Quick search Find article
Quick search
Find article

Comparison between radiological and artificial neural network diagnosis in clinical screening

A Degenhard1,3, C Tanner1,2, C Hayes1,4, D J Hawkes1,2, M O Leach1 and the UK MRI Breast Screening Study1,5

Show affiliations


The imaging protocol of the UK multicentre magnetic resonance imaging study for screening in women at genetic risk of breast cancer aims to assist in detecting and diagnosing malignant breast lesions. In this paper, we evaluate a three-layer, feed-forward, backpropagation neural network as an artificial radiological classifier using receiver operating characteristic (ROC) curve analysis and compare the results with those obtained using a proposed radiological scoring system for the study which currently supplements the radiologist's clinical opinion, in comparison with histological diagnosis. Based on the 76 symptomatic cases evaluated, descriptive features scored by radiologists showed considerable overlap between benign and malignant, although some features such as irregular contours and heterogeneous enhancement were more often associated with malignant pathology. In this preliminary evaluation, ROC analysis showed that the proposed scoring scheme did not perform well, indicating further refinement is required. When all 23 features were used in the neural network, its performance was poorer than that of the scoring scheme. When only ten features were used, limited to descriptors of enhancement characteristics, the neural network performed similar to the scoring scheme. This comparison shows that the neural network approach to clinical diagnosis has considerable potential and warrants further development.


PACS

87.61.Tg Clinical applications

02.70.Rr General statistical methods

87.19.X- Diseases

07.05.Mh Neural networks, fuzzy logic, artificial intelligence

87.57.C- Image quality

87.57.R- Computer-aided diagnosis

Subjects

Computational physics

Instrumentation and measurement

Biological physics

Medical physics

Dates

Issue 4 (November 2002)

Received 20 May 2002

Published 18 October 2002



  1. Comparison between radiological and artificial neural network diagnosis in clinical screening

    A Degenhard et al 2002 Physiol. Meas. 23 727

  2. Finite element modelling for materials with size effect

    S Swaddiwudhipong et al 2006 Modelling Simul. Mater. Sci. Eng. 14 1127

  3. The collagen structure of bovine intervertebral disc studied using polarization-sensitive optical coherence tomography

    Stephen J Matcher et al 2004 Phys. Med. Biol. 49 1295

  4. A covariant formalism of spin precession with respect to a reference congruence

    Rickard Jonsson 2006 Class. Quantum Grav. 23 37

  5. Gaseous diffusion in porous media. Part 3 - Wet granular materials

    J A Currie 1961 Br. J. Appl. Phys. 12 275

  6. A lactate electrochemical biosensor with a titanate nanotube as direct electron transfer promoter

    Mingli Yang et al 2008 Nanotechnology 19 075502

  7. Stochastic gravity: a primer with applications

    B L Hu and E Verdaguer 2003 Class. Quantum Grav. 20 R1

  8. Hopf bifurcations on cubic lattices

    T K Callahan 2003 Nonlinearity 16 2099

  9. Constraints on the Inner Cluster Mass Profile and the Power Spectrum Normalization from Strong Lensing Statistics

    Dragan Huterer and Chung-Pei Ma 2004 ApJ 600 L7

  10. String theory and noncommutative geometry

    Nathan Seiberg and Edward Witten JHEP09(1999)032

Related review articles

What's this?
View review articles related to this research to gain an insight into the key trends in this subject area. Related review articles are selected based on PACS/MSC codes, and are no more than three years old.

  1. Physics of magnetic resonance imaging: from spin to pixel
  2. Interaction of MRI field gradients with the human body

View by subject




Export








Please login to access our web services, or create an account if you don't yet have one.

You must have cookies enabled in your web browser to be able to login.

Username
Password

Forgotten your password? Get a new one here.