CN105433945A - Bone mineral density detection equipment and detection method - Google Patents

Abstract

The invention discloses a bone mineral density detection equipment and detection method based on a magnetic resonance T2 relaxation time spectrum. A bone mineral density is detected based on the magnetic resonance T2 relaxation time spectrum. The detection equipment comprises a computer control terminal, a one-board magnetic resonance controller unit, a signal amplification and switch control unit and a magnet unit. According to the bone mineral density detection equipment and detection method, water molecule magnetic resonance transverse relaxation deamplification is measured firstly, the corresponding T2 relaxation time spectrum is obtained through a mathematical inversion method, structural characteristics of pores among bone trabeculae are determined by analyzing all peak values on the T2 relaxation time spectrum, and biological bone mineral density data is obtained through a regression analysis data processing method finally. The bone mineral density detection equipment and detection method can be applied to identification of osteoporosis, the functions of rapidly and nondestructively analyzing and measuring the bone mineral density, evaluating the bone quality and the like are achieved, and the bone mineral density detection equipment and detection method have the advantages of being safe and harmless to organisms, simple in equipment structure, small in size, light in weight, high in measurement precision, easy and convenient to operate, high in repeatability of measurement method and the like.

Description

A kind of Bone mineral density equipment and detection method

Technical field

The invention belongs to magnetic resonance arts, relate to biological bone Density Detection, particularly relate to a kind of based on magnetic resonance T ₂the Bone mineral density equipment of relaxation time spectrum and detection method.

Background technology

Skeleton is substantially one of the most of paramount importance physiological structure in human body.With advancing age, mostly there is the situation that bone amount runs off in people; If, will osteoporosis be there is in bone amount loss especially severe.Osteoporosis not only can increase the risk of fracture, and also can cause multiple complications, human health in serious threat.

In general, the skeletal structure of patients with osteoporosis does not especially significantly change, and change is the quality (Quality) of skeleton more significantly.In order to quality (Quality) quantification by skeleton, generally choose bone mineral density BMD (BoneMineralDensity) as the parameter weighing bone mass.The main method of existing measurement BMD data has Single Photon Absorption algoscopy, dual-energy x-ray absorption measurement method, quantitive CT method and ultrasonic measuring method.

Single Photon Absorption algoscopy, because accuracy is low, poor repeatability, is eliminated gradually.In most cases, dual-energy x-ray absorption measurement method is adopted to measure bone density BMD data.Dual-energy x-ray absorption measurement method, also known as DEXA (DualEnergyX-rayAbsorptiometry), its ultimate principle is the minimizing of number of electrons and energy in the x-ray light beam according to the two bundle different-energies through tissue, carrys out quantitative analysis through the thickness organized and density.DEXA is existing main flow diagnostic method, technology maturation and its bone density data measured can reflect the quality of bone mass to a certain extent.But the product of different manufacturers, different operating person go to measure, the bone density data property of there are differences.More seriously there is the X-radiation of doses in the method, has certain damage for human body, should not frequently measure, and comparing across the time of bone density data is restricted.

The method of quantitive CT, also known as QCT (QuantityComputedTomography), can be improved realization, but need newly-increased special software on existing computed tomographic scanner.Different from DEXA, that its is measured is the bone density data (g/cm of volume ³), be real density data.But the radiation dose of QCT is higher than DEXA, and equipment is huge, somewhat expensive, is difficult to promote.

What in fact ultrasonic method for measuring measured is not bone density data, but according to the acoustic velocity value SOS (SpeedofSound of sound wave measured, unit: m/s) and broadband ultrasonic wave pad value BUA (BroadbandUltrasoundAttenuation, unit: dB/MHz) through calculating quantitative ultrasound ripple bone strength QUS (QuantityUltrasoundStiffness) data.Compared with DEXA measuring method, there is not ionizing radiation in QUS measuring method, harmless to health.But be through Experimental Comparison, compared with the BMD data that QUS data and DEXA record, dependency is lower, and accuracy cannot ensure, only has certain reference value, be difficult to clinical measurement diagnosis.

In recent years along with the development of mr imaging technique, due to the attention that it has complete noinvasive, degree of accuracy high is more and more subject to medical domain.The application development of mr imaging technique in orthopaedics is especially rapid, is mainly used in discriminating and the diagnosis of the aspects such as skeleton trauma, bone tumor and soft tissue lesions.But it is longer that existing MR imaging apparatus and method gather the echo time, can not collect the osseous tissue signal under the extremely short echo time, in the less detection be applied to bone density.

