Download PDF
Nuklearmedizin 2008; 47(01): 48-55
DOI: 10.3413/nukmed-0098
Originalarbeiten
Schattauer GmbH

Computer-aided diagnosis in two-phase 201Tl-SPECT of thoracic lesions

Computer-gestützte Diagnostik mit der Zwei-Phasen-201Tl-SPECT von Thoraxläsionen
Y.-H. Yu
1   Departments of Internal Medicine
,
R.-F. Chang
2   Department of Computer Science and Information Engineering, National Chung Cheng University, Chiayi, Taiwan
,
W.-C. Shen
2   Department of Computer Science and Information Engineering, National Chung Cheng University, Chiayi, Taiwan
,
D.-K. Lin
2   Department of Computer Science and Information Engineering, National Chung Cheng University, Chiayi, Taiwan
,
S.-S. Sun
3   Departments of Nuclear Medicine, Division of Pulmonary and Critical Care Medicine, China Medical University Hospital, Taichung
,
C.-Y. Tu
1   Departments of Internal Medicine
,
K.-L. Chiu
1   Departments of Internal Medicine
,
W.-H. Hsu
1   Departments of Internal Medicine
› Author Affiliations
Further Information

Publication History

Received: 23 February 2007

accepted in revised form: 02 June 2007

Publication Date:
02 January 2018 (online)

    Permissions and Reprints

Summary

Aim: The retention index, a traditionally quantitative analysis of two-phase 201Tl single photon emission computed tomography (SPECT) of the chest, is manually calculated by experienced physicians from comparable 2-D ROI. However, a 3-D ROI would provide more information than a 2-D ROI extracted from a single frame of SPECT. We propose a new diagnostic system, computer-aided diagnosis (CAD), to automatically detect suspicious lesions as 3-D objects on chest 201Tl-SPECT, and assist the physician in interpreting these images. Patients, methods: Seventy patients with thoracic lesions and confirmed diagnoses were enrolled to test this automatic CAD system. The reliability of the CAD system in detecting lesions as 3-D objects was compared to the 2-D ROI of 201Tl-SPECT found by the manually visualized method. Furthermore, we also proposed a novel index, the retention index using the heart (RIH), to differentiate high retention (slow clearance, increasing target to heart ratio) as a criterion for a malignant lesion, from low retention (faster clearance, small or no increase of the target to heart ratio) for benign lesions. Results: The CAD system can achieve a detection rate of 100% in automatically searching for thoracic lesions in 201Tl-SPECT. In diagnostic performance, the CAD system with the RIH of comparable 3-D objects has an area under the ROC curve of 0.86, higher than the 0.78 of the traditional RI method (p = 0.198). Conclusion: The CAD system of two-phase 201Tl-SPECT is a promising tool for detecting and diagnosing thoracic lesions with a diagnostic accuracy of 0.81.

