Portal:Complex Systems Digital Campus/E-Laboratory on Machine Learning in Medicine



.
.
the repository for Open Questions, Challenges and Ressources of the
.




e-Laboratory on Machine Learning in Medicine


Challenges edit

The challenge of this e-laboratory is to provide a platform to register, process, analyse and model clinical signal and images in a standardized repository complying technical requirements to improve the development of Complex System technologies. In particular to build up an adequate training set for the definition of Tailored to the Problem (TPS) Mathematical Transforms, in the fields of Engineering applied to Medicine and to Life and Cognition Sciences.

During a first phase Artificial Neural Networks (ANN), which are Complex adaptive systems with inference learning emergent properties, are to be used. Later on, other more advanced methodologies are to be tested.

Trained ANN have computational properties enabling them to be considered Universal Approximators as mapping networks. As stated (Hornik, K., 1989) “any lack of success in applications must arise from inadequate learning, insufficient numbers of hidden units or the lack of a deterministic relationship between input and target”. Therefore this e-lab point towards getting adequate learning and managing possible non- deterministic relations between input and target.

Recently, ANN has been used to define new Mathematical Transforms; Tailored to the Problem Specificity (TPS) (Glaría, A., Taramasco, C., Demongeot, J., 2010) and Dynalet Transforms (DT) (Demongeot, J., Glade N., Forest L., 2007).

TPS methodology can be envisioned as part of the development of new Complex System based technologies which has been used in the field of Engineering applied to Medicine and to Life and Cognitive Sciences (IMVC).

Inadequate learning due to the lack of deterministic relationship between data and its classification makes the accuracy of this method to be critically sensitive, on one part, to the representativeness and completeness of data in the input vectors of the ANN learning set and, on the other hand, to the existence of data clusters associated with different classifications in the target vectors.

Collaborative efforts between new transform developers -both, from the Academy and from the Industry- are required to have access to more representative, complete and well documented data. We believe that UNITWIN UNESCO “Complex System Digital Campus”, aimed to potentiate collaborative networks, could be an adequate environment to enrich completeness in ANN Learning Set data.

The main goal of the proposed e-lab is to register, process, analyse and model clinical data, either signals or images, and its classification on systems for which TPS or Dynalet transforms are being developed. A reliable repository, duly completed and properly provisioned with representative data, is a necessity for the development of reliable Mathematical Transforms -such as TPS or Dynalet- in the context of complex systems technologies with inference learning emergent properties in IMVC domain.


Name, e-mail, website and institution edit

of the responsible for the e-laboratory edit

Antonio Glaría | antonio.glaria@uv.cl | Universidad Valparaíso, Chile

list of the teams participating in the e-laboratory edit

  • PhD. Pablo Reyes, pablo.reyes@uv.cl. Universidad Valparaíso (Chile)
  • PhD. Alejandro Veloz, alejandro.veloz@uv.cl. Universidad Valparaíso (Chile)
  • PhD. Alejandro Weinstein, alejandro.weinstein@uv.cl. Universidad Valparaíso (Chile)
  • PhD. Stéren Chabert, steren.chabert@uv.cl. Universidad de Valparaíso (Chile)
  • PhD. Rodrigo Salas, rodrigo.salas@uv.cl. Universidad de Valaparaíso (Chile)
  • PhD. Carla Taramasco, carla.taramasco@polytechnique.edu. Universidad de Valparaíso (Chile) – CNRS – EHESS (France)
  • PhD. Marta Barría, marta.barria@uv.cl Universidad de Valparaíso (Chile)
  • PhD. Pablo Pérez, pablo.perez@uv.cl Universidad de Valparaíso (Chile)
  • PhD. Carlos Becerra, carlos.becerra@uv.cl Universidad de Valparaíso (Chile)
  • PhD. Carlos Felipe Henríquez, carlos.henriquez@uv.cl Universidad de Valparaíso (Chile)
  • PhD. Mónica Catalán, monica.catalan@uv.cl Universidad de Valparaíso (Chile)
  • PhD. Jacques Demongeot, jacques.demongeot@agim.eu . Université Joseph Fourier, (France)


Coordination committee edit

(to be completed)


