[Magnetic field sensors for medical diagnosis]

Abstract

Magnetic field sensors are considered promising in medical diagnostics. They are grouped into two types: typeI – operating at room temperature; type II – requiring cryogenic cooling. It is noted that among type I sensors,laser�pumped atomic magnetometers are suitable, and among type II – SQUIDs (Superconducting QuantumInterference Devices). Also particularly promising are combined sensors consisting of a superconducting film witha nanostructured active band serving as a magnetic field hub and a structure with magnetoresistance as a mag�netically sensitive element.

Full text

Highly sensitive magnetic field sensors (MFS) are
now used in many fields of human activity, e.g. electron
ic compasses, archaeological research, space devices,
and in medical diagnostic systems [1]. In the latter case,
high sensitivity is required, as many organisms generate
weak but measurable magnetic field B10 nT. These
biomagnetic signals could be static fields caused by direct
currents or small magnetic particles in tissues and oscil
lating electrical activity. MFS are used in many areas of
medicine, such as clinical diagnosis [2], gastroenterolo
gy [3], recognition of biomolecules [4], magnetocardio
graphy (MCG) [5], magnetoencephalography (MEG)
[6], etc. Two moststudied sources of biomagnetic signals
in the human body are the brain and the heart. MEG
and MCGsignals are caused by electrical currents flow
ing in nerve cells of the brain and the heart muscles,
respectively. Timing accuracy of these methods is very
high (in the millisecond range), and the recording of sig
nals in several spatial positions usually allows locating
their sources. Biomagnetic fields generated by the brain
and heart mainly lie in the range Bfrom 10 pT to
1 fT.
Modern medicine practice employs numerous pas
sive (e.g. bone tissue substitutes) and active implants (cir
culatory assist devices, artificial hearts, various stimu
lants, etc.). Noninvasive inspection of their functional
characteristics, resources and other properties can be per
formed by highly sensitive magnetic field sensors.
This article discusses the most promising lowfre
quency (1 kHz) MFS in the field of medical diagnos
tics. MFS are systematized into two types: operating at
room temperature – type I, and requiring cryogenic
cooling – II type. It briefly describes physical principles
of operation and possibilities of using the devices, as well
as provides some numerical and experimental data: reso
lution of magnetic field δB, resolution of magnetic flux
δφ, energy resolution ε, dynamic measurement range Dr,
etc.
Type I
1.1. Magnetoresistance – the property of materials
that change their resistance when exposed to an external
magnetic field. The most significant magnetoresistive
effects are anisotropic magnetoresistance, giant magne
toresistance (GMR), extraordinary magnetoresistance
(EMR), and tunneling magnetoresistance. Numerous
commercial MFS are based on these effects. Magneto
resistive MFS can record magnetic fields in range from
1 mT to 1 nT. For example, they are used in gastroen
terology when magnetoresistive MFS systems and mag
netic field concentrators (MFC) track magnetic markers
in the esophagus [3].
GMR effectbased sensors are used for identification
of biomolecules [4]. Figure 1 shows a magnetoresistive
MFS for studying DNA hybridization.
Biomedical Engineering, Vol. 48, No. 6, March, 2015, pp. 305309. Translated from Meditsinskaya Tekhnika, Vol. 48, No. 6, Nov.Dec., 2014, pp. 1923.
Original article submitted October 17, 2014.
305 00063398/15/48060305 ©2015 Springer Science+Business Media New York
1 National Research University of Electronic Technology MIET,
Zelenograd, Moscow, Russia; Email: leo852@inbox.ru
2Base Technologies Co, Zelenograd, Moscow, Russia.
* To whom correspondence should be addressed.
Magnetic Field Sensors in Medical Diagnostics
L. P. Ichkitidze1*, N. A. Bazaev1, D. V. Telyshev1, R. Y. Preobrazhensky1, and M. L. Gavrushina2
Magnetic field sensors are considered promising in medical diagnostics. They are grouped into two types: type
I – operating at room temperature; type II – requiring cryogenic cooling. It is noted that among type I sensors,
laserpumped atomic magnetometers are suitable, and among type II – SQUIDs (Superconducting Quantum
Interference Devices). Also particularly promising are combined sensors consisting of a superconducting film with
a nanostructured active band serving as a magnetic field hub and a structure with magnetoresistance as a mag
netically sensitive element.
