Author
Rafael Molina
Other affiliations: Intel, Northwestern University
Bio: Rafael Molina is an academic researcher from University of Granada. The author has contributed to research in topics: Image restoration & Iterative reconstruction. The author has an hindex of 52, co-authored 381 publications receiving 10765 citations. Previous affiliations of Rafael Molina include Intel & Northwestern University.
Papers published on a yearly basis
Papers
More filters
••
TL;DR: This paper model the components of the compressive sensing (CS) problem, i.e., the signal acquisition process, the unknown signal coefficients and the model parameters for the signal and noise using the Bayesian framework and develops a constructive (greedy) algorithm designed for fast reconstruction useful in practical settings.
Abstract: In this paper, we model the components of the compressive sensing (CS) problem, i.e., the signal acquisition process, the unknown signal coefficients and the model parameters for the signal and noise using the Bayesian framework. We utilize a hierarchical form of the Laplace prior to model the sparsity of the unknown signal. We describe the relationship among a number of sparsity priors proposed in the literature, and show the advantages of the proposed model including its high degree of sparsity. Moreover, we show that some of the existing models are special cases of the proposed model. Using our model, we develop a constructive (greedy) algorithm designed for fast reconstruction useful in practical settings. Unlike most existing CS reconstruction methods, the proposed algorithm is fully automated, i.e., the unknown signal coefficients and all necessary parameters are estimated solely from the observation, and, therefore, no user-intervention is needed. Additionally, the proposed algorithm provides estimates of the uncertainty of the reconstructions. We provide experimental results with synthetic 1-D signals and images, and compare with the state-of-the-art CS reconstruction algorithms demonstrating the superior performance of the proposed approach.
718 citations
••
TL;DR: The popular neural network architectures used for imaging tasks are reviewed, offering some insight as to how these deep-learning tools can solve the inverse problem.
Abstract: Traditionally, analytical methods have been used to solve imaging problems such as image restoration, inpainting, and superresolution (SR). In recent years, the fields of machine and deep learning have gained a lot of momentum in solving such imaging problems, often surpassing the performance provided by analytical approaches. Unlike analytical methods for which the problem is explicitly defined and domain-knowledge carefully engineered into the solution, deep neural networks (DNNs) do not benefit from such prior knowledge and instead make use of large data sets to learn the unknown solution to the inverse problem. In this article, we review deep-learning techniques for solving such inverse problems in imaging. More specifically, we review the popular neural network architectures used for imaging tasks, offering some insight as to how these deep-learning tools can solve the inverse problem. Furthermore, we address some fundamental questions, such as how deeplearning and analytical methods can be combined to provide better solutions to the inverse problem in addition to providing a discussion on the current limitations and future directions of the use of deep learning for solving inverse problem in imaging.
496 citations
••
TL;DR: In oestrogen receptor-positive, HER2-negative, lymph node-negative patients, multianalyte tests such as urokinase plasminogen activator (uPA)-PAI-1, Oncotype DX, MammaPrint, EndoPredict, Breast Cancer Index (BCI) and Prosigna (PAM50) may be used to predict outcome and aid adjunct therapy decision-making.
356 citations
••
TL;DR: Recommendations are presented for the routine clinical use of serum and tissue-based markers in the diagnosis and management of patients with breast cancer and recently validated prognostic markers for lymph node-negative breast cancer patients may be of value in selecting node- negative patients that do not require adjuvant chemotherapy.
Abstract: Recommendations are presented for the routine clinical use of serum and tissue-based markers in the diagnosis and management of patients with breast cancer. Their low sensitivity and specificity preclude the use of serum markers such as the MUC-1 mucin glycoproteins (CA 15.3, BR 27.29) and carcinoembryonic antigen in the diagnosis of early breast cancer. However, serial measurement of these markers can result in the early detection of recurrent disease as well as indicate the efficacy of therapy. Of the tissue-based markers, measurement of estrogen and progesterone receptors is mandatory in the selection of patients for treatment with hormone therapy, while HER-2 is essential in selecting patients with advanced breast cancer for treatment with Herceptin (trastuzumab). Urokinase plasminogen activator and plasminogen activator inhibitor 1 are recently validated prognostic markers for lymph node-negative breast cancer patients and thus may be of value in selecting node-negative patients that do not require adjuvant chemotherapy.
