| Peer-Reviewed

Neural Method Based on Kohonen Topological Maps Applied to the Whole-body Scintigraphy Image

Received: 11 October 2021     Accepted: 2 November 2021     Published: 12 November 2021
Views:       Downloads:
Abstract

Objective: The aim of this study was to describe the stages of learning Kohonen's self-organizing maps applied to scintigraphy imaging in order to perform classification for medical diagnostic aid. Method: To achieve these goals, the neurons, arranged on a regular grid, are connected to each other by a neighbor relationship, which creates the topology of the map. The input layer consisted of pixels from the scintigraphy images. Results: During the iteration rounds of learning, we have seen a deployment of neurons on the nodes of the map that becomes more and more important. And it is the same for the winning neurons. After 750 iterations, the Davies Bouldin index attests to the end of the training with a quantization error that goes from 0.175 at the beginning of the training to 0.0225 at the end of the training. After this study, we find that neurons 41, 62, 121, 101 and 145 have captured most of the data with a peak uptake achieved by neuron 41 which has captured 1048 data. This individualizes the class of high intensities undoubtedly corresponding to metastatic hyperfixations. Conclusion: This innovative method could undoubtedly be integrated into the link in the chain highlighting periarticular metastases in developing countries, most of which do not have a SPECT-CT.

Published in International Journal of Medical Imaging (Volume 9, Issue 4)
DOI 10.11648/j.ijmi.20210904.14
Page(s) 189-192
Creative Commons

This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.

Copyright

Copyright © The Author(s), 2021. Published by Science Publishing Group

Keywords

Artificial Neural Networks, Kohonen Self Organizing Maps, Bone Scintigraphy

References
[1] NDONG B. Place de la scintigraphie osseuse dans le bilan d’extension des cancers de la prostate au Sénégal: étude préliminaire à propos de 45 cas. Médecine Nucléaire Imagerie fonctionnelle et Métabolique 2012 vol. 36 pp. 586-590.
[2] M. L. Mboup, K. Tall, B. Ndong, I. Diop, M. B. Gueye, F. Kaly, M. S. Camara, P. A. Mboup, M. Mbodj, S. M. Farssi “Système d’aide au Diagnostic des Métastases du Cancer de la Prostate par le traitement de l’Image de Scintigraphie Osseuse”, 1ere Edition de la Journée Scientifique des Doctorants, COJEC, Ecole Supérieure Polytechnique (ESP), 11 Juin 2016, Dakar, Sénégal.
[3] Dryfus G., Martinez J. M., Samuelides M. B., Gordon M. B., Badran F., Thiria S. Apprentissage statistique des réseaux de neurones, cartes topologiques, machine à vecteur supports, 2008.
[4] Kohonen T. Self-organized formation of topologically correct feature Maps. Biological cybernetics springer-verla. 1982 vol. 43 pp. 59-69.
[5] Guèye M. B. Inversion neuronale pour la reconstruction de profils de salinité océanique en atlantique tropical à partir de mesures de surface et de hauteurs d’eau, Thèse soutenue à l’université Pierre et Marie Curie, Paris, 2015 pp. 7-8.
[6] Jainn- Shiun Chiu, Yuh-Feng Wang, Yu-Cheih Su, Ling-Huei Wei, Jian Guo Liao. Artificial Neural Network to predict skeletal Metastasis in patients with Prostate cancer, Journal of medical systems, vol. 33, Issue 2 April 2009, pp 91-100.
[7] Rigau J. Tiguert, R. le Normand L., Karam G. Glimain P., Buzelin J. M. et al. Prognostic value of bone scan in patients with metastatic prostate cancer treated initially with androgen deprivation therapy J. urol. 168 (4pt 1): 1423-1426, 2002.
[8] J. T. Batuello, E. J. Gamito, E. D. Crawford, M. Han, A. W. Partin, D. G. McLeod, et C. O’Donnell, «Artificial neural network model for the assessment of lymph node spread in patients with clinically localized prostate cancer», Urology, vol. 57, no 3, p. 481–485, 2001.
[9] Kerkeni N. Classification des stades de sommeil par des réseaux de neurones artificiels hiérarchiques IRBM, 2012 vol. 33, pp. 25-40.
[10] S. Xingjian, Z. Chen, H. Wang, D.-Y. Yeung, W.-k. Wong, and W.-c. Woo. Convolutional lstm network: A machine learning approach for precipitation nowcasting. In Advances in Neural Information Processing Systems, 2015.
[11] A. J. Tatem, H. G. Lewis, P. M. Atkinson, et M. S. Nixon, «Super-resolution target identification from remotely sensed images using a Hopfield neural network», IEEE Trans. Geosci. Remote Sens., vol. 39, no 4, p. 781-796, 2001.
[12] M. Negreiros et A. Palhano, «The capacitated centred clustering problem», Comput. Oper Res., vol. 33, no 6, p. 1639-1663, 2006.
[13] C. L. Forgy, «Rete: A fast algorithm for the many pattern/many object pattern match problem», in Readings in Artificial Intelligence and Databases, Elsevier, 1989, p. 547-559.
[14] Kohonen T. Self organizing maps, Berlin Heidelberg: Springer Verlag 2001 pp. 501 pages.
Cite This Article
  • APA Style

