Academic Open Internet Journal
Volume 10, 2003



New Devices and Approaches for Medical Imaging


A.Yu. Alevanau

Mid Sweden University



Prospective way to build new detectors and mechanisms for medical imaging is described. Consideration is based on interdisciplinary approach joining general concepts of computer science and models of its physical realization.




Mostly known concept of computer science is concept of memory. Turing computer is based on it [1]. Both its data and programs are stored in memory and are used by central processing unit for operation. Different sequential computer architectures implement different algorithms of access and use of memory [2]. Parallel computation in neural architectures implies use of distributed memory that differs from memory in digital binary systems mostly in regard of its analogous presentation and distributed localization [3]. However, both computational paradigms – of sequential and parallel computations have the same general characteristic. This is ones’ open character in regard of unavoidable presence of inputs and outputs.

The other necessary property of any useful information processing system is localization of information how to process and temporary storage of what to process in internal memory of the system. These two characteristics – open character of input and output channels and localized character of memory, even if networking paradigm and distributed networked calculations develop, are the basic elements of the well established way of thinking in design of computer architectures.

How do these well known facts relate to medical imaging and detectors for it?

The link is not straight. It appears from sequential consideration of paradigms for distributed computations yielding physical possibilities to create detectors for medical imaging. That means description of some physical ideas for this from analogies taken in consideration of computational concepts. Such a useful concept is concept of memory in distributed information medium that is logically closed [4]. The later one objectivity implies application of the concept of memory in a medium, where is no hardware to be regarded as memory. Medium itself should be constructed so, that external observer of its dynamics can select appeared objects inside the medium that can be considered as objects whose condition is recorded in the same medium. The observer has to select objects carrying information about mentioned above objects of other type. One has to select also mechanisms to store this information in the medium and to recall it.

However, this way of what the observer has to look for is coming from traditional paradigm of what memory is and how it should be used. This paradigm is supported by existence of the definite construction and design of computer hardware that has memory chips storing bits of information, buses to move and change these bits and processing units to operate with these bits in useful way. Not much difference is observed if take a look for the way of use of neural networks. Coded in different ways information is stored in neural networks, processed by it, and extracted from it to output.

The approach for creation of functional systems in distributed medium presented in [4] is different. The medium itself can be constructed so, that it will have parts able to perform some intellectual tasks in dynamics inside the medium. This intellectual character of behavior of the selected parts of the medium can be deduced as result of observations of the dynamics of these parts in time and from ones’ form of exchange with the whole medium. The property of intellectuality can be concluded if one will observe properties of memorizing of some conditions of the parts in medium and recalling of these conditions from one in dynamics. The dynamics of the selected parts of medium represents some function, performed by ones in exchange with the medium. The medium provides the parts with memory and mechanism of access to it. The whole medium is consisted from different parts made up from the very elementary ones. This differentiation provides with possibility to divide parts on the ones whose condition is recorded in slight changes of condition of other parts. Also one can select forms of motion correspondent to these changes and select carriers of the exchange between the parts and the whole recording and recalling medium.

The task to construct such a closed distributed information medium was formulated in [4] with assumption of possibility to use multi layer logical division in the medium, where objects of one logical layer interact with objects of other layers in described above way of exchange and perform some intellectual functions. Usage of such a medium can be made up on one of these layers by interfacing it to external informational inputs. However, such a usage is limited in time in principle, as interference of external perturbations to logically closed system will change laws and logic of its functioning and will destroy functioning of chosen systems on chosen layer together with all the other ones on all other layers.

This is analogous to loosing of associative memory property in Hopfield neural networks when it exceeds limit of inputs [3].

Nevertheless, consideration of logically closed distributed information medium is valuable from the general point of view. It brings unexpected conclusions for physics in general and medical imagery in particular. This consideration and its conclusions are presented below in the main parts of the paper.


1.      Associative principle and its physical counterparts


What kind of intellectual dynamics can exist in the environment described above?

The first logical step to think about such a dynamics is to divide the selected system in the medium on two parts. These two parts can have joint functional dynamics revealing intellectual properties in exchange with the medium. Obvious characteristic of intellectuality of the exchange is revealing of associative memory’s algorithms in the exchange. That is described in details and relatively to example physical systems of self-organization in [4]. To represent shortly its main idea here while taking to account its new physical yield, we can directly introduce analogies with properties of physical space into consideration of distributed memory properties in the medium.