DOI 10.1007/s10527-015-9475-0

306 Ichkitidze et al.
Magnetoresistive sensors are also used in clinical
diagnostics. Sensors track the position of functionalized
magnetic microgranules [6].
An EMRbased sensor is demonstrated whose main
noise sources are thermal Johnson noises only [7]. These
sensors are of size less than 50 ×50 μm2, their resolution
is δφ ~ 10–6 φ0and δl< 50 μm, where φ0≈2⋅10–15 Wb –
magnetic flux quantum. Selfnoise can be reduced to the
level of Bn1 pT/Hz1/2, which would allow using these
sensors as MFS in magnetocardiography.
1.2. In the beginning of this century, production of
laserpumped atomic magnetometers (LPAM) began; the
volume of their working cell was 1 cm3. LPAM with cell
volume of 4 ×19 ×40 mm3containing potassium atoms
Focusing
Magnetic markers
Target DNA
DNAprobes
Passivation
Substrate
Structure of magnetic
marker transport Spintronic
transducer
Hybridization
of DNAprobe
and target
Magnetic
moment
of the marker
Fig. 1. Example of chemical identification of biomolecules using magnetoresistive MFS [4].
Fig. 2. LPAM sensor head location near human skull [4].
Sensor
cell
Sensor head
< 4 mm
Fig. 3. MEG Elekta Neuromag. The helmet contains a SQUID
array; the head of the patient is located inside the helmet [16].
Number of sensors
in an array: 306
Sensor price:
Cooling liquid consumption:
12 liters/day
Distance between sensor center
and skin of head
Price of MEG Elekta
Neuromag:
SQUID array
in MEG Elekta Neuromag
MEG Elekta Neuromag

Magnetic Field Sensors in Medical Diagnostics 307
and using circularly polarized laser diodes have resolution
of the order of magnetic selfnoise Bn~ 7 fT/Hz1/2, ε≈
7·10–29 J/Hz for “magnetometer” configuration and Bn~
0.54 fT/Hz1/2, ε≈7·10–31 J/Hz for “gradiometer” config
uration [8]. The spatial precision δlis ~1.5 cm for both
configurations.
LPAM with free spinexchange relaxation atoms are
the most sensitive [9]. Bn~ 1 fT/Hz1/2 with the possibility
of reducing selfnoise to the fundamental level of
0.01 fT/Hz1/2 was shown. Atomic magnetometers for bio
medical research are available in two configurations. The
first involves placing the person in a shielded environ
ment, as described in [10]. The second configuration uses
miniature integrated atomic magnetometers in gradiome
ter mode [11]. Systems of this type do not require shield
ing environment and can realize high spatial resolution
due to the small size of the sensor. An example of location
of the LPAMsensor for MEG recording is shown in
Fig. 2.
1.3. Values of δB~ 100 pT, Dr~ 70 dB are reached in
fluxgate MFS with magnetosensitive element (MSE)
made of singledomain monocrystalline Permalloy. These
sensors directly measure absolute values of the projection
of the magnetic field, and these values allow recording
accumulation of ferromagnetic particles in biomedical
objects [12]. Disadvantages of these MFS include large
size and limited dynamic range.
Type II
2.1. SQUID consists of basic MSE elements –
Josephson junctions – two weakly coupled pieces of
superconductor, between which superconducting tunnel
current flows. In SQUIDs based on hightemperature
superconducting (HTS) materials of the YBaCuO sys
tem with working temperature Tw~ 77K, the resolutions
δφ ~ 10–510–6 φ0, δB~ 10–1410–13 T, and ε~ 10–27 J/Hz
were achieved, which is several times inferior to those for
SQUIDs based on lowtemperature superconducting
materials (LTS), e.g. in Nbbased SQUIDs with Tw~ 4K:
δφ ~ 10–610–7 φ0, δB~ 10–15 T, Dr140 dB. For them,
the estimated εreaches the record low value of
~10–32 J/Hz, and experimental ε~ 10–30 J/Hz [13].