Summary of the invention

In order to overcome above-mentioned the deficiencies in the prior art, the invention provides a kind of based on magnetic resonance T ₂the Bone mineral density equipment of relaxation time spectrum and detection method, the equipment echo time is shortened to ultrashort echo time UTE (Ultra-shortEchoTime by design magnetic resonance hardware device, scope and echo time are shorter than 100 microseconds), can T be collected ₂the osseous tissue signal that relaxation time is extremely short, then obtain T through mathematical inversion ₂the data such as relaxation time spectrum, further by T ₂relaxation time spectrum combines with the bone density data of quantitative assay, finally reaches the object of correct measurement bone density.

Technical scheme provided by the invention is:

A kind of Bone mineral density equipment, described checkout equipment is based on magnetic resonance T ₂relaxation time spectrum detects bone density, comprise computer control terminal 1, veneer magnetic resonance controller unit 2, signal amplification and switch control unit 3 and magnet unit 4, described computer control terminal 1 connecting single board magnetic resonance controller unit 2, veneer magnetic resonance controller unit 2 amplifies with signal and switch control unit 3 is connected, and signal amplifies and switch control unit 3 is connected with magnet unit 4; Described computer control terminal 1 for controlling the workflow of complete equipment system, receive simultaneously and process described veneer magnetic resonance controller unit 2 gather signal, obtain T2 relaxation time spectrum; Described veneer magnetic resonance controller unit 2 is for generation of resonance and collect original magnetic resonance signal; Described signal amplification and switch control unit 3 are for amplifying the signal collected and the switch controlling radio-frequency coil; Described magnet unit 4 is for keeping the constant temperature of surveyed area environment and building region magnetostatic field.

In above-mentioned Bone mineral density equipment, further, described veneer magnetic resonance controller unit 2 comprises frequency synthesis and pumping signal radiating portion 5 and digital detection and signal process part 6; Described signal amplifies and switch control unit 3 comprises low-noise preamplifier 7, radio frequency amplifier 8, radio-frequency (RF) switch 9 and Q-switch switch 10; Described frequency synthesis and signal are launched driver unit 5 and are produced magnetic resonance excitation signal sequence, and transfer to radio-frequency coil through radio frequency amplifier 8, now radio-frequency (RF) switch 9 is signal foment; Radio-frequency coil through the co-controlling of radio-frequency (RF) switch 9 and Q-switch switch 10, switches to signals collecting state, collects original magnetic resonance signal after signal sequence excitation sample; Described original magnetic resonance signal is after low-noise preamplifier 7 amplifies, computer control terminal 1 is transmitted back after carrying out pretreatment by digital detection and signal process part 6, process described signal by computer control terminal 1 and obtain T2 relaxation time spectrum, as detecting the final data result obtained.

In above-mentioned Bone mineral density equipment, further, electronic chip integrated for multiple height is integrated on one piece of circuit board by described veneer magnetic resonance controller unit 2, for realizing magnetic resonance measurement function; In the embodiment of the present invention, described circuit board is of a size of " 220mm × 100mm ", greatly simplifies the structure of magnetic resonance equipment, makes equipment have portability and mobility; The integrated electronic chip of multiple height mainly comprises the power supply chip (model is TPS74401) of on-site programmable gate array FPGA (Field-ProgrammableGateArray) chip EP3C55F484C8, FPGA, the configuring chip EPCS16 of FPGA and digital-to-analogue conversion (DA) chip AD9742 that Digital Signal Processing (DSP) chip LF2407, ALTERA company that Texas Instrument (TI) produces produces.

In above-mentioned Bone mineral density equipment, further, described veneer magnetic resonance controller can shorten the equipment echo time to ultrashort echo time UTE (Ultra-shortEchoTime, scope and echo time are shorter than 100 microseconds), can collect T ₂the osseous tissue signal that relaxation time is extremely short.

In above-mentioned Bone mineral density equipment, further, Q-switch switch 10 is with coil quality factor change-over switch (Q-switch), coil can be made to receive rapid translating between two kinds of mode of operations in pumping signal transmitting and magnetic resonance signal by this Q-switch switch, very big shortening acquisition of magnetic resonance signals time, thus the magnetic resonance signal collecting skeletal tissue extremely short die-away time itself.

