PDF herunterladen
Methods Inf Med 2004; 43(03): 239-246
DOI: 10.1055/s-0038-1633864
Original Article
Schattauer GmbH

Elicitation and Representation of Expert Knowledge for Computer Aided Diagnosis in Mammography

E. Alberdi
1   CHIME, University College London, UK
,
P. Taylor
1   CHIME, University College London, UK
,
R. Lee
1   CHIME, University College London, UK
› Institutsangaben
Weitere Informationen

Publikationsverlauf

Publikationsdatum:
05. Februar 2018 (online)

    Lizenzen und Reprints

Summary

Objectives: To study how professional radiologists describe, interpret and make decisions about micro-calcifications in mammograms. The purpose was to develop a model of the radiologists’ decision making for use in CADMIUM II, a computerized aid for mammogram interpretation that combines symbolic reasoning with image processing.

Methods: In Study 1, eleven radiologists were asked to ‘think out loud’ as they interpreted 20 sets of calcifications. Participants used 159 terms to describe calcifications. We used these data to design a scheme with 50 descriptors. In Study 2, ten radiologists used the scheme to describe 40 sets of calcifications. We assessed the capacity of the terms to discriminate between benign and malignant calcifications, testing them against radiologists’ assessments of malignancy and follow-up data.

Results: The descriptors that were found to be the most discriminating in Study 2 were included in CADIMUM II’s knowledge base. They were represented as arguments for either a benign or a malignant diagnosis. These arguments are the central component of the decision support provided by the system. Other components are: image processing algorithms for the detection and measurement of calcifications and a set of rules that use the measures to decide which of the arguments apply to a given set of calcifications.

Conclusions: Preliminary evaluations of the CADMIUM II prototype reinforce the value of representing explicitly decision making processes in computer aided mammography and of deriving these processes from image processing measurements. Decision support is presented here at a level of description that is both relevant and meaningful to the user.

Keywords

Computer aided diagnosis - radiology - knowledge elicitation - knowledge representation - image processing
 