Commercial SQUIDs have different characteristics:
the noise, i.e. resolution of the magnetic field and mag
netic flux (Bnand δφn), size (surface area 30 mm2, plate
thickness 4 mm), operating temperatures, number of
projections of recorded magnetic field, materials, and
price range (Table 1).
Usually HTS SQUIDs are several times more expen
sive than LTS SQUIDs. This price difference significantly
affects the systems in which the number of SQUIDs can
reach several hundreds. Figure 3 shows the typical commer
cial MEG Elekta Neuromag (Elekta, Finland) [16]. The
helmet contains a SQUID array, and the greater the num
ber of SQUIDs, the higher the resolution of the device.
2.1.1. Electrocardiographic diagnostic method is
widely used in medical practice, but the magnetocardio
graphic method is more informative and allows for early
detection of pathology (Fig. 4). The modern MCG in
most cases are equipped with several dozens of SQUIDs,
and their resolution of δB~ 1 pT is considered accept
able. The typical MCG Mesuron Avalon90 of Mesuron
LLC Company includes 90 SQUIDs [17]. Magnetic sig
nal processing techniques and their connections with var
ious pathologies are being developed. The price range for
the MCG is much lower than for the MEG, so they are
installed in many central hospitals.
2.1.2. The magnetic microscope contains SQUID as
the MSE [18]. When the MSE gets closer to the object of
study, the microscope allows recording magnetic particles
of micron and submicron size depending on their depth
relative to the surface. To control large integrated circuits,
magnetic microscopes are used that record microcurrents
and design flaws. For example, Neocera (USA) produces
the MAGMA30 magnetic microscope with HTS
SQUID [19]. After structural changes and increase in
TABLE 1. Some Parameters of Commercial SQUIDs
Name
Parameter
Bn, fT/Hz1/2
δφn, μφ0/Hz1/2
Dimensions, mm
Tw, K
Material
Price, $
Tristan HTM8 [14]
50
8
8 ×8
77
YBCO
3500
Cryo GA1165 [15]
2
2
6 ×6
06
NbAlAlOxNb
1645
Cryo M1000 [15]
100
10
9 ×9
77
YBCO
2510
Tristan LSQ/20 [14]
< 1
1
7.2 ×7.2
07
NbAlAlOxNb
3500

308 Ichkitidze et al.
sensitivity such microscopes might be used to control
directional drug delivery using magnetic particles in the
patient’s organs.
2.2. The magnetomodulation magnetometer
(MMM) is similar to a fluxgate MFS, but a core made of
ceramic HTS material with the property of “Josephson
medium” is used as the MSE [20]. MMM at Tw~ 77 K is
characterized by acceptable absolute magnetosensitivity
(105V/T), values δB10–13 T, and δφ  10–4 φ0, but are
inferior to SQUIDs. However, they are much cheaper
than SQUID and can directly measure the absolute value
of the magnetic field, unlike SQUIDs.
A layer of the Josephson medium formed in film
MFC in the form of superconducting rings acquires
parameters that rival MMM parameters and have very
small mass and dimensions [2125]. Further improve
ment in layer materials technology, e.g. Y123 or Bi2223
systems with properties of “Josephson medium”, will bring parameters of MFS based on them closer to the
parameters of SQUIDs.
2.3. Combined MFS based on “superconductor
(MFC)/nonsuperconductor (MSE)” structures are being
actively developed, in particular HTS/GMR, LTS/GMR,
etc. (Fig. 5). In combination “film LTS/GMR” (niobium
film/permalloy film), the value of δB~ 1 fT at Tw~ 4K is
achieved, which is much better compared to the resolution
of HTS SQUIDs (δB5 fT at Tw~ 4K) [2628].
2.4. The authors of [2933] propose nanostructuring
of active superconducting film bands of MFC in the com
bined MFS. When splitting (nanostructuring) an active
band of MFC in the form of parallel superconducting
branches and slots with width in the range of 201400 nm
(Fig. 6), parameters of the combined MFS are signifi
cantly improved. For example, δBdecreases by more than
an order of magnitude, and Drincreases severalfold
compared to MFS with solid (notnanostructured) MFC.