In above-mentioned Bone mineral density equipment, further, described magnet unit 4 comprises attemperating unit 11, permanent magnet 12 and transmitting and receiving coil 13; Described permanent magnet 12 provides dominant frequency to be the magnetic field of 10.71MHz; Described permanent magnet 12 is a kind of open moveable magnets, sample bore reaches 120mm, the osseous tissue not being only applicable to various toy is measured, and is applicable to the larger situation of the measuring object volumes such as human body forearm osseous tissue, meets the measurement demand under multiple condition.

The present invention also provides a kind of and utilizes above-mentioned Bone mineral density equipment to carry out the method for Bone mineral density, first measures hydrone magnetic resonance transverse relaxation deamplification in skeleton, then obtains corresponding T by mathematical inversion method ₂relaxation time spectrum, then by analyzing T ₂the architectural feature of hole between each peak value determination bone trabecula on relaxation time spectrum, obtains biological bone density data finally by regression analysis data processing method; The method comprises the steps:

1) the magnetic field center position that one group of measurement sample is placed in described Bone mineral density equipment is measured, often organized the original magnetic resonance signal measuring sample, as often organizing the spin echo data measuring sample;

In the embodiment of the present invention, carry out measuring the spin echo signal data obtained, to eliminate the impact of main field inhomogeneities especially by employing self-rotary echo-pulse series;

2) spin echo data recorded is carried out mathematical inversion process, obtain T ₂relaxation time spectrum;

3) by T ₂the linear corresponding relation existed between relaxation time and biological bone density is represented by graticule formula (formula 4):

BMD=KT ₂+ C (formula 4)

In formula 4, BMD is bone density (BoneMineralDensity, g/cm ²); T ₂for the T of sample cortical bone ₂relaxation time data, unit is μ s; K is constant, and unit is g/ (cm ²﹒ μ s); C is constant, and unit is g/cm ²; K value and C value are along with biological species is different, sex is different, age different value is different.By the T of different sample ₂relaxation spectrum and dual-energy x-ray borne densitometers measure the corresponding bone density data obtained and carry out regression analysis, obtain the K value in formula 4 and C value;

4) calculate detection bone density according to formula 4, complete Bone mineral density.

Above-mentioned Bone mineral density method, further, step 2) spin echo (CPMG) data recorded are carried out mathematical inversion process, obtain T ₂relaxation time spectrum; Described mathematical inversion adopt document " multi-exponential Inversion of NMR relaxation signal " (Wang Weimin, Li Pei, Ye Chaohui, Chinese science A, 2001:, 31 (6): 730-736) the conversion inversion algorithm recorded.

In the present invention, the conversion inversion algorithm of employing is following detailed process:

Owing to there is the hole of different size in skeleton, and the T that different large fine pore is corresponding ₂relaxation time length is different, therefore, measures the superposition that spin echo signal y (t) obtained is a series of single hole spin echo signal in the present invention, as shown in Equation 11:

$y\left(t\right)={\Σ}_i{f}_i\·\exp (-\frac{t}{T_{2i}})+\mathrm{\ϵ}\left(t\right),t=n\·\mathrm{\τ}$ (formula 11)

In formula 11, f _ibe the share of the i-th class hole shared by total pore space; T _2ibe the T of the i-th class hole ₂relaxation time; τ is the echo sounding time; ε (t) is random noise sequences.By solving the T of all kinds of hole ₂relaxation time T _2iand the share f that all kinds of hole is shared in total pore space _i, corresponding T can be obtained ₂relaxation time spectrum.

In order to solve aforesaid equation 11, first do object function as shown in the formula 12:

${\mathrm{\χ}}^2={\Σ}_{i=1}^n{\[y_i-{\Σ}_{j=1}^m(f_j\·m_{ij})\]}^2+\mathrm{\λ}\·{\Σ}_{j=1}^m{f}_j=\left|\right|y-Mf|{|}^2+\mathrm{\λ}\·|\left|f\right|{|}^2$ (formula 12)

