Breast Cancer Diagnosis and Prognosis Via Linear Programming

By Olvi L. Mangasarian, W. Street, William H. Wolberg. 1995

Subfields: Medical Imaging, Medical & Health AI

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Medical Imaging

In Medical & Health AI

Citations in the corpus

A practical tutorial on autoencoders for nonlinear feature fusion: Taxonomy, models, software and guidelines (2018, background): “The original data [138] has 30 variables describing each pattern.”
Learning Optimized Risk Scores (2016, methodology): “We ran the computational experiments in Sections 3.3 and 4 using a collection of synthetic datasets that we generated from the breastcancer dataset of Mangasarian et al. (1995), which contains n = 683 samples and d = 9 features xij ∈ {0, . . . , 10}.”
Supersparse linear integer models for optimized medical scoring systems (2015, neutral): “roducibility, we include all processed datasets in Online Resource 1. Dataset Source N P Classication Task adult Kohavi (1996) 32561 36 predict if a U.S. resident earns more than $50000 breastcancer Mangasarian et al. (1995) 683 9 detect breast cancer using a biopsy bankruptcy Kim and Han (2003) 250”
Node harvest: simple and interpretable regression and classication (2009, methodology): “As an example of binary classiﬁcation, take the Wisconsin breast cancer data [Mangasarian, Street and Wolberg (1995)].”
Genetic programming for simultaneous feature selection and classifier design (2006, methodology): “We do not include the feature weights for WDBC in Table X because the number of features is too large to accommodate in Table X.”
Machine Learning via Polyhedral Concave Minimization (1996, methodology): “11)], which is very e ective in discriminating between sets arising from real world problems [21], and motivate our formulation by the perturbation results of linear programming [20] to suppress as many of the coe cients w of the separating plane fx j xTw = g as possible, we obtain the following pro”
A fuzzy-genetic approach to breast cancer diagnosis (1999, background): “2 Fine needle aspiration is an outpatient procedure that involves using a small-gauge needle to extract fluid directly from a breast mass [20].”
Model-Based Clustering, Discriminant Analysis, and Density Estimation (2002, methodology): “Figure 2: Pairs plots of the Wisconsin Diagnostic Breast Cancer Data from the DCI Machine Learpjng Repository, showing only the 3 explanatory variables used by Mangasarian et al. (1995). There are 569 observations.”
Rule extraction from linear support vector machines (2005, background): “Doctors identified nine visually assessed characteristics or attributes of an FNA sample which they considered relevant to diagnosis (for more detail please refer to [12]).”
SSVM: A Smooth Support Vector Machine for Classification (2001, methodology): “We ran all tests on six publicly available datasets: the Wisconsin Prognostic Breast Cancer Database [24] and four datasets from the Irvine Machine Learning Database Repository [26].”
On Ensuring that Intelligent Machines Are Well-Behaved (2017, methodology): “algorithms is important because, for example, they have been used for medical applications where undesirable behavior could delay cancer diagnoses [Zacharaki et al., 2009; Mangasarian et al., 1995], and because they have been shown to sometimes cause racist, sexist, and other discriminatory behavior”
Preventing undesirable behavior of intelligent machines (2019, background): “5, 11] and [10, 15] respectively, as shown by the blue box in Fig.”