded 100 mg/dl. During the post-training period, with return to ad libitum food and water, consumption declined from these peaks, but remained significantly higher than pretraining consumption; the ethanol solution was established as a reinforcer. Monkeys differed significantly in the extent of this pre/post increase. In both pre- and post-training periods, consumption was significantly greater when all three drinkometers operated, compared with only one. On pretraining days when only one drinkometer operated, more dominant animals drank significantly more than less dominant animals; this difference disappeared by the post-training period, as less dominant animals learned to use the drinkometers at times when the dominant animals eschewed the machines. Pretraining consumption did not predict (among animals) post-training consumption. Before, during, and after training, these animals drank less of both alcohol and control solutions than did members of another species, which we had studied in smaller indoor pens. We discuss possible explanations and implications.

Alcohol Drinking↗
PubMed2006

Distributed computing methodology for training neural networks in an image-guided diagnostic application.

Distributed computing is a process through which a set of computers connected by a network is used collectively to solve a single problem. In this paper, we propose a distributed computing methodology for training neural networks for the detection of lesions in colonoscopy. Our approach is based on partitioning the training set across multiple processors using a parallel virtual machine. In this way, interconnected computers of varied architectures can be used for the distributed evaluation of the error function and gradient values, and, thus, training neural networks utilizing various learning methods. The proposed methodology has large granularity and low synchronization, and has been implemented and tested. Our results indicate that the parallel virtual machine implementation of the training algorithms developed leads to considerable speedup, especially when large network architectures and training sets are used.

Algorithms↗
PubMed1993

A training simulator for detecting equipment failure in the anaesthetic machine.

Simulation is often used for training personnel in activities where the consequences of inappropriate actions are serious. We report a realistic training simulator, which can reproduce practically all potential malfunctions in the anaesthetic machine. Using actual standard equipment (Dameca 10750), the interior of the anaesthetic machine has been profoundly modified, whereas the external appearance remains virtually unchanged. The concealed alterations allow 20 different pre-set technical faults to be activated selectively from a mobile control unit. While assisted by an instructor, the trainee performs hands-on interactive experimentation with the simulator, while being exposed to 'unexpected' machine faults, which prompt for interpretation of error symptoms. Alternatively, the trainee can personally activate the simulated symptoms of different component failures, to enhance learning of the functional principles of the apparatus. The latter approach also allows a systematic presentation of defects to be identified by each step in a formal safety checklist for anaesthetic machines.

Anesthesiology↗
PubMed2005

Recognition of a familiar place by the honeybee (Apis mellifera).

Recent work shows that at any one place bees detect a limited variety of simple cues in parallel. At each choice point, they recognize a few cues in the range of positions where the cues occurred during the learning process. There is no need to postulate that they re-assemble the surrounding panorama in memory; only that they retain memories of the coincidences of cues in the expected retinotopic directions. The cues could be stimuli that excite groups of peripheral visual neurons. All the experimentally known cues are described, including modulation of the receptors, the locations of areas of black or colour, the nearness, size, averaged edge orientation, and radial and tangential edges. Cues of each type are separately summed within large fields, the size of which varies with the cue. Local orientation cues from edges at right angles cancel each other within each field, which also suggests that the discrimination of shape and texture is limited. Resolution depends on lateral interactions and the number of ommatidia required for each cue. To identify a new place, a few sparse cues, together with their directions, are learned in orientation flights. When the bee returns, the cues in the panorama are progressively matched as they coincide with the cues in memory. The limited number of cues, though economical for memory, may restrict the foraging behaviour and lead to flower constancy. This kind of a visual system is a candidate model for other animals or machines with economical processing systems.

Animals↗
PubMed1985

A user-friendly biological workstation.

Learning methods developed by artificial intelligence research teams are very efficient for biological sequences analysis but they need running on large computers accessed by terminals. These computers are interfaced with standard displays involving long and unpleasant alphanumerical data handling. The "biological work station" is a personal computer with a color graphic screen providing a user-friendly interface for the artificial intelligence learning programs running on large computers. It provides to biologist a graphical convenient tool for sequence analysis built with efficient man-machine communication methods such as multiwindows, icons and mouse selection. It allows the biologist to edit and display sequences in an efficient and natural way, showing off directly on color pictures the data and the results of learning programs.

Biology↗
PubMed2006

Robotic cholecystectomy: learning curve, advantages, and limitations.

BACKGROUND: Robotic cholecystectomy is safe, feasible procedure. Initial studies showed significant set up time and operating time but no clear clinical advantage of the robotic involvement. We have investigated the learning curve, advantages and limitation of the procedure. MATERIAL AND METHODS: We reviewed all (n = 51) robotic cholecystectomies performed between July 2004 and December 2005. The surgery was performed using the da Vinci system. We recorded operative time, setup time of robotics instrumentation, conversion to laparoscopic or open cholecystectomy and complication of the procedure. RESULTS: Forty-eight of the 51 procedures (94%) were completed robotically. We did not experience any significant complications directly related to robotics surgery. The mean +/- SD operating time was 77 +/- 22.3 min. The mean setup time for robotics (from incision until robot was in place, including draping the robot) was 24 +/- 8.8 min. However, the setup time significantly improved as we gained more experience: from 30.6 +/- 10.7 min (first 16 cases) to 18.3 +/- 4.0 min (cases 33-48). The mean robotic time was 34 +/- 16.1 min. We observed no significant improvement in robotic procedure time. CONCLUSIONS: Robotic cholecystectomy offers significant advantages such as three-dimensional view, easier instrument manipulations and possibility of remote site surgery. We observed some shortcomings of robotic surgery such as need for larger and additional ports, and need for undocking the machine in case of cholangiography or change of patient position. Our data shows that the learning curve is between 16 to 32 procedures to significantly decrease the setup time and total operating time.

Adolescent↗
PubMed2005

Vessel enhancement in digital X-ray angiographic sequences by temporal statistical learning.

In this paper, we present a vessel enhancement method, SVM temporal filtering (STF), for X-ray angiographic (XA) images using Support Vector Machine (SVM). We show that the linear SVM applied to vessel enhancement can be regarded as a matched linear filter optimizing the contrast-to-noise ratio in XA images. We propose a non-linear kernel function for the SVM leading to good enhancement with noisy, varying grey-level dynamics at vessel pixels. One key advantage over the matched filters is that an optimal filter is learnt from images, not estimated at design stage. Results on clinical XA images show that learning-based enhancement achieves better results compared to simple subtraction and other image stacking methods.

Algorithms↗
PubMed2004

A novel approach for short-term load forecasting using support vector machines.

A support vector machine (SVM) modeling approach for short-term load forecasting is proposed. The SVM learning scheme is applied to the power load data, forcing the network to learn the inherent internal temporal property of power load sequence. We also study the performance when other related input variables such as temperature and humidity are considered. The performance of our proposed SVM modeling approach has been tested and compared with feed-forward neural network and cosine radial basis function neural network approaches. Numerical results show that the SVM approach yields better generalization capability and lower prediction error compared to those neural network approaches.

Artificial Intelligence↗

Refine your search to explore more results.