Research projects in the e-laboratory edit

  • Formalización y aplicaciones de una metodología para el cálculo de Transformadas Matemática Hechas a la medida de la Especificidad de un Problema (TPS, del inglés: Tailored to the Problem Specificity). (Formalizing and applications of a methodology to estimate Tailored to the Problem Specificity Mathematical Transform) Antonio Glaría, Carla Taramasco, Pablo Reyes. Code U. de Valparaíso DIUV 44/11. 2013 – 2015.
  • Sistema de monitoreo no intrusivo de señales biomédicas. (Non intrusive Biomedical signal Monitoring System) Alejandro Weinstein, Pablo Reyes, Antonio Rienzo. Appear- Network/Universidad de Valparaíso. Code Conicyt (Chile)/ FONDEF IT13110035. 2013 – 2015.
  • Plataforma de Integración Tecnológica para el Registro, Vigilancia y Alerta de Enfermedades de Notificación Obligatoria (Technology integration platform for the notifiable diseases registration, monitoring and alert), Carla Taramasco, Marta Barria, Anibal Vivaceta, Rodrigo Vergara, Universidad de Valparaíso. Code Conicyt (Chile)/ FONDEF IT13110059. 2013 – 2015.
  • Fortalecimiento del postgrado en Departamento de Ingeniería Biomédica de la Universidad de Valparaíso. (Biomedical Engineering Department postgraduate programs enforcement), Jacques Demongeot, Université Joseph Fourier (France), Stéren Chabert, Carla Taramasco, Antonio Glaría, Universidad de Valparaíso (Chile). Code Conicyt(Chile)/ MEC 80110027. 2012 – 2013.
  • Sistema Biométrico y de detección de personal en línea para Minería Subterránea. (Biometric and staff detection system for underground mining), Pablo Reyes, Antonio Glaría, Universidad de Valparíso/ Appear Network. I & D Project supported by APPEAR NETWORKS CHILE. 2012 -2013
  • CORtex and reTINA modelling from an engineering and computational perspective. CORTINA Team. Frederic Alexandre, INRIA (France), Adrián Palacios, Rodrigo Salas, Stéren Chabert. Universidad de Valparaíso (Chile). INRIA Associate Teams Programme. Universidad de Valparaíso. INRIA. 2011 – 2013.
  • Identificación de Parámetros Espacio-Temporales en Imágenes Funcionales. (Spatio- temporal Parameter identification in functional Magneto- Resonance imaging), A. Veloz, S. Chabert, R. Salas. Code U. de Valparaíso DIUV 2012 - 2014


e-Laboratory Scientific Committee edit

  • Carla Taramasco, PhD
  • Jacques Demongeot, PhD

URL for the Website and/or Wiki of the e-laboratory edit

Grid, Cloud, or other network utilities to be used edit

(to be completed)


Data and/or Tools to be shared edit

  • arterial pulse signal and functional MRI Hemodynamic functions and preliminary codes for TPS analysis written in MATLAB to be published soon
  • http://nimi.uv.cl/ Training Set for non-Invasive and minimally-Intrusive (nImI) Blood Pressure estimates