322 citations
••
TL;DR: All tumor markers showed a clear relationship with tumor stage and histology and therefore enabled a better histological diagnosis and helped in the diagnosis of non-small cell lung cancer.
Abstract: CEA, CA 125, SCC, CYFRA 21-1 and NSE were prospectively studied in 211 patients with non-small cell lung cancer and compared with clinical parameters (age, sex, Karnofsky Index, symptoms and smoking status), histopathological parameters (stage, histology, tumor size and nodal involvement), biological parameters (LDH and albumin) and the therapy used (surgery, chemotherapy or radiotherapy). Tumor marker sensitivity was CYFRA 21-1: 76%, CA 125: 55%, CEA: 52%, SCC: 33% and NSE: 22%. One of the tumor markers was abnormally high in 87% of the patients with locoregional disease and in 100% of the patients with metastases. Except for NSE, all tumor markers showed a clear relationship with tumor stage and histology and therefore enabled a better histological diagnosis. Abnormal CEA serum levels were mainly found in adenocarcinomas, CA 125 in large-cell lung cancers (LCLC) and adenocarcinomas and SCC in squamous tumors. Eighty-five percent of the patients with SCC levels >2 ng/ml had squamous tumors. Likewise, CA 125 levels <60 U/ml or CEA <10 ng/ml excluded adenocarcinoma or LCLC with a probability of 82 and 91%, respectively.
265 citations
Cited by
More filters
•
28,685 citations
•
02 Aug 1996TL;DR: In this paper, a density-based notion of clusters is proposed to discover clusters of arbitrary shape, which can be used for class identification in large spatial databases and is shown to be more efficient than the well-known algorithm CLAR-ANS.
Abstract: Clustering algorithms are attractive for the task of class identification in spatial databases. However, the application to large spatial databases rises the following requirements for clustering algorithms: minimal requirements of domain knowledge to determine the input parameters, discovery of clusters with arbitrary shape and good efficiency on large databases. The well-known clustering algorithms offer no solution to the combination of these requirements. In this paper, we present the new clustering algorithm DBSCAN relying on a density-based notion of clusters which is designed to discover clusters of arbitrary shape. DBSCAN requires only one input parameter and supports the user in determining an appropriate value for it. We performed an experimental evaluation of the effectiveness and efficiency of DBSCAN using synthetic data and real data of the SEQUOIA 2000 benchmark. The results of our experiments demonstrate that (1) DBSCAN is significantly more effective in discovering clusters of arbitrary shape than the well-known algorithm CLAR-ANS, and that (2) DBSCAN outperforms CLARANS by a factor of more than 100 in terms of efficiency.
17,056 citations
•
01 Jan 1996TL;DR: DBSCAN, a new clustering algorithm relying on a density-based notion of clusters which is designed to discover clusters of arbitrary shape, is presented which requires only one input parameter and supports the user in determining an appropriate value for it.
Abstract: Clustering algorithms are attractive for the task of class identification in spatial databases. However, the application to large spatial databases rises the following requirements for clustering algorithms: minimal requirements of domain knowledge to determine the input parameters, discovery of clusters with arbitrary shape and good efficiency on large databases. The well-known clustering algorithms offer no solution to the combination of these requirements. In this paper, we present the new clustering algorithm DBSCAN relying on a density-based notion of clusters which is designed to discover clusters of arbitrary shape. DBSCAN requires only one input parameter and supports the user in determining an appropriate value for it. We performed an experimental evaluation of the effectiveness and efficiency of DBSCAN using synthetic data and real data of the SEQUOIA 2000 benchmark. The results of our experiments demonstrate that (1) DBSCAN is significantly more effective in discovering clusters of arbitrary shape than the well-known algorithm CLARANS, and that (2) DBSCAN outperforms CLARANS by a factor of more than 100 in terms of efficiency.
14,297 citations
01 Jan 2006
TL;DR: Probability distributions of linear models for regression and classification are given in this article, along with a discussion of combining models and combining models in the context of machine learning and classification.
Abstract: 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.
10,141 citations