Sinus node dysfunction syndrome: making decision by quantum genetic algorithm, graph neural networks

Размещено на http://www.allbest.ru/

Sinus node dysfunction syndrome: making decision by quantum genetic algorithm, graph neural networks

Kulishov Sergii Kostyantynovych D.Ph., D.Sci., professor

Poltava State Medical University, Poltava, Ukraine

Summary. Quantum genetic algorithms, graph neural networks using for making decisions by sinus node dysfunction syndrome diagnosis are present in this publication. dragon graph neural network sinus node dysfunction

Key words: Graph neural networks, sinus node dysfunction syndrome

Introduction

It's known that genetic algorithms, graph neural networks are basic directions [1] to artificial intellect formation. Deep Neural Networks for ECG analysis was close to the standard in clinical practice [2]. The purpose of this study was to determine principles of quantum genetic algorithm, graph neural networks using for using for making decisions by sinus node dysfunction syndrome.

Methods

Subjective, objective and additional investigations were as diagnostic criteria for sinus node dysfunction syndrome diagnosis [3,4]. Etiology of sinus node dysfunction syndrome presented on visual programming language "Dragon" [5] (fig. 1).

Preliminary diagnosis of sinus node dysfunction syndrome presented on visual programming language "Dragon" [5] (fig. 2).

Additional investigation for diagnosis of sinus node dysfunction syndrome as predisposing to formulation of clinical diagnosis presented on visual programming language "Dragon" [5] (fig. 3).

We were used Typical Conceptual Spaces by some principles [6] as information is organized by quality dimensions that are sorted into domains; domains are endowed with a topology or metric; similarity is represented by distance in a conceptual space [6].

We used our modified domains [7] for analysis of electrical myocardial instability:

Domain 1. Qualitative characteristics of ECG waves, segments, intervals (Types of ECG elements by forms as arcs, triangulates and others);

Domain 2. Quantitative characteristics of ECG waves, segments, intervals (Values of amplitudes, durations of ECG elements);

Domain 3. ECG elements and their origin (Heart morphological origins of ECG elements: sinus, atrioventricular nodes, conduction system and others);

Fig. 1. Sinus node dysfunction syndrome: etiology by visual programming on "Dragon" language [5].

Fig. 2. Preliminary diagnosis of sinus node dysfunction syndrome by visual programming on language "Dragon" [5].

Domain 4. Qualitative characteristics of "Gift wrapping" algorithm of ECG element analysis (The sequence of the connection ECG elements' as "Gift wrapping";

Domain 5. Quantitative characteristics of "Gift wrapping" algorithm of ECG element analysis (Values of surface, volume characteristics of ECG elements "Gift wrapping" [1,3,4,5] and distances between points, angles of polygons);

Domain 6. Qualitative and quantitative characteristics of ECG elements (Data of Delaunay triangulation, Voronoi Diagrams, convex analysis;

Domain 7. Qualitative and quantitative characteristics of 2D ECG elements;

Example of the quantum genetic algorithm using for differential diagnosis of antonym, oxymoron like heart electrical instabilities for sinus node dysfunction syndrome or binodal syndrome by some qubit chromosomes [7]:

|q> = a1|0> + a2|1>

|q1 > = a1 (sinoatrial blockade II stage)|0> + a2 (atrial fibrillation) |1>

|q2> = a1 (tachy- brady- syndrome)|0> a2 (atrial flutter) |1>

|q3 > = a1 (reentrant supraventricular tachycardia)|0> + a2 (Trifascicular Block - Right Bundle Branch Block with Both Left anterior fascicular block and Left posterior fascicular block) |1>

|q4 > = a1 (Atrioventricular III block)|0> + a2 (atrial fibrillation) |1>

|q5 > = a1 (sinus bradycardia)|0> + a2 (atrial tachyarrhythmia) |1>

|q6 > = a1 (sinus pause )|0> + a2 (atrial fibrillation) |1>

|q 7> = a1 (sinoatrial exit block|0> + a2 (sinus bradycardia) |1>

|q 8> = a1 (sinoatrial blockade II stage)|0> + a2 (atrial fibrillation) 11>

|q9> = a1 (pair supraventricular extrasystoles)|0> + a2 (short supraventricular tachycardia) |1>

|q10> = a1 (pair atrial extrasystoles)|0> + a2 (Short runs of atrial arrhythmia| 1>

|q11> = a1 (tachycardia syndrome)|0> + a2 (bradycardia syndrome ) |1>

And so on

Fig. 3. Additional investigation for diagnosis of sinus node dysfunction syndrome as predisposition to formulation of clinical diagnosis by visual programming on language "Dragon" [5].

Results

Implementation of our algorithms on visual programming language "Dragon", using modified domains of Typical Conceptual Spaces, quantum genetic algorithms gave us possibilities to graph neural networks decision-making of sinus node dysfunction syndrome diagnosis (Fig. 4):

We used Shannon [8,10,11] and Renyi entropy [9] for sinus node dysfunction

Fig. 4. Graph neural network using for making decisions by sinus node dysfunction syndrome diagnosis

Discussion

Principles of conceptual spaces domains, Shannon and Renyi entropy, quantum genetic algorithms, graph neural networks using for making decisions for sinus node dysfunction diagnosis presented in this article.