    Ndong Boucar, Djigo Mamoudou Salif, Mboup Mamadou Lamine, Tall Khaly, Bathily El Hadji Amadou Lamine, et al. (2021). Neural Method Based on Kohonen Topological Maps Applied to the Whole-body Scintigraphy Image. International Journal of Medical Imaging, 9(4), 189-192. https://doi.org/10.11648/j.ijmi.20210904.14

    Copy | Download

    ACS Style

    Ndong Boucar; Djigo Mamoudou Salif; Mboup Mamadou Lamine; Tall Khaly; Bathily El Hadji Amadou Lamine, et al. Neural Method Based on Kohonen Topological Maps Applied to the Whole-body Scintigraphy Image. Int. J. Med. Imaging 2021, 9(4), 189-192. doi: 10.11648/j.ijmi.20210904.14

    Copy | Download

    AMA Style

    Ndong Boucar, Djigo Mamoudou Salif, Mboup Mamadou Lamine, Tall Khaly, Bathily El Hadji Amadou Lamine, et al. Neural Method Based on Kohonen Topological Maps Applied to the Whole-body Scintigraphy Image. Int J Med Imaging. 2021;9(4):189-192. doi: 10.11648/j.ijmi.20210904.14

    Copy | Download

  • @article{10.11648/j.ijmi.20210904.14,
      author = {Ndong Boucar and Djigo Mamoudou Salif and Mboup Mamadou Lamine and Tall Khaly and Bathily El Hadji Amadou Lamine and Diop Ousseynou and Akpo Géraud Léra Kelvin and Badji Nfally and Mbaye Gora and Diouf Augustin Louis Diaga and Sy Pape Mady and Djiboune Alphonse and Fashinan Herbert and Farssi Mohamed and Ndoye Oumar and Diarra Mounibé and Mbodji Mamadou},
      title = {Neural Method Based on Kohonen Topological Maps Applied to the Whole-body Scintigraphy Image},
      journal = {International Journal of Medical Imaging},
      volume = {9},
      number = {4},
      pages = {189-192},
      doi = {10.11648/j.ijmi.20210904.14},
      url = {https://doi.org/10.11648/j.ijmi.20210904.14},
      eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ijmi.20210904.14},
      abstract = {Objective: The aim of this study was to describe the stages of learning Kohonen's self-organizing maps applied to scintigraphy imaging in order to perform classification for medical diagnostic aid. Method: To achieve these goals, the neurons, arranged on a regular grid, are connected to each other by a neighbor relationship, which creates the topology of the map. The input layer consisted of pixels from the scintigraphy images. Results: During the iteration rounds of learning, we have seen a deployment of neurons on the nodes of the map that becomes more and more important. And it is the same for the winning neurons. After 750 iterations, the Davies Bouldin index attests to the end of the training with a quantization error that goes from 0.175 at the beginning of the training to 0.0225 at the end of the training. After this study, we find that neurons 41, 62, 121, 101 and 145 have captured most of the data with a peak uptake achieved by neuron 41 which has captured 1048 data. This individualizes the class of high intensities undoubtedly corresponding to metastatic hyperfixations. Conclusion: This innovative method could undoubtedly be integrated into the link in the chain highlighting periarticular metastases in developing countries, most of which do not have a SPECT-CT.},
     year = {2021}
    }
    