It can have parameters at the level of HTS SQUIDs (ε~
10–27 J/Hz) and be close to the LTS SQUID parameters
(Bn~ 1 fT/Hz1/2). These parameters are related to the
combined MFS having MFC with critical current Jc
106A/cm2. However, in some HTS materials, e.g. the Bi
2223 system, to enhance Jcin addition to traditional syn
thesis methods additional processing is required, such as
inclusion of nano and microsized additions [34],
shockwave plasma effect, etc. [35].
In SQUIDs, high magnetosensitivity is realized by
MFC with a diameter of ~10 mm. In nanostructured
combined MFS, a MFC with diameter of ~12 mm is suf
ficient. This means that such MFS can be positioned
more compactly, e.g. on the MEG helmet (see Fig. 2),
thus increasing its spatial resolution. The price of nano
structured combined MFS is expected to be several times
lower than that of HTS SQUIDs (ca. $500).
Fig. 5. Typical structure of combined MFS. Lower part, GMR
MSE with measurement probes (two current and two potential
probes). Upper part, MFC of film superconductor; the narrow
part covers the MSE (the arrow shows the direction of the meas
ured magnetic field [17]).
Fig. 6. Diagram of MFS and its elements: 1) superconducting ring
of MFC; 2) dielectric substrate; 3) active band of MFC (enlarged;
the proportions are not maintained); 4) MSE; 5) insulating film;
6) branches of the active band; 7) slits of the active band [29].
Fig. 4. Signals recorded by electrocardiograph (upper plot) and
magnetocardiograph (lower plot) [17].
Time, s
MCG, pT
ECG, mV

Magnetic Field Sensors in Medical Diagnostics 309
Conclusion
This article describes some of magnetic field sensors
with possibility of their use in magnetic systems for med
ical diagnostics. From MFS type I (not requiring cryo
genic cooling) in magnetocardiography, probably most
suitable are laserpumped magnetometers. They have
rather large dimensions and can be incorporated in mag
netoencephalographs with numerous LPAM sensors sim
ilar to a SQUIDbased one, but it is a difficult task [16].
Among universal MFS type II (requiring cryogenic cool
ing) are commercial SQUIDcontaining magnet systems.
These systems have very wide dynamic measurement
range (140 dB), so their potential is very high.
For example, a SQUIDcontaining magnetoen
cephalograph is the only hardware that enables noninva
sive recording of brain neuronal activity and diagnosing
epilepsy. Indeed, the spatial functional and resolution
capabilities of these systems are at the same level with
other modern methods of diagnosis (e.g. magnetic reso
nance imaging, positron emission tomography), or even
surpass them. However, in practice most magnetic sys
tems (magnetocardiography, magnetoencephalography,
etc.) contain multiple SQUIDs, which are very expensive
and difficult to obtain. The high price of these systems is
mainly due to the high cost of SQUIDs (Table 1).
A new combined magnetic field sensor based on the
phenomena of superconductivity and spintronics was
reviewed. These sensors with nanostructured elements
have characteristics close to those of SQUIDs, while being
several times smaller and less expensive. Consequently,
new magnetic systems combined with nanostructured
magnetic field sensors will significantly increase the avail
ability of noninvasive magnetic diagnostics.
New methods of treatment and diagnostics, new bio
compatible materials, in particular directional drug deliv
ery using magnetic nanoparticles, nanomaterials with fer
romagnetic or supermagnetic particles, carbon nano
tubes, and others are now being developed. Undoubtedly,
regular monitoring of active implanted devices is
required, for example, of an artificial heart, circulatory
assist devices, various stimulants, etc. Suitable for this
purpose are magnetic systems with highly sensitive mag
netic field sensors that enable safe, noninvasive diagnosis
and monitoring. The problem of high demand might be
solved with the use of the magnetic field sensors analyzed
in this article.
The authors thank Professors V. M. Podgaetsky and
S. V. Selishchev for useful advice.
This study was supported by a grant from the Russian
Science Foundation (project No. 143900044).
REFERENCES
1. S. Tumanski, Handbook of Magnetic Measurements, CRC Press
(2011), pp. 335379.