In formula 12, M=[m _ij]=[exp (-t _i/ T _2j)], m _ij=exp (-t _i/ T _2j); λ is smoothing factor; F=(f ₁, f ₂..., f _m) ^tfor amplitude; χ ²for corresponding object function; y _ifor the n chosen ties up the magnetic resonance signal that in time domain, the i-th dimension is corresponding; Y is the y (t) in formula 11, the spin echo signal namely recorded; f _jcorresponding hole share shared in total pore space is tieed up for m ties up jth in T2 spatial domain.To a kth component of amplitude f ask extreme value (k=1,2 ..., m) and make it equal 0 can to obtain formula 13:

(M ^tm) f+ λ I _{m × m}f=M ^ty (formula 13)

In formula 13, I _{m × m}for m × m rank unit matrix; Linear change is done to formula 13, makes f=M ^tc, can make the T that m ties up ₂the solution of spatial domain transforms to n dimension time domain space and separates, and brings this conversion into formula 13, and selects suitable λ value, get final product the least square solution that through type 14 obtains equation 11:

$\hat{f}=M^T\·{(M\·M^T+{\mathrm{\λI}}_{n\×n})}^{-1}\·y$ (formula 14)

In formula 14, for n ties up the solution of range value; The number of n is cloth and counts, a corresponding n T ₂relaxation time values.

In embodiments of the present invention, adopt above-mentioned conversion inversion algorithm, choosing applicable smoothing factor is λ=0.05, inverse time scope 1 ~ 10000 μ s, and cloth counts 100, and the log cloth point mode that it is the end with 10 that points distributing method adopts, carries out mathematical inversion process, obtain T ₂relaxation time spectrum.

Above-mentioned Bone mineral density method, further, step 3) described graticule formula carries out linear fit particular by method of least square and obtains.

Above-mentioned Bone mineral density method is method for fast measuring, measures required time and is less than 0.5 minute.

Compared with prior art, the invention has the beneficial effects as follows:

The present invention, by design magnetic resonance hardware device, shortens the equipment echo time to ultrashort echo time UTE (Ultra-shortEchoTime, scope and echo time are shorter than 100 microseconds), can collect T ₂the osseous tissue signal that relaxation time is extremely short, then obtain T through suitable mathematical inversion ₂the data such as relaxation time spectrum, further by T ₂relaxation time spectrum combines with the bone density data of quantitative assay, finally reaches the object of correct measurement bone density.

The invention provides checkout equipment and first detection method measures hydrone magnetic resonance transverse relaxation deamplification in skeleton, then obtain corresponding T by mathematical inversion method ₂relaxation time spectrum, then by analyzing T ₂the architectural feature of hole between each peak value determination bone trabecula on relaxation time spectrum, obtains biological bone density data finally by regression analysis data processing method; The present invention can be applicable to the discriminating of osteoporosis, there is quick, nondestructive analysis mensuration bone density, evaluate the functions such as bone mass, compared with the equipment of the use X-ray measurement method of routine, have organism completely harmless, device structure is simple, volume is little, lightweight, certainty of measurement is high, easy and simple to handle, the repeatable advantage such as by force of measuring method.

Accompanying drawing explanation

Fig. 1 is that magnetic resonance spins echo (CPMG) measures sequence timing diagram.

Fig. 2 is the structured flowchart of the ultrashort echo biological skeleton magnetic resonance relaxation spectrum measurement device in the embodiment of the present invention;

Wherein, 1-computer control terminal 1; 2-veneer magnetic resonance controller unit; 3-signal amplifies and switch control unit; 4-magnet unit; ; 5-frequency synthesis and pumping signal radiating portion; 6-digital detection and signal process part; 7-low-noise preamplifier; 8-radio frequency amplifier; 9-radio-frequency (RF) switch controls; 10-Q-switch switch; 11-attemperating unit; 12-permanent magnet; 13-transmitting and receiving coil.

Fig. 3 is magnetic resonance T provided by the invention ₂spectrum detects the FB(flow block) of bone density method.

Fig. 4 is 12 months old rats femur typical sample T in the embodiment of the present invention ₂relaxation time spectrum.

Fig. 5 is 12 months old rats femur T in the embodiment of the present invention ₂relaxation time-mean bone density least square fitting straight line.

Detailed description of the invention

Below in conjunction with accompanying drawing, further describe the present invention by embodiment, but the scope do not limited the present invention in any way.