The first and mostly primitive memory property in such a medium is constant existence of the selected systems in it. The medium can be modeled, for example, by cellular automata space presented in [5]. The system in it is phase difference between locally connected counters by module selected because of its steady form. The network of counters has hidden internal mechanism to preserve synchronization among counters in the whole network. The defects of global phase synchronization are analogous to ones described in model of space [6]. This model models space by abstract lattice analogous to researched in physics of solid state.

Both models have these properties of global support for synchronization and solid character of lattice as analogies to introduced here logically closed and united character of distributed medium. The close character appears to be a base for consideration of the very existence of any system in space considered as yield of work of distributed dynamic memory in that space. Existence of any selected combination of elements of medium is supported by layered motion in the medium in cyclic and closed paths.

Regarding two mentioned above models, this motion in closed paths corresponds to action of mechanism to preserve synchronization in the network of counters [5] and to motions of phonons or gravitational waves in the lattice in model [6]. The supported systems in this motion are the structures of phase difference between the counters in [5] and structures of defects in [6].

What does allow designation of this mechanism of existence in dynamics as memory property? It has direct analogy with work of hardware of dynamic memory [7]. The difference is just in logically closed and cyclic character of dynamical processes of regenerating of system’s structure in the exchange with the whole medium.

So one can call the carriers of motion from selected systems to the whole medium and back as carriers of distributed memory in the medium. Here we come to conclusion that these processes of recording and recalling are going concurrently and simultaneously providing the systems inside medium with ones’ basis of formation and existence.

What is related to description of system divided on associatively interacting parts [4], this property looks obvious if one considers internal dynamics of the system preserving its structure and form in exchange with the memory in space. This intrinsic property of any system in distributed medium to preserve its form in a process of associative recall of conditions of all its internal parts in exchange between ones and with the medium is designated in [4] in terms of action of associative principle. The principle is applied to analysis of internal dynamics of the functional systems discussed there. And even more, to analysis of natural systems having fractal properties of surrounding neighborhood in dynamics of big time scales [4]. The memory properties of natural distributed media with fractals in output and input channels feeding dynamical systems observed inside the media were described in [8]. Sample dynamical system with fractal input and output channels described are natural basins, whose changes in level of water on long periods in time are governed by Hurst statistics [8]. This statistics is observed also on various natural dynamical processes of large space and time scales [8]. Application of the statistics allows detection of so-called self-sustainable tendencies in dynamics of observed parameters. These tendencies elucidate long-range memory effects in the exchange of observed system with surrounding environment possessing with fractal properties of channels feeding the system input and output [8].

This note could be very illustrative for the following logical assumption about properties of motion of memory carriers moving out from the selected system to medium and vice versa.

This motion necessarily has to pass energy through spatial scales. Transformations of energy motional forms during passing through different scales are called in [4] information-scaling transformations. Ones’ physical existence reveals itself in creation of fractals. Fractals present direct evidence of interaction between different scale and logical levels of objects’ existence in the medium. Let us say so: memory carriers from one level of scales move out and back in the medium and interact during this motion with objects of other logical and scale levels, helping to form a fractal.

This assumption and observation together with the general explanation of the phenomenon of fractals itself is a key for the experiments proposed below and prospective devices for medical imaging. It allows for researcher to search forms of dynamics on relative macro scale for interaction with input and output regenerative flows of memory carriers for dynamical processes and forms on relative micro scale.

It looks obvious from previous material that this interaction can be found not only using fractals described in theory and consisted from the forms replicated on every consecutive scale. One has to find out just form of transformation or reflection of objects and processes of scale of interest to scale of measurement.

Knowledge of laws of motion of memory carriers, designated in [4] as reflecting structures may help for this task very much. General theory applicable for this task is theory of scale relativity [9-10]. It has several experimental proofs on astrophysical scales and its developers are working on new proofs and applications on micro and medium scales. As this theory has not yet proofs and applications for medical imaging, the logical reasoning and steps towards the experiment and such applications are presented in this paper below.


2.      Methodology and prospective experimental proofs of the idea


Let us interpret some known phenomena in context of the idea about division of every memory system onto two parts. These parts were mentioned above relatively to associative principle [4]. Their roles can be designated as correspondent to address and data in conventional memory. The difference is described in [4] on example of system of differential equations having associative relationships between variables. To tell the essence in words, it could be said that which part of the system is considered as receiving or sending data, and which part is considered as address for these data is determined by character of dynamic and nonlinear relationships between the parts of the system and the whole medium. Due to this character, roles of the parts can change with the time. Mostly appropriate systems to observe and use such a dynamics in exchange with the medium are so-called systems of self-organization [11]. Example of associative memory made up on typical such system of Benar cells in context of the approach is described in [4,12].