Zusammenfassung

Ziel: Der Retentionsindex, eine traditionell quantitative Analyse der Zwei-Phasen 201Tl-SPECT (Single-Photonen-Emissions-Computertomographie des Thorax, wird von erfahrenen ärzten manuell anhand von vergleichbaren 2D-ROIs berechnet. Eine 3D-ROI würde jedoch mehr Informationen liefern als eine 2D-ROI, die einer SPECT-Einzelaufnahme entnommen wird. Wir schlagen ein neues diagnostisches System vor, die Computergestützte Diagnostik (CAD), um verdächtige Läsionen automatischals3D-ObjekteimThorax-201Tl-SPECTzuerkennenund um dem Arzt bei der Auswertung dieser Bilder zu helfen. Patienten, Methoden: 70 Patienten mit Thoraxläsionen und bestätigtenDiagnosenwurdenindieStudiezurPrüfungdiesesautomatischen CAD-Systems aufgenommen. Die Zuverlässigkeit des CAD-Systems bei der Erkennung von Läsionen als 3D-Objekte wurde mit der 2D-ROI der 201Tl-SPECT verglichen, die mit der manuell visualisierten Methode erhalten wurden. Außerdem schlugen wir einen neuen Index vor, den Retentionsindex unterVerwendungdesHerzens(RIH),umzwischen einerhohen Retention (langsame Clearance, steigender Zielorgan-Herz-Quotient) als Kriterium für eine maligne Läsion von einer niedrigen Retention (schnellere Clearance, geringer oder kein Anstieg des Zielorgan-Herz-Quotienten) bei benignen Läsionen unterscheiden zu können. Ergebnisse: Mit dem CAD-System lässt sich eine Detektionsrate von 100% bei der automatischen Suche nach Thoraxläsionen im 201Tl-SPECT erzielen. Bezüglich der diagnostischen Leistungsfähigkeit besitzt das CAD-System mit dem RIH vergleichbarer 3D-Objekte eine Fläche unter der ROC-Kurve von 0,86, die größer ist als die mit der traditionellen Risiko-Methode erzielte Fläche von 0,78 (p = 0,198). Schlussfolgerung: Das CAD-System des Zwei-Phasen-201Tl-SPECT ist ein vielversprechendes Hilfsmittel für die Erkennung und Diagnostik von Thoraxläsionen und besitzt eine diagnostische Genauigkeit von 0,81.