The invention provides a kind of equipment for bone density measurement and detection method, for promoting the accuracy of bone density DATA REASONING, repeatability and safety, make detection data can reflect the quality of bone mass more accurately, and reduce the radiation hazard to medical worker and sufferer.

The present invention, by design magnetic resonance hardware device, shortens the equipment echo time to ultrashort echo time UTE (Ultra-shortEchoTime, scope and echo time are shorter than 100 microseconds), can collect T ₂the osseous tissue signal that relaxation time is extremely short, then obtain T through suitable mathematical inversion ₂the data such as relaxation time spectrum, further by T ₂relaxation time spectrum combines with the bone density data of quantitative assay, finally reaches the object of correct measurement bone density.

T ₂relaxation time, when being hydrogen atom generation spin-spin relaxation, Mxy component (i.e. the component in magnetization vector x, y direction in three-dimensional system of coordinate) decayed to the time needed for 0 by maximum also known as T2.The cross stream component M of magnetization intensity vector between relaxation period _x, M _y, shown in 1 and formula 2:

$M_x\left(t\right)=M_x\left(0\right)sin\left({\mathrm{\ω}}_{0}t\right)e^{-t/T_2}$ (formula 1)

$M_y\left(t\right)=M_y\left(0\right)cos\left({\mathrm{\ω}}_{0}t\right)e^{-t/T_2}$ (formula 2)

In formula 1 and formula 2, T ₂for the relaxation time; M _xand M _ybe respectively the cross stream component of magnetization intensity vector between relaxation period; M _xand M (0) _y(0) cross stream component of magnetization intensity vector during t=0 is respectively; T is from T ₂the time that when relaxation starts, (t=0) experiences; ω ₀for the Larmor precession frequency of magnetization vector component.

In actual measurement, because main field always exists certain inhomogeneities, M _x, M _yattenuation pole the earth accelerate, corresponding T ₂relaxation time becomes as shown in Equation 3:

$\frac{1}{{T_2}^{*}}=\frac{1}{T_2}+\frac{1}{T_{2m}}$ (formula 3)

In formula 3, T _2mbecause main field inhomogeneities causes T ₂the part of relaxation time change; T ₂for real T ₂relaxation time; for T _2mand T ₂the T measured under combined effect ₂relaxation time.In order to eliminate the impact of main field inhomogeneities, self-rotary echo-pulse series is adopted to measure.Fig. 1 is that magnetic resonance spins echo (CPMG) measures sequence timing diagram, and as shown in Figure 1, the process of this measurement sequence is:

(1) within the long recovery time for TR, 90 ° of radio-frequency pulses are launched when time t=0;

(2) 180 ° of radio-frequency pulses are launched when t=τ=TE/2, due to Mxy component first poly-phase of faling apart afterwards mutually, now in receiving coil, appearance amplitude is first increased, the magnetic resonance signal (i.e. spin echo signal) of rear decay, there is maximum at t=2 τ=TE time place in this signal;

(3) per interval 2 τ launches 180 ° of radio-frequency pulses, repeats n (n=1,2,3 altogether ..., the numerical value of n can be free), namely obtain one group of ever-reduced spin echo signal of signal amplitude.

Carry out mathematical inversion by the spin echo signal obtained said process measurement and can try to achieve T ₂relaxation time spectrum, passes through T ₂relaxation time spectrum can obtain T corresponding to different sample component ₂distribution of relaxation times.

The femur main component that animal comprises the mankind is cortical bone, and its basic feature is the hole shape structure of bone trabecula composition.And there is a small amount of Bound moisture and Free water in hole between bone trabecula.T during hydrone generation surface relaxation in different aperture size ₂relaxation time is not quite similar, the T that macrovoid is corresponding ₂relaxation time is longer, the T that fine pore is corresponding ₂relaxation time is shorter.By by conventional bone density measurement data and T ₂relaxation time spectrum Value Data is analyzed, and just can realize using T ₂relaxation time spectrum reflects the object of the bone density size of tested sample.In practical application, by by conventional bone density data and T ₂peak-data on relaxation time spectrum between 300 ~ 500 μ s is analyzed, just available T ₂relaxation time reflects the bone density size of tested sample.