If one comes to more simply examples just to the level of single system in exchange with the medium, as it was described above, some physical parallels should be made.

Let us say that the system in abstract distributed medium discussed above is physical body in space. The space is considered as medium above, providing the body with means of its formation and exchange with the space in the whole. The fact that space provides selected as its body area with means to be formed is evident for observer. The fact that it provides ones for the body in constantly going on exchange is not so evident. One can say that properties of motion of body moving in space are preserved by exchange with the space. Logical base of such a point is assumption of existence of some agent preserving unity of the whole space. Logic of this assumption is not a proof for observer, who doesn’t see motion of derivatives from this agent. One can say, if there is no mechanism to preserve all the parts of space in unity, why the space does not break on its parts? Logic of this question is also not proof for the partly blind observer.

Even if author doesn’t see any flaw in logic of reasoning why any object in space is existent in exchange with it in the whole, there is need to provide this partly blind and skeptical observer with other proofs of the point.

 Unity of space means existence of some interaction between its elements that links ones with infinite speed into the kind of solid object having universal logic of changes. This solid object is analogous to models of closed finite size cellular space in [5] and solid lattice as space and absolute system of reference in [6]. However, if we just assume described above property of dynamic memory as basis for any object’s existence and dynamics in real physical space, we can prove existence of the closed cyclic and regenerative motion of memory carriers by simple experiments on interaction between pairs of colliding bodies.

To provide with the properties of formation and motion for any object in space, the memory carriers have to move very fast. At least, to provide us with photons and with their properties of motion, ones have to move much faster than photons. How one can detect existence of such a fast motion and energy transfer in it? The answer is: by detection of interaction between also very fast processes. Such fast processes are processes of collisions. These processes also are the direct application of memory paradigm for ones’ consideration. Let us say so: as there are reflecting structures made up from memory carriers for composite bodies in space, the composition of two colliding bodies also is recorded and recalled.

Next logical step is to assume possibility for reflecting structures moving out from colliding bodies to influence directly on reachable collision with the close dynamical parameters of its “address” and “data” parts.

This mechanism of energy transfer between objects of different scale levels of motion is basis for described above mechanism of formation of fractals. It is necessary to note, that derivatives from the infinite speed and common for all the elements of space joint logic of changes can have different speeds and specifics of motion laws in the space. So reflecting structures or memory carriers can be different and move differently for different combinations of elements of space. And thus could be considered as objects of different not only scale but also logical levels. What is most important in this – the possibility of direct interaction between objects of various scales and logical levels during motion of reflecting structures or memory carriers through the levels. This assumption allows approach for creation of systems for medical imaging based on scale resolved observations of correlated fluctuations in dynamical characteristics of interactions between objects on the scales. Measured dynamics of objects on macro scale levels has to be linked with dynamics of objects on micro scale levels.

By the idea, the laws of interaction of electric currents can be considered as direct consequence of this energy transfer between the systems of colliding particles on micro scale level to macro scale one. Specific property of collisions in electric currents is that at least one from the colliding particles has property linked to electric charge.

This property can be considered as “address” or “data” in the exchange. As any current can be considered as process made up from collisions (if not with other current carriers in conductors, then with virtual particles in vacuum), the laws of electrodynamics can be yielded by this idea from detailed consideration of energy transfer between the collisions possessing with charge as “address” or “data” characteristic.

To prove this possibility in part, and to provide the idea with more proofs, one should propose experiment on interaction between currents made up from collisions with different “address” and “data” parts. These parts can be chosen as not electrical charges, but, for example, baryonic charges carried by neutral nuclear particles. The experiment to prove possibility of such an interaction is described below. Together with described above possibility for interactions between dynamical processes of different scales, one should provide with basis for new applications in medical imaging.

This experiment should be done using neutrons as current curriers.

As measurements analogous to measurements of interactions between electric currents are not possible to reproduce directly using neutrons as current carriers, mostly suitable way is to measure interactions between small elements of current. Such elements of current are areas of neutron reflection or scattering. The experimental setups can look as follows:




Reflecting crystals should have separate position or temperature control to vary characteristics of “conductors” for “currents” and see influence from these variations on one crystal to scattering on the other one.

This scheme looks different from general experimental methodology in physics of high energies. There is absence of direct search for correlations between spatially distant events occurring almost simultaneously. Einstein’s theory has direct ban for such correlations to be, as energy transfer for this must happen between and during very fast events with speed bigger than speed of light [13]. Collective behavior and long-range correlations take place in such processes as synchrotron (Chrenekov’s) radiation and are searched using wavelets for gluons in nuclear analogy of the effect [14]. Yet the direct experiment to test the long-range interaction propagating faster than light has its main obstacle in Einstein’s theoretical ban.