Keywords

201Tl - SPECT - CAD - lung cancer

Schlüsselwörter

201Tl - SPECT - CAD - Lungenkrebs
 

  • References

  • 1 Berk F, Demir H, Aktolun C. Thallium-201 imaging in the assessment of tumor response to antitumor treatments. QJNucl Med 2003; 47: 63-74.
    Google Scholar
  • 2 Black MJ, Sapiro G, Marimont DH. et al. Robust anisotropic diffusion. Image Processing, IEEE Transactions 1998; 7: 421-432.
    Google Scholar
  • 3 Cox PH, Belfer AJ, van der Pompe WB. Thallium 201 chloride uptake in tumours, a possible complication in heart scintigraphy. Br J Radiol 1976; 49: 767-768.
    CrossrefPubMedGoogle Scholar
  • 4 Fletcher BD, Kauffaan WM, Kaste SC. et al. Use of Tl-201 to detect untreated pediatric Hodgkin disease. Radiology 1995; 196: 851-855.
    CrossrefPubMedGoogle Scholar
  • 5 Floyd Jr CE, Jaszczak RJ, Manglos SH. et al. Compensation for collimator divergence in SPECT using inverse Monte Carlo reconstruction. Nuclear Science, IEEE Transactions 1988; 35: 784-787.
    Google Scholar
  • 6 Gauch JM. Image segmentation and analysis via multiscale gradient watershed hierarchies. Image Processing, IEEE Transactions 1999; 8: 69-79.
    Google Scholar
  • 7 Higashi K, Nishikawa T, Seki H. et al. Comparison of fluorine-18-FDG PET and thallium-201 SPECT in evaluation of lung cancer. J Nucl Med 1998; 39: 9-15.
    PubMedGoogle Scholar
  • 8 Higashi K, Ueda Y, Sakuma T. et al. Comparison of [18F]FDG PET and 201Tl SPECT in evaluation of pulmonary nodules. J Nucl Med 2001; 42: 1489-1496.
    PubMedGoogle Scholar
  • 9 Hisada K, Tonami N, Miyamae T. et al. Clinical evaluation of tumor imaging with 201TI chloride. Radiology 1978; 129: 497-500.
    CrossrefPubMedGoogle Scholar
  • 10 Liang Z. Implementation of linear filters for iterative penalized maximum likelihood SPECT reconstruction. Nuclear Science, IEEE Transactions 1991; 38: 606-611.
    Google Scholar
  • 11 Manglos SH, Jaszczak RJ, Floyd CE. Weighted backprojection implemented with a non-uniform attenuation map for improved SPECT quantitation. NuclearScience, IEEE Transactions 1988; 35: 625-628.
    Google Scholar
  • 12 Quek FKH, Kirbas C. Vessel extraction inmedical images by wave-propagation and traceback. Medical Imaging, IEEE Transactions 2001; 20: 117-131.
    Google Scholar
  • 13 Salvatore M, Carratu L, Porta E. Thallium-201 as a positive indicator for lung neoplasms: preliminary experiments. Radiology 1976; 121: 487-488.
    CrossrefPubMedGoogle Scholar
  • 14 Schwartz AG, Ruckdeschel JC. Familial lung cancer: genetic susceptibility and relationship to chronic obstructive pulmonary disease. Am J RespirCrit Care Med 2006; 173: 16-22.
    CrossrefPubMedGoogle Scholar
  • 15 Singh M, Horne M, Maneval D. et al. Non-uniform attenuation and scatter correction in SPECT. Nuclear Science, IEEE Transactions 1988; 35: 767-771.
    Google Scholar
  • 16 Sumiya H, Taki J, Tsuchiya H. et al. Midcourse thallium-201 scintigraphy to predict tumor response in bone and soft-tissue tumors. J Nucl Med 1998; 39: 1600-1604.
    PubMedGoogle Scholar
  • 17 Tonami N, Shuke N, Yokoyama K. et al. Thallium-201 single photon emission computed tomography in the evaluation of suspected lung cancer. J Nucl Med 1989; 30: 997-1004.
    PubMedGoogle Scholar
  • 18 Tonami N, Yokoyama K, Shuke N. et al. Evaluation of suspected malignant pulmonary lesions with 201Tl single photon emission computed tomography. Nucl Med Commun 1993; 14: 602-610.
    CrossrefPubMedGoogle Scholar
  • 19 Tonami N, Yokoyama K, Taki J. et al. Thallium-201 SPECT depicts radiologically occult lung cancer. J Nucl Med 1991; 32: 2284-2285.
    PubMedGoogle Scholar
  • 20 Vincent L, Soille P. Watersheds in digital spaces: an efficient algorithm based on immersion simulations. Pattern Analysis and Machine Intelligence, IEEE Transactions 1991; 13: 583-598.
    CrossrefGoogle Scholar
  • 21 Vos MJ, Hoekstra OS, Barkhof F. et al. Thallium-201 single-photon emission computed tomography as an early predictor of outcome in recurrent glioma. J Clin Oncol 2003; 21: 3559-3565.
    CrossrefPubMedGoogle Scholar
  • 22 Yamamoto Y, Nishiyama Y, Satoh K. et al. Comparative study of technetium-99m-sestamibi and thallium-201 SPECT in predicting chemothera-peutic response in small cell lung cancer. J Nucl Med 1998; 39: 1626-1629.
    PubMedGoogle Scholar
  • 23 You Y-L, Xu W, Tannenbaum A. et al. Behavioral analysis of anisotropic diffusion in image processing. Image Processing, IEEE Transactions 1996; 5: 1539-1553.
    Google Scholar
  • 24 Yu Y, Acton ST. Speckle reducing anisotropic diffusion. Image Processing, IEEE Transactions. 2002 11. 1260-1270.
    PubMedGoogle Scholar
  • 25 Yu YH, Hsu WH, Hsu NY. et al. Comparison of two-phase 201Tl SPECT with chest CT to differentiate thoracic malignancies from benign lesions. Q J Nucl Med Mol Imaging. 2007 Epub ahead of print.
    PubMedGoogle Scholar
  • 26 Yu YH, Hsu WH, Hsu NY. et al. Comparison of two-phase Tl-201 SPECT with chest CT to differentiate thoracic malignancies from benign lesions. Q J Nucl Med. 2006 in press.
    PubMedGoogle Scholar
  • 27 Yu YH, Hsu WH, Hsu NY. et al. The use of dual phase 201Tl SPECT for differentiating pulmonary malignancies from benign lesions. Jpn J Clin Oncol 2004; 34: 445-451.
    CrossrefPubMedGoogle Scholar