The present invention adopts T ₂relaxation time spectrometry and analytical method, provide the complete device for biological bone density measure and Analysis of bone mineral density method, comprising:

1) a ultrashort echo time magnetic resonance measurement device, this device mainly comprises a calibrating 120mm, the packaged type permanent magnet of main field frequency 10MHz, high speed coil quality factor change-over switch and highly integrated small-sized veneer magnetic resonance controller;

Fig. 2 is magnetic resonance T provided by the invention ₂the structure chart of the equipment of relaxation time analysis mensuration biological bone density, comprises computer control terminal 1, veneer magnetic resonance controller unit 2, signal amplification and switch control unit 3 and magnet unit 4 amounts to four parts.Wherein, veneer magnetic resonance controller unit 2 mainly comprises frequency synthesis and pumping signal radiating portion 5 and digital detection and signal process part 6.Signal amplifies and switch control unit 3 comprises low-noise preamplifier 7, radio frequency amplifier 8, radio-frequency (RF) switch control 9 and Q-switch switch 10.Magnet unit 4 comprises attemperating unit 11, permanent magnet 12 and transmitting and receiving coil 13.

The equipment of said determination biological bone density operationally, computer control terminal 1 controls the workflow of whole system, veneer magnetic resonance controller unit 2 is responsible for the generation of resonance and the collection of signal, signal amplifies and switch control unit 3 amplifies for the signal that laboratory collects and the on-off control effect of radio-frequency coil, the constant temperature of magnet unit 4 for Experimental Area and the structure in magnetostatic field region.During specific works, computer control terminal assigns instruction, launch driver unit from the synthesis of veneer magnetic resonance control unit medium frequency and signal and produce magnetic resonance excitation signal sequence, transfer to launching and receiving coil through radio frequency amplifier, now radio-frequency (RF) switch is signal foment; Radio-frequency coil controls and the co-controlling of Q-switch through radio-frequency (RF) switch after signal sequence excitation sample, switch to signals collecting state, gather original magnetic resonance signal, amplify after primary signal through low-noise preamplifier, transmit back computer control terminal by after digital detection and signal process part pretreatment, finally control terminal processes signal by computer and obtain T ₂relaxation time spectrum, as final data result.

2) one T ₂relaxation time analysis of spectrum measures the method for bone density data, can correctly simulate bone density-T ₂relaxation time calibration line the measurements and calculations being applied to bone density obtain the bone density of corresponding skeleton.By above-mentioned magnetic resonance T provided by the invention ₂relaxation time analyzes the equipment measuring biological bone density, collects T ₂the osseous tissue signal (i.e. original magnetic resonance signal) that relaxation time is extremely short, then obtain T through mathematical inversion ₂the data such as relaxation time spectrum; In detection method, by T ₂the linear corresponding relation existed between relaxation time and biological bone density is represented by graticule formula (formula 4):

BMD＝K·T ₂+C(4)

In above formula, BMD is bone density (BoneMineralDensity, g/cm ²), T ₂the T that (unit: μ s) is sample cortical bone ₂relaxation time data, K (unit: g/ (cm ²﹒ μ s)), C (unit: g/cm ²) be constant.According to our result of study, K value and C value are along with biological species is different, sex is different, age not equal factor has different values.In actual applications, need the bone density of the biological bone tissue of the certain sample size of first quantitative measurement, then adopt regression analysis to draw K value and the C value of biological place group, then obtain bone density according to formula 4 mensuration.Because this measuring method is safe and harmless, can take multiple measurements, thus person easy to use determines whether according to the size of bone density the process that osteoporosis and osteoporosis occur and develop.

Fig. 3 uses T in the embodiment of the present invention ₂relaxation time analysis of spectrum measures the particular flow sheet of biological bone density method, below for isolated rat femur (but being not limited to femur) bone densitometry (biological species is not limited to rat), describes T of the present invention in detail ₂relaxation time analyzes the method and concrete work process that measure biological bone density.