  • References

  • 1 Taylor P. Computer aids for decision making in diagnostic radiology – a literature review. Brit J Radiol 1995; 68: 945-57.
    CrossrefPubMedGoogle Scholar
  • 2 Roehrig J, Castellino RA. The promise of computer aided detection in digital mammography. Eur J Radiol 1999; 31 (01) 35-9.
    CrossrefPubMedGoogle Scholar
  • 3 Freer TW, Ulissey MJ. Screening mammography with computer-aided detection: prospective study of 12,860 patients in a community breast center. Radiology 2001; 220: 781-6.
    CrossrefPubMedGoogle Scholar
  • 4 Schmidt F, Sorantin E, Szepesvàri C, Graif E, Becker M, Mayer H, Hartwagner K. An automatic method for the identification and interpretation of clustered micro-calcifications in mammograms. Phys Med Biol 1999; 44 (05) 1231-43.
    CrossrefPubMedGoogle Scholar
  • 5 Hara T, Yamada A, Fujita H. et al. Automated classification method of mammographic calcifications by using artificial neural network and ACR BIRADS criteria for microcalcification distribution. In Yaffe M. (ed.) Proceedings of the 5th International Workshop on Digital Mammography. Medical Physics Publishing. 2000: 198-204.
    Google Scholar
  • 6 Kaufmann GH, Salifty MF, Granitto P, Cecccatto HA. Automated detection and classification of clustered calcifications using morphological filtering and statistical techniques. In Yaffe M. (ed.) Proceedings of the 5th International Workshop on Digital Mammography; Medical Physics Publishing. 2000: 253-8.
    Google Scholar
  • 7 Markopoulos C, Kouskos E, Koufopoulos K. et al. Use of artificial neural networks (computer analysis) in the diagnosis of microcalcifications on mammography. Eur J Radiol 2001; 39: 60-5.
    CrossrefPubMedGoogle Scholar
  • 8 Hartswood M, Procter R, Willimas J. Prompting in practice: How can we ensure radiologists make best use of computer-aided detection systems in screening mammography?. In Karssemeijer E, Thijsen M, Hendriks J. (eds.) Digital Mammography. Kluwer Publishing, Dordrecht. 1998: 363-70.
    Google Scholar
  • 9 Taylor P, Alberdi E, Lee R. et al. Incorporating Image Processing in a Clinical Decision Support System. In Insana MF, Leahy RM. (eds.) Information Processing in Medical Imaging. Springer Verlag; 2001: 134-40.
    Google Scholar
  • 10 Taylor P, Fox J, Todd-Pokropek A. The development and evaluation of CADMIUM: a prototype system to assist in the interpretation of mammograms. Medical Image Analysis 1999; 3: 321-37.
    CrossrefPubMedGoogle Scholar
  • 11 Cook HM, Fox MC. Application of expert systems to mammographic image analysis. Am J Physiol Imag 1989; 4: 16-22.
    PubMedGoogle Scholar
  • 12 Stewart BD, Cairns AY, Ricketts IW. The analysis of breast abnormalities by computer. In Gale A, Astley S, Dance D, Cairns A. (eds.) Digital Mammography. Elsevier Science; Amsterdam: 1994: 251-61.
    Google Scholar
  • 13 Floyd CE, Lo JY, Tourassi GD. Case-based reasoning algorithm that uses mammographic findings for breast biopsy decisions. AJR 2000; 175: 1347-52.
    CrossrefPubMedGoogle Scholar
  • 14 Fox J, Myers CD, Greaves MF, Pegram S. Knowledge acquisition for expert systems: Experience in leukaemia diagnosis. Methods Inf Med 1985; 24: 65-72.
    Artikel in Thieme ConnectPubMedGoogle Scholar
  • 15 Cooke NJ. Varieties of knowledge elicitation techniques. International Journal of Human Computer Studies 1994; 41: 801-49.
    CrossrefPubMedGoogle Scholar
  • 16 Khan AS, Hoffmann A. Building a case based diet recommendation system without a knowledge engineer. Artificial Intelligence in Medicine 2003; 27 (02) 155-79.
    CrossrefPubMedGoogle Scholar
  • 17 Guimaraes G, Peter JH, Penzel T, Ultsch A. A method for automated temporal knowledge acquisition applied to sleep-related breathing disorders. Artificial Intelligence in Medicine 2001; 23 (03) 211-37.
    CrossrefPubMedGoogle Scholar
  • 18 Ceusters W, Rogers J, Consorti F, Rossi-Mori A. Syntactic-semantic tagging as a mediator between linguistic representations and formal models: an exercise in linking SNOMED to GALEN. Intelligence in Medicine 1999; 15 (01) 5-23.
    PubMedGoogle Scholar
  • 19 Chute CG. Clinical classification and terminology: Some history and current observations. JAMIA 2000; 7 (03) 298-303.
    PubMedGoogle Scholar
  • 20 American College of Radiology. Breast imaging reporting and data system (BI-RADS). 2nd ed. Reston, VA: ACR; 1995
    Google Scholar
  • 21 Baker JA, Kornguth PJ, Floyd CE. Breast imaging reporting and data system standardized mammography lexicon: observer variability in lesion description. AJR 1996; 166: 773-8.
    CrossrefPubMedGoogle Scholar
  • 22 Liberman L, Abramson AF, Lassman JR, Morris EA, Dershaw DD. The breast imaging reporting and data system: Positive predictive value of mammographic features and final assessment categories. AJR 1997; 171: 35-40.
    PubMedGoogle Scholar
  • 23 Kerlikowske K, Grady D, Barclay J, Frankel SD, Ominsky SH, Sickles EA, Ernster V. Variability and accuracy in mammographic interpretation using the American College of Radiology Breast Imaging Reporting Data System. Journal of the National Cancer Institute 1998; 90 (23) 1801-9.
    CrossrefPubMedGoogle Scholar
  • 24 Starren J, Johnson SM. Expressiveness of the Breast Imaging Reporting and Database System (BI-RADS). JAMIA 1997; Suppl. S 655-9.
    PubMedGoogle Scholar
  • 25 Simpson W, Neilson F, Kelly PJ. Descriptive Terms for mammographic abnormalities: Observer variation in application. Clin Radiol 1996; 51: 709-71.
    CrossrefPubMedGoogle Scholar
  • 26 Whatmough P, Gale A, Wilson ARM. Do radiologists agree on the importance of mammographic features. In Doi K, Giger ML, Nishikawa RM, Schmidt RA. (eds.) Digital Mammography’96. Elsevier Science; Amsterdam: 1996: 111-6.
    Google Scholar
  • 27 D’Orsi CJ, Getty DJ, Swets JA. et al. Reading and decision aids for improved accuracy and standardization of mammographic diagnosis. Radiology 1992; 184: 619-22.
    CrossrefPubMedGoogle Scholar
  • 28 Alberdi E, Taylor P, Lee R. et al. CADMIUM II: Acquisition and representation of radiological knowledge for computerized decision support in mammography. In Overhage JM. (ed.) Proceedings of the American Medical Informatics Association Symposium. 2000: 7-11.
    Google Scholar
  • 29 Alberdi E, Taylor P, Lee R. et al. Eliciting a terminology for mammographic calcifications. Clinical Radiology 2002; 57: 1007-13.
    CrossrefPubMedGoogle Scholar
  • 30 Skinner MA, Swain M, Simmons R, McCarty KS, Sullivan DC, Iglehart JD. Nonpalpable breast lesions at biopsy. Ann Surg 1988; 208 (02) 203-8.
    CrossrefPubMedGoogle Scholar
  • 31 Franceschi D, Crowe J, Zollinger R, Duchesneau R, Shenk R, Stefanek G. Biopsy of the breast for mamographically detected lesions. Surgery, Gynecology and Obstetrics 1990; 171 (06) 449-55.
    PubMedGoogle Scholar
  • 32 Monostori Z, Herman PG, Carmody DP, Eacobacci TM, Capece NR, Cruz VM, Fentin S, Vernace FM. Limitations in distinguishing malignant from benign lesions of the breast by systematic review of mamograms. Surgery, Gynecology and Obstetrics 1991; 173: 438-42.
    PubMedGoogle Scholar
  • 33 Harkins K, Tartter PI, Hermann G, Squitieri R, Bower ST, Keller RJ. Multivariate analysis of roentgenologic characteristics and risks factors for nonpalpable carcinoma of the breast. Journal of the American College of Surgeons 1994; 178: 149-54.
    PubMedGoogle Scholar
  • 34 Lee R, Alberdi E, Taylor P. et al. A comparative study of four techniques for calcification detection. In Yaffe M. (ed.) Proceedings of the 5th International Workshop on Digital Mammography. Medical Physics Publishing; 2000: 264-71.
    Google Scholar