However, the experiments described above can have unexpected results to commonly accepted theory and nuclear phenomenology due to unknown properties of interaction between neutron currents. Nobody did experiments with ones. It can be as in the case of electric currents, which properties of interaction are not obviously seen from the properties of charges.

PSD on the drawings is position sensitive detector. Its purpose is to detect changes in reflection angles produced by energy transfer between pairs of neutrons in beam colliding with nuclei of distantly placed targets.

Qualitatively, character of expected changes in scattering angles can be understood from the following schematics of interaction via memory carriers moving between combined “address” and “data” parts.

Let impulse vectors P1 and P2 of neutrons n1 and n2 be parallel and equal by amplitude. Vectors P3 and P4 for n3 and n4 are zeros in the same system of reference. The difference between the two systems is that the lines between centers of the neutrons to be collided are not parallel. That produces different angles of scattering after collision for all the neutrons in the picture. Due to the interaction between the pairs during collisions via the memory carriers, scattering will be different in comparison with expected one in absence of interaction. The expected difference is designated on the picture by the angles between the solid and dashed lines of impulse vectors from n3 and n4. If we take the equality of impulses P1 and P2 as the same “address” characteristic in two memory systems, their “data” parts will come to interaction towards closer values during collisions. The indexes e and r of impulses on the picture correspond to expected and real impulses obtained during interaction between the pairs. Their vectors and amplitudes should have more close directions and values comparatively to the case of absence of interaction.

It also should be noted that the drawing assumes current of free neutrons scattered on other free neutrons. In the case of neutron scattering on crystal targets the role of “address” parts take particles in nuclei of targets. Exchange between the “data” would be seen on changes of scattering angles for neutrons in the beam. The possibility to detect it is dependent from intensity of the beam and yielded number of simultaneous collisions. If the intensity is too low, exchange between the targets and yielded alteration of angles in scattering can be neglected in observations. The concept of simultaneity between events is applied in the Netwon’s absolute reference system [13]. If the events were started simultaneously in this reference system, the energy transfer between the pairs of particles should be considered as dependent from speed of memory carriers, from duration of the events, from total energy difference between the two systems, and in general from ones’ similarities of “address” and “data” combinations.

The suggested experimental methodology can present different approach to experiments on quantum entanglement (QE) [15-18]. The experiments on neutron scattering performed by authors [16-17] provide evidence that cross-sections of Compton neutron scattering on protons in mixture of normal and heavy water are decreased due to QE up to 30% comparatively to normal water. From the point of view of the proposed in this paper idea, that is due to the energy transfer between systems with protons in different environments.

In other words, the environment of H2 atom is changed by the link with environment of D2. The associative exchange between ones changes normal conditions for interaction with incoming neutrons via motion of memory carriers between the systems with H2 and D2. As it is written in the [16], QE seems to effectuate destructive quantum interference between protons in the sub-femtosecond time scale.

Methodological point of view discussed in this paper gives idea to consider the phenomenon of QE like interaction between two elements of specific currents. It moves point of view from ideas of experimental setups and methodology given by quantum mechanics to simply idea analogous to routine electric measurements. Thus, this shift opens new ways to think about applications.

For example, the first one that looks mostly approached by experiments with quantum cryptography and communication on entangled photons [15] is application in area of communication. Main disadvantage of approach currently used is in impossibility to establish channel based on quantum entanglement without auxiliary classical one [15]. Result of measurement for photon on “transmitting” side has several probabilities for outcome.

From the point of view of communication between the photons via their memory carriers, QE takes place because of the very good geometrical and time conditions for carriers linked with one photon to influence on the other photon. Such conditions do not need to be reproduced in the case of macro currents. When considering macro-current of photons, several conditions for ones to interact should be present. The first one: photons in the currents have to be scattered and change ones’ conditions in this process. These changes should be controllable by changing, for example, temperature or other dynamic parameters of the “address” or “data” parts combined with photons and being in connection via motion of memory carriers with correspondent part on the other side in communication. Changing of conditions of memory systems of both sides should have some protocol of control and measurement allowing detection of correlations between the sides and thus transmission of information. This protocol has to be based purely on detection of differences between expected and observed dynamics of “data” parts of memory systems.

Speed of such a hypothetical communication channel won’t be dependent from speed of memory carriers. Main dependence for speed originates from volume of energy transfer per unit of time via the carriers to equalize difference between the conditions of memory systems.