Some authors [12] developed an algorithm, which exceeds the performance of board certified cardiologists in detecting a wide range of heart arrhythmias from electrocardiograms recorded with a single-lead wearable monitor. On dataset, they

[12] train a 34-layer convolutional neural network, which maps a sequence of ECG samples to a sequence of rhythm classes. They [12] exceed the average cardiologist performance in both recall (sensitivity) and precision (positive predictive value). Key to exceeding expert performance is a deep convolutional network, which can map a sequence of ECG samples to a sequence of arrhythmia annotations along with a novel dataset two orders of magnitude larger than previous datasets of its kind [12].

Different investigations [13,14,15,16] gave data that antisymmetry is the symmetry of elements that are identical in shape, but contrasting in "content".

Sinus node dysfunction of different etiology is the symmetry of elements that are identical in shape, but contrasting in "content" by etiological peculiarities. It's necessary to build additional neuronets by others criteria of subjective, objective and additional investigations, according to etiology, pathogenesis of this syndrome.

Conclusion

1. Quantum genetic algorithm, graph neural networks using for making decisions for sinus node dysfunction diagnosis presented in this investigation.

2.

3. Shannon and Renyi entropy of heart electrical instabilities were as criteria

for sinus node dysfunction syndrome diagnosis, for graph neural networks building.

4. It's necessary to build additional neuronets by others criteria of subjective,

objective and additional investigations, according to etiology, pathogenesis of this syndrome.

References:

[1] Demuth H.B., Beale M.H., De Jesus O. (2014). Neural Network Design. e-Book 2nd ed.

Edition ISBN-10: 0971732116 ISBN-13 : 978-0971732117

[2] Ribeiro, A.H., Ribeiro, M.H., Paixao, G.M.M., Oliveira, D.M., Gomes, P.R., Canazart, J.A.,

Ferreira, J.M.P.S., Andersson, C.R., Macfarlane, P.W., Wagner Meira Jr., Schon, T.B. &

Ribeiro, A.L.P. (2020). Automatic diagnosis of the 12-lead ECG using a deep neural network. Nat. Commun. (11), 1760. https://doi.org/10.1038/s41467-020-15432-4

[3] Kantharia, B.K. (2018). Sinus Node Dysfunction.Medscape.

https://emedicine.medscape.com/article/158064-overview

[4] Mitchell, L.B. (2021). Sinus Node Dysfunction (Sick Sinus Syndrome). MSD manuals. https://www.msdmanuals.com/professional/cardiovascular-disorders/arrhythmias-and- conduction-disorders/sinus-node-dysfunction

[5] Parondzhanov V.D. (2017) Pochemu vrachi ubivayut i kalechat patsiyentov ili Zachem vrachu blok-skhemy algoritmov? Illyustrirovannyye algoritmy diagnostiki i lecheniya - perspektivnyy put' razvitiya meditsiny. Klinicheskoye myshleniye vysokoy tochnosti i yuyezopasnost'patsiyentov. M.: DMK Press.

[6] Gardenfors, P, Williams, M-A. (2015) Reasoning about Categories in Conceptual Spaces, Proceeding IJCAI'01 Proceedings of the 17th international joint conference on Artificial intelligence, Morgan Kaufmann Publishers Inc. San Francisco, CA, (1), 385-392.

[7] Kulishov, S.K. (2021) Sinus node dysfunction syndrome: mechanisms, diagnosis, treatment Poltava, ResearchGate, RG. http://dx.doi.org/10.13140/RG.2.2.28442.31688

[8] Kozlowski, L. Shannon entropy calculator. www.shannonentropy.netmark.pl

[9] Gao, J, Hu, J, Buckley, T, White, K, Hass, C (2011). Shannon and Renyi Entropies to Classify Effects of Mild Traumatic Brain Injury on Postural Sway. PLoS ONE 6(9): e24446. https://doi.org/10.1371/journal.pone.0024446

[10] Shannon, C. (1948). A Mathematical Theory of Communication. Bell System Technical Journal 27, 329-423, 623-56.

[11] Guiard, Y., Gori, J., Roy, Q., Rioul, O. (2017). Not Just Pointing: Shannon's Information Theory as a General Tool for Performance Evaluation of Input Techniques. (hal-02090158)

[12] Rajpurkar, P., Hannun, A.Y., Haghpanahi, M., Bourn, C., Andrew Y. Ng. (2017). Cardiologist-Level Arrhythmia Detection with Convolutional Neural Networks arXiv:1707.01836v1 [cs.CV] 6Jul 2017

[13] Padmanabhan, H., Munro, J.M., Dabo, I., Gopalan, V. (2020) Antisymmetry: Fundamentals

and Applications. Annual Review of Materials Research,(50),255-281.

https://www.annualreviews.org/doi/abs/10.1146/annurev-matsci-100219-101404

[14] Schmied, R. (2019). Using Mathematica for Quantum Mechanicd. A Student's Manual University of Basel, Switzerland, 164 p. arXiv:1403.7050v3 [quant-ph] https://doi.org/10.48550/arXiv.1403.7050

[15] Klimenko, A.Y. (2016). Symmetric and antisymmetric forms of the Pauli master equation for interaction of matter and antimatter quantum states. Nature.com - Scientific Reports 6, 29942. Quantum Physics (quant-ph)arXiv preprint arXiv:1611.04582, 2016 - arxiv.org

[16] Petitjean M. (2021). Symmetry, Antisymmetry, and Chirality: Use and Misuse of Terminology. Symmetry. 13(4):603. https://doi.org/10.3390/sym13040603

Размещено на Allbest.ru