    Copy | Download

  • TY  - JOUR
    T1  - Neural Method Based on Kohonen Topological Maps Applied to the Whole-body Scintigraphy Image
    AU  - Ndong Boucar
    AU  - Djigo Mamoudou Salif
    AU  - Mboup Mamadou Lamine
    AU  - Tall Khaly
    AU  - Bathily El Hadji Amadou Lamine
    AU  - Diop Ousseynou
    AU  - Akpo Géraud Léra Kelvin
    AU  - Badji Nfally
    AU  - Mbaye Gora
    AU  - Diouf Augustin Louis Diaga
    AU  - Sy Pape Mady
    AU  - Djiboune Alphonse
    AU  - Fashinan Herbert
    AU  - Farssi Mohamed
    AU  - Ndoye Oumar
    AU  - Diarra Mounibé
    AU  - Mbodji Mamadou
    Y1  - 2021/11/12
    PY  - 2021
    N1  - https://doi.org/10.11648/j.ijmi.20210904.14
    DO  - 10.11648/j.ijmi.20210904.14
    T2  - International Journal of Medical Imaging
    JF  - International Journal of Medical Imaging
    JO  - International Journal of Medical Imaging
    SP  - 189
    EP  - 192
    PB  - Science Publishing Group
    SN  - 2330-832X
    UR  - https://doi.org/10.11648/j.ijmi.20210904.14
    AB  - Objective: The aim of this study was to describe the stages of learning Kohonen's self-organizing maps applied to scintigraphy imaging in order to perform classification for medical diagnostic aid. Method: To achieve these goals, the neurons, arranged on a regular grid, are connected to each other by a neighbor relationship, which creates the topology of the map. The input layer consisted of pixels from the scintigraphy images. Results: During the iteration rounds of learning, we have seen a deployment of neurons on the nodes of the map that becomes more and more important. And it is the same for the winning neurons. After 750 iterations, the Davies Bouldin index attests to the end of the training with a quantization error that goes from 0.175 at the beginning of the training to 0.0225 at the end of the training. After this study, we find that neurons 41, 62, 121, 101 and 145 have captured most of the data with a peak uptake achieved by neuron 41 which has captured 1048 data. This individualizes the class of high intensities undoubtedly corresponding to metastatic hyperfixations. Conclusion: This innovative method could undoubtedly be integrated into the link in the chain highlighting periarticular metastases in developing countries, most of which do not have a SPECT-CT.
    VL  - 9
    IS  - 4
    ER  - 

    Copy | Download

Author Information
  • Nuclear Medicine Department of Dalal Jamm Hospital, Cheikh Anta Diop University, Dakar, Senegal

  • Nuclear Medicine Department of Dalal Jamm Hospital, Cheikh Anta Diop University, Dakar, Senegal

  • Bio-informatics Laboratory ESP, Cheikh Anta Diop University, Dakar, Senegal

  • Bio-informatics Laboratory ESP, Cheikh Anta Diop University, Dakar, Senegal

  • Idrissa Pouye General Hospital of Grand Yoff, Cheikh Anta Diop University, Dakar, Senegal

  • Nuclear Medicine Department of Dalal Jamm Hospital, Cheikh Anta Diop University, Dakar, Senegal

  • Radiology Hospital Aristide Le Dantec, Cheikh Anta Diop University, Dakar, Senegal

  • Radiology Hospital Aristide Le Dantec, Cheikh Anta Diop University, Dakar, Senegal

  • Pharmaceutical Biophysics Department, Cheikh Anta Diop University, Dakar, Senegal

  • Pharmaceutical Biophysics Department, Cheikh Anta Diop University, Dakar, Senegal

  • Pharmaceutical Biophysics Department, Cheikh Anta Diop University, Dakar, Senegal

  • Pharmaceutical Biophysics Department, Cheikh Anta Diop University, Dakar, Senegal

  • Nuclear Medicine Department of Dalal Jamm Hospital, Cheikh Anta Diop University, Dakar, Senegal

  • Bio-informatics Laboratory ESP, Cheikh Anta Diop University, Dakar, Senegal

  • Idrissa Pouye General Hospital of Grand Yoff, Cheikh Anta Diop University, Dakar, Senegal

  • Idrissa Pouye General Hospital of Grand Yoff, Cheikh Anta Diop University, Dakar, Senegal

  • Idrissa Pouye General Hospital of Grand Yoff, Cheikh Anta Diop University, Dakar, Senegal

  • Sections