Progressive artificial neural network for medical applications
Description
This article is presented by professionals, working in diverse fields and combining their knowledge in artificial neural networks with decades of experience in application of TRIZ, the Theory of Inventive Problem Solving. The article describes problems associated with the impossibility of effectively using existing neural networks for the analysis of medical information. Their solution proposes utilization of the recently invented, fundamentally new system - Progressive Artificial Neural Network (PANN). A description of PANN and its advantages is presented. The example of PANN implementation is shown with some possible applications of PANN, together with TRIZ, for conducting important research in the field of medicine, invention, and the development of new medical equipment and diagnostic and treatment methods. In particular, the authors propose:
• Intelligent Database Management System in Medicine (IDBMSM),
• System "Medical Advisor"
• System "Personal Physician"
• System "Physician Assistant - Researcher"
• Medical Control System “Identify an error (mistake)”
• System "Development of medical technology"
• System of mass medical monitoring
• Medical training system
The authors believe that the use of the PANN system will be a very important step in the development of both the theory and practical aspects of medicine.
Files
farber1FIN.pdf
Files
(431.8 kB)
Name | Size | Download all |
---|---|---|
md5:5bd5bec59fe4c8335858772da7e39862
|
431.8 kB | Preview Download |
Additional details
References
- 1. Altshuller customers. The Innovation Algorithm: TRIZ, systematic innovation and technical creativity, 1999
- 2. Altshuller G.S.Creativity as an Exact Science 1984
- 3. Zlotin B.L., Zusman A.V. Solving research problems, 1991
- 4. Boris Zlotin Alla Zusman, Stan Kaplan, Svetlana Visnepolschi, New Tools for Failure and Risk Analysis / Anticipatory Failure Determination 1999
- 5. Farber, B.S.; Martynov, A.V.; Kleyn, I.R. Creation of new medical drugs based on TRIZ and computer mathematical modeling. Annals of Mechnikov Institute, N 4, 2018 p.15-32. https://zenodo.org/record/2547580#.Ya7ZgFXMI8l
- 6. Farber, B.S.; Martynov, A.V.; Osolodchenko T.I, Kleyn, I.R. Application of synergetic set of TRIZ principles for developing cAMP - accumulation activators and their influence on multi-drug resistance microorganisms. Annals of Mechnikov Institute, N 1, 2019 p.47-54. https://zenodo.org/record/2639505#.XLMwluj7SUk
- 7. Jean-Christophe B. Loiseau. Rosenblatt's perceptron, the first modern neural network. A quick introduction to deep learning for beginners. Mar 11, 2019· https://towardsdatascience.com/rosenblatts-perceptron-the-very-first-neural-network-37a3ec09038a
- 8. Boris L. Zlotin, Vladimir N. Proseanic, Anatoly A. Guin, Ivan I. Nehrishnyi Sr, Ivan I. Nehrishnyi Jr, Vladimir S. Matsenko, Anton N. Belyakov, Gafur A. Zainiev. A New Approach to Building Artificial Neural Networks and Medicine.
- 9. A SPECIAL REPORT: The Great AI Reckoning. Why Is AI So Dumb? IEEE Spectrum, October 2021. https://spectrum.ieee.org/files/11920/10_Spectrum_2021.pdf
- 10. Patent US 9390373; 2016 Neural network and method of neural network training.
- 11. Patent US 9619749; 2017 Neural network and method of neural network training.
- 12. Patent US 10423694; 2019 Neural network and method of neural network training.