The other valuable moment in building of such a communication system is easily expected dependence from distance of energy transfer via memory carriers analogous to Newton’s law of gravitation and to electric current interactions. To obtain such a selected anisotropy in space like in QE, one needs to find out and implement special conditions for communicating memory systems. One clue for this is possibility to use characteristic scales of information-scaling transformations [4] for some kind of dynamic processes on micro scale, for example. And build macro device for interaction with memory structures spreading out and back from source processes at micro scale.




General possibilities to build systems for medical imaging based on measurements of correlated dynamics of accelerations and decelerations of interacting objects were described. It is assumed that correlated character of the changes in dynamic parameters of these objects can be found due to motion of so-called reflecting structures or memory carriers via scale and logic transformations. The structures move from objects of interest for imaging out and back in space in some kind of regenerative cyclic motion. Fractal like interaction with objects in dynamics on different scales takes place during this motion. Measurements of this interaction are basis for creation of new methods and devices for medical imaging.




Author thanks his managing supervisor Professor Hans-Erik Nilsson for possibility to express these ideas in discussions of preliminary stage before definition of his PhD thesis theme.




[1] Boolos, G. and Jeffrey, R., Computability and Logic, 2nd ed., Cambridge: Cambridge University Press, 1980.

[2] Copeland, B.J. Super Turing-Machines. Complexity 4, pp. 30-32, 1998

[3] Siegelmann, H.T. & E.D. Sontag, Computational Power of Neural Networks.  Journal of Computer System Sciences 50(1), pp. 132-150, 1995

[4] A.Yu. Olevanov, About one of approaches for constructive modelling of functional systems and media, Advances in Synergetics, Vol. 5, p.p. 148-160, Belarus State University Press, Minsk, Belarus, 1995

[5] S. Berkovich, Cellular automation as a model of reality: search for new representations of physical and informational processes, Moscow, Moscow University Publisher, 1993, (in Russian)

[6] V.F. Shipicin, A.A Zhivoderov, A.G. Gorbich, Hypothesis of the space structure,, 2002

[7] M. Shalan and V. Mooney, "Hardware Support for Real-Time Embedded Multiprocessor System-on-a-Chip Memory Management," Proceedings of the Tenth International Symposium on Hardware/Software Codesign (CODES'02), pp. 79-84, May 2002

[8] J. Feder. Fractals, Plenum Press, New York and London, 1988

[9] L. Nottale & G. Schumacher, "Scale relativity, fractal space-time and gravitational structures", in "Fractals and Beyond : Complexities in the Sciences", Proceedings of 'Fractal 98' conference, (World Scientific), Ed. M.M. Novak, p. 149, 1998

[10] L. Nottale, "Scale relativity and quantization of extrasolar planetary systems", Astron. Astrophys. Lett., 315, L9, 1996

[11] H. Haken, Information and Self-organization. A Macroscopic Approach to Complex  Systems, Moscow, Mir Publisher, 1991, (in Russian)

[12] A.A. Kalnin, About possibility of distributed recording in non-equilibrium cooperative systems (structural reflexia), Pub. Of AS USSR, Ser. Biophysics, T. XXIX, v. 1., 1984

[13] W. Pauly, Theory of Relativity, Moscow, "Nauka", 1991, (in Russian)

[14] I.M. Dremin, Long-range particle correlations and wavelets, Uspekhi Fizicheskikh Nauk, vol. 170, issue 11, 2000 (Physics-Uspekhi, vol. 43, issue 11, 2000)

[15] Harry Buhrman, Quantum Computing and Communication Complexity, Center for Mathematics and Computer Science Quantum Computing and Advanced Systems Research, 2000

[16] C. Halcoussis, T. Abdul-Redah, H. Naumann, G. Monaco and C.A. Chatzidimitriou-Dreismann, First Observation of Molecular Vibrational Excitations of Water with Inelastic X-ray Scattering, ESRF Newsletter Oct. 2000

[17] C. A. Chatzidimitriou-Dreismann, T. Abdul Redah, R. M. F. Streffer, and J. Mayers,  Anomalous Deep Inelastic Neutron Scattering from Liquid H2O-D2O: Evidence of Nuclear Quantum Entanglement, Physical Review Letters, V.79, N. 15, pp. 2840-2842, 1997

[18] C. A. Chatzidimitriou-Dreismann, T. Abdul-Redah, and R. M. F. Streffer, Sub-femtosecond dynamics and dissociation of C–H bonds in solid polystyrene and liquid benzene, Journal of Chemical Physics, V. 116, N. 4, 2002


  Technical College - Bourgas,
All rights reserved, © March, 2000