2. R. Ramli, F. Haryanto, K. Khairurrijal, and M. Djamal, InTech,
149164 (2011); http://www.intechopen.com/books/biosensors
forhealthenvironmentandbiosecurity/gmrbiosensorsforclin
icaldiagnostics.
3. F. C. Paixao, Ann. Int. Conf. of the IEEE Engineering in Medicine
and Biology Society, Conference Proceedings, 2006, Vol. 2007, pp.
29482951.
4. P. P. Freitas et al., J. Physics: Condensed Matter, 19, 165221
(2007).
5. M. A. Primin, I. V. Nedaivoda, Yu. V. Maslennikov, and Yu. V.
Gulyaev, Radiotekhnika Elektronika, 55, 12501269 (2010).
6. J. C. Rife, M. M. Miller, P. E. Sheehan, C. R. Tamanaha, M.
Tondra, and L. J. Whitman, Sensors and Actuators A: Physical,
107, 209218 (2003).
7. M. Hamalainen, R. Hari, R. J. Ilmoniemi, J. Knuutila, and O. V.
Lounasmaa, Rev. Modern Phys., 65, 413497 (1993).
8. I. K. Kominis, T. W. Kornack, J. C. Allred, et al., Nature, 422,
596599 (2003).
9. J. P. Wikswo, Physics Today, 57, 1517 (2004).
10. D. Budker and M. Romalis, Nature Physics, 3, 227234 (2007).
11. T. H. Sander et al., Biomed. Optics Express, 3, 981990 (2012).
12. S. K. Prishchepov and K. I. Vlaskin, Nauch. Priborostr., 21, 151
155 (2011).
13. D. Robbes, Sensors and Actuators A: Physical, 129, 8693 (2006).
14. www.tristantech.com.
15. www.starcryo.com.
16. www.elekta.com.
17. M. PannetierLecoeur, SuperconductingMagnetoresistive
Sensor: Reaching the Femtotesla at 77K: PhD Thesis, Universite
Pierre et Marie CurieParis VI (2010), pp. 3234.
18. http://perst.issp.ras.ru/Control/Inform/tem/HiTech/squid.htm.
19. http://www.neocera.com/.
20. Y. E. Grigorashvili, L. P. Ichkitidze, and N. N. Volik, Physica C:
Superconductivity, 435, 140143 (2006).
21. L. P. Ichkitidze, Weak Magnetic Field Sensor Based on a
Superconducting Film, RF Patent No. 2258275.
22. L. P. Ichkitidze, Superconducting Weak Magnetic Field Film
Sensor with Magnetic Flux Transformer, RF Patent No. 22899870.
23. L. P. Ichkitidze, Bull. Russ. Acad. Sci.: Physics, 71, 11801182
(2007).
24. L. P. Ichkitidze, Bull. Russ. Acad. Sci.: Physics, 71, 11451147
(2007).
25. L. P. Ichkitidze, Physica C: Superconductivity, 435, 136139
(2006).
26. M. PannetierLecoeur et al., Science, 304, 16481650 (2004).
27. M. PannetierLecoeur et al., J. Magnetism Magnetic Materials,
322, 16471650 (2010).
28. M. PannetierLecoeur et al., Appl. Phys. Lett., 98, 153705 (2011).
29. L. Ichkitidze and A. Mironyuk, Physica C: Superconductivity, 472,
5759 (2012).
30. L. P. Ichkitidze and A. N. Myronyuk, Nano Mikrosist. Tekhn., 1,
4750 (2012).
31. L. P. Ichkitidze and A. N. Myronyuk, Superconducting Film Flux
Transformer, RF Patent No. 2455732.
32. L. P. Ichkitidze and A. N. Myronyuk, J. Physics: Conf. Ser., 400,
022032 (2012).
33. L. P. Ichkitidze, AIP Advances, 3, 062125 (2013).
34. B. P. Mikhailov, L. P. Ichkitidze, et al., Inorg. Mater., 39, 749754
(2003).
35. B. P. Mikhailov, V. Ya. Nikulin, P. V. Silin, et al., Perspekt. Mater.,
10, 7075 (2013).