Results edit

Bibliography edit

  • Contreras, G., Glaría, A. Codifying Temporal Characteristics of Jewett Components to improve Jewett Transform. J. of Physics: Conf. Series; 90 012075. 2007.
  • Daubechies, I. Grossman, A., Meyer, Y. Painless nonorthogonal Expansions. J. Math. Phys. (27): 1271-1283. 1986.
  • Demongeot, J., Hamie A., Glaría, A., Taramasco, C. Dynalet: a new representation of periodic biological signals and spectral data. In: IEEE AINA'13. IEEE Proceedings, Piscataway: 1525-1532. 2013.
  • Demongeot, J., Glade N., Forest L. Liénard systems and Potential-Hamiltonian decomposition: (a) I. Methodology. C. R. Acad. Sci. Paris, Ser. I. (344): 121-126. 2007; (b) II. Algorithm. C. R. Acad. Sci. Paris, Ser. I. (344): 191-194. 2007; (c) III. Applications. C. R. Acad. Sci. Paris, Ser. I. (344): 253-258. 2007; (d) Applications in Biology. C. R. Biologies. (330): 97-106. 2007.
  • Dobai, B.M., Iantovics, L.B., Paiu, A. Exploratory Factor Analysis for Identifying Comorbidities as Risk Factors Among Patients with CIED. Acta Marisiensis. Seria Technologica. 18 (1):47-51, 2021.
  • Enăchescu, C., Iantovics, L.B. Decision Support Systems Based on Methods of Artificial Intelligence in the XXI Century Medicine, Romanian Medical Journal, 69(1):369-374, 2022
  • Fekete, L., Iantovics, L.B., Fekete, G.L. Exploratory Axis Factoring for identifying the self-esteem latent factors and their correlation with the life quality of persons suffering from Vitiligo, Frontiers in Psychology, 14:1200713, 2023.
  • Fourier, J.B.J. Digression sur la manière d'exprimer les fonctions arbitraires par des séries de quantités périodiques in Théorie mathématique de la chaleur. Chapitre VII. Firmin Didot Père et Fils, Paris. 1822.
  • Gecow, A., Iantovics, L.B., Tez, M. Cancer and Chaos and the Complex Network Model of a Multicellular Organism, Biology, 11(9), 1317, 2022.
  • Georgieva, V., Petrov, P., Iantovics, L.B. X-ray image processing for tissue involvement-based caries detection, X-ray image processing for tissue involvement-based caries detection, October 2019, Communication, Electromagnetics and Medical Application (CEMA'19), Sofia, Bulgaria, 22-26, 2019.
  • Glaría, A. Upgrading Fourier: Alamedas.Tutorial 6. EVIC 2012, at http://www.evic.cl . 2012.
  • Glaría, A., Taramasco, C., Demongeot, J. Methodological Proposal to estimate a Tailored to the Problem Specificity Mathematical Transformation. Use of Computer Intelligence to optimize Algorithm Complexity and Application to Auditory Brainstem Responses Modeling. IEEE AINA'10. IEEE Proceedings, Piscataway: 775-781. 2010.
  • Glaría, A., Zepeda H., Chabert S., Hidalgo M., Demongeot J., Taramasco C. Complex adaptive systems with inference learning emergent property to estimate Tailored to the Problem Specificity Mathematical Transforms: three study cases. European Conferences on Complex Systems, Barcelona. 11- 16 Septiembre. 2013.
  • Glover, G. H. Deconvolution of impulse response in event-related BOLD fMRI. NeuroImage. (9): 416–429. 1999.
  • Hornik, K. Multilayer Feedforward Networks are Universal Approximators. Neural Networks. (2): 359-366; 1989.
  • Iantovics, L.B. An Agent-Based Hybrid Medical Complex System, Information - An International Interdisciplinary Journal, 16(6A): 3709-3726, 2013.
  • Iantovics, L.B.: Cognitive Medical Multiagent Systems, BRAIN. Broad Research in Artificial Intelligence and Neuroscience, 1(1): 12-21, 2010.
  • Iantovics, L.B., Iakovidis, D.K., Nechita, E., II-Learn-A Novel Metric for Measuring the Intelligence Increase and Evolution of Artificial Learning Systems, International Journal of Computational Intelligence Systems, 12(2): 1323-1338, 2019.
  • Iantovics, L.B., Marusteri, M., Kountchev, R., Zamfirescu, C-B, Crainicu, B.: Intelligent CMDS Medical Agents with learning Capacity, In Proc. of the 5th Int. Conf. on Virtual learning (ICVL 2010), 29-31 Oct. 2010, Tg. Mures, Romania, In: Vlada, M., Albeanu, G., Popovici, D.M. (Eds.), Bucharest University Press, 2010, pp. 325-331.
  • Iantovics, L.B., Rotar, C., Morar, F.. Survey on establishing the optimal number of factors in exploratory factor analysis applied to data mining, Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery 9(2): e1294, 2019.
  • Kreider D.L., Kuller R.G., Oestberg D.R., Perkins F.W. An introduction to linear analysis. Adisson -Wesley Pub. Co. Reading, USA. 1966.
  • Kountchev, R., Iantovics, L.B. (Eds.): Advances in Intelligent Analysis of Medical Data and Decision Support Systems, Studies in Computational Intelligence (series), Springer, 473, 2013.
  • Kounchev, R., Iantovics, L.B., Kountcheva, R.: Efficient Management of Medical Image Databases Based on Inverse Pyramid Decomposition, In Proc. of the 5th Int. Conf. on Virtual learning (ICVL 2010), 29-31 Oct. 2010, Tg. Mures, Romania, In: Vlada, M., Albeanu, G., Popovici, D.M. (Eds.), Bucharest University Press, 2010, pp.395-402.
  • RK Kountchev, BL Iantovics, RA Kountcheva, Hierarchical third‐order tensor decomposition through inverse difference pyramid based on the three‐dimensional Walsh–Hadamard transform with applications in data mining, Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 10(2):e1314, 2020.
  • Lazar. N. The Statistical Analysis of Functional MRI Data". Springer. New York. Chapter 5: 74. 2008.
  • Lemarié P.G. y Meyer, Y. Ondelettes et bases hilbertiennes. Rev. Matemática Iberoamericana (2): 1-18. 1986.
  • Macia, N. Modeling and Control of Dynamic Systems. Thomson Delmar Learning. Clifton Park, NY: 52-73. 2005.
  • MacKay, D.J.C. Information Theory, Inference and Learning algorithms. Cambridge University Press. 2003
  • Revett, K., Iantovics, L.B. A Survey of Electronic Fetal Monitoring: A Computational Perspective. In: Iantovics, B., Kountchev, R. (eds) Advanced Intelligent Computational Technologies and Decision Support Systems. Studies in Computational Intelligence, vol 486. Springer, Cham, 135–141, 2014.



Return to the Portal of the Complex Systems Digital Campus