Step 1) measure sample preparation: the isolated rat femur getting the 12 monthly ages ovariectomy being numbered 1 ~ 10 is tested, and wherein 1 ~ 5 group is osteoporosis group, and 6 ~ 10 groups is matched group (sclerotin is normal); The long 20cm of rat body, body weight 600 grams;

Step 2) gauge preparation: open ultrashort echo biological bone and organize magnetic resonance relaxation spectrum measurement device, treat that device temperature is constant; Choose and add the water of relaxation reagent as substantially calibrating thing, determine the parameter such as magnetic field bias, gain size, 90 ° and 180 ° of pulse lengths;

Step 3) instrument parameter selection: the concrete measurement parameter chosen is magnetic field dominant frequency 10.71MHz (being provided by permanent magnet 12), echo sounding time TE=100 μ s, scanning accumulative frequency 4 times, recovery time 1000 milliseconds, sampling number 1024 (above-mentioned parameter data correspond to magnetic resonance excitation signal sequence basic parameter).One group (often organize a femur) is measured the magnetic field center position that sample is placed in magnetic resonance equipment and measures by each measurement, is often organized corresponding spin echo data (i.e. original magnetic resonance signal);

Step 4) calibration data process: the spin echo data recorded is carried out mathematical inversion process by the inverted parameters of setting, obtains T ₂distribution of relaxation times and T ₂relaxation time spectrum, accompanying drawing 4 is three groups of typical T in this experiment ₂relaxation time spectrum.Then by the T of different sample ₂relaxation spectrum and dual-energy x-ray borne densitometers measure the corresponding bone density data obtained and carry out regression analysis, obtain the K value in graticule formula (4) and C value;

Step 5) bone density measurement: obtain K value and C value by above-mentioned regression analysis; After determining K and C value, by measuring T ₂relaxation time spectrum; Carry out linear fit by method of least square and obtain graticule formula; In conjunction with graticule formula, namely Quick Measurement (Measuring Time is less than 0.5 minute) can be carried out to other rat femur.Concrete operation step is: first use the present invention to record the femur T of corresponding rat sample ₂relaxation time spectrum, gets T ₂between 300-500 μ s, the peak value T at peak is composed in time spectrum ₂relaxation time, by the T of respective sample ₂relaxation time is corresponding with the bone density data that it records by DEXA method, records 10 groups of samples altogether.To the T one to one recorded ₂relaxation time-bone density data carry out least-squares algorithm linear fitting, can obtain corresponding graticule formula.

T ₂there is linear corresponding relation between relaxation time and biological bone density, corresponding graticule formula is formula 4.

As the data in following table 1, along with the increase of femoral bmd, the T between 300 ~ 500 μ s of its correspondence ₂relaxation time peak value shortens gradually.For the data in table 1, take method of least square to carry out linear fit, gained graticule formula is as shown in above-mentioned formula (4).The schematic diagram that accompanying drawing 5 is the calibration straight line that utilizes least square fitting to go out in this experiment.

The thighbone density data of table 112 months old rats and the T of correspondence thereof ₂relaxation time

Table 2 lists K value corresponding to 12 months old rats femurs and C value.Can judge according to gained correlation coefficient, T ₂peak value in relaxation time spectrum between 300 ~ 500 μ s and the degree of correlation of bone density data very high, the contact of the two can be described with the linear relationship of formula 4, thus can find out adopt T ₂relaxation time spectral method can Accurate Determining bone density.

Table 212 months old rats thighbone density-T ₂relaxation time fit parameter values

Group	K value (g/ (cm 2﹒μs))	C value (g/cm 2)	Coefficient R value
				12 monthly ages	-0.0005	0.4950	0.9255

For the accuracy of checking gained graticule formula, choose and carry out confirmatory experiment with a collection of in vitro femur of 12 monthly age Ovariectomized Rats being numbered 1 ~ 5.First the T obtaining 5 groups of samples is measured ₂relaxation time spectrum, then according to the graticule formula that above-mentioned experiment records, calculate corresponding bone density data.As shown in Table 3 below, BMD1 is the rat femur bone density data utilizing DEXA method to record, T ₂relaxation time is the corresponding relaxation time data utilizing this method to record, and BMD2 is the bone density data utilizing calibration line formulae discovery to draw, Δ is the difference of BMD2 and BMD1, and error rate is the percentage ratio that Δ accounts for BMD2.According to error rate, the error rate absolute value of 5 groups of samples is all less than 5%, and visible this method precision is higher, therefore can substitute existing DEXA method Measurement accuracy bone density data by the method.

Table 3BMD=KT ₂+ C bone density computing formula verification msg

Numbering	BMD1(g/cm 2)	T 2Relaxation time (μ s)	BMD2(g/cm 2)	Δ(g/cm 2)	Error rate (%)
						1	0.2658	460	0.2650	-0.0008	-0.30
2	0.2734	425	0.2825	0.0091	3.22
						3	0.2852	400	0.2950	0.0098	3.32
4	0.2955	370	0.3100	0.0145	4.68
						5	0.3020	400	0.2950	-0.0070	-2.37

It should be noted that the object publicizing and implementing example is to help to understand the present invention further, but it will be appreciated by those skilled in the art that: in the spirit and scope not departing from the present invention and claims, various substitutions and modifications are all possible.Therefore, the present invention should not be limited to the content disclosed in embodiment, and the scope that the scope of protection of present invention defines with claims is as the criterion.

Claims (10)

1. a Bone mineral density equipment, described checkout equipment is based on magnetic resonance T ₂relaxation time spectrum detects bone density, comprise computer control terminal 1, veneer magnetic resonance controller unit 2, signal amplification and switch control unit 3 and magnet unit 4, described computer control terminal 1 connecting single board magnetic resonance controller unit 2, veneer magnetic resonance controller unit 2 amplifies with signal and switch control unit 3 is connected, and signal amplifies and switch control unit 3 is connected with magnet unit 4; Described veneer magnetic resonance controller unit 2 is for generation of resonance and collect original magnetic resonance signal; Described signal amplification and switch control unit 3 are for amplifying the signal collected and the switch controlling radio-frequency coil; Described magnet unit 4 is for keeping the constant temperature of surveyed area environment and building region magnetostatic field; Described computer control terminal 1, for controlling the workflow of complete equipment system, receives simultaneously and processes signal that described veneer magnetic resonance controller unit 2 collection treats thus obtain T2 relaxation time spectrum.

2. Bone mineral density equipment as claimed in claim 1, is characterized in that, described veneer magnetic resonance controller unit 2 comprises frequency synthesis and pumping signal radiating portion 5 and digital detection and signal process part 6; Electronic chip integrated for multiple height is integrated on one piece of circuit board by described veneer magnetic resonance controller unit 2, for realizing magnetic resonance measurement function.

3. Bone mineral density equipment as claimed in claim 2, it is characterized in that, described electronic chip comprises dsp chip, fpga chip, the power supply chip of FPGA, the configuring chip of FPGA and DA chip.

4. Bone mineral density equipment as claimed in claim 1, it is characterized in that, described signal amplifies and switch control unit 3 comprises low-noise preamplifier 7 for being carried out amplifying by described original magnetic resonance signal, be used for the magnetic resonance excitation signal sequence transmission that produced by veneer magnetic resonance controller unit 2 to the radio frequency amplifier 8 of radio-frequency coil, radio-frequency (RF) switch 9 and Q-switch switch 10; Described Q-switch switch 10 is with coil quality factor change-over switch, makes radio-frequency coil receive rapid translating between mode of operation in pumping signal transmitting mode of operation and magnetic resonance signal by described Q-switch switch.

5. Bone mineral density equipment as claimed in claim 1, is characterized in that, described magnet unit 4 comprises attemperating unit 11, permanent magnet 12 and transmitting and receiving coil 13; Described permanent magnet 12 is a kind of open moveable magnets, provides dominant frequency to be the magnetic field of 10.71MHz.

6. Bone mineral density equipment as claimed in claim 5, it is characterized in that, the sample bore of described permanent magnet 12 reaches 120mm.

7. utilize Bone mineral density equipment described in claim 1 ~ 6 to carry out a method for Bone mineral density, it is characterized in that, comprise the steps:

1) measure one group the magnetic field center position that sample is placed in described Bone mineral density equipment, adopt self-rotary echo-pulse series to measure, often organized the original magnetic resonance signal measuring sample, as often organizing the spin echo data measuring sample;

2) by step 1) the spin echo signal data that obtain by mathematical inversion, obtain T ₂relaxation time spectrum;

3) by T ₂the linear corresponding relation existed between relaxation time and biological bone density is represented by graticule formula formula 4, and the K value in formula 4 and C value are by the T by different sample ₂relaxation spectrum and dual-energy x-ray borne densitometers measure the corresponding bone density data obtained to carry out regression analysis and obtains;

BMD＝K·T ₂+C(4)

In formula 4, BMD is bone density (BoneMineralDensity, g/cm ²); T ₂for the T of sample cortical bone ₂relaxation time data, unit is μ s; K is constant, and unit is g/ (cm ²﹒ μ s); C is constant, and unit is g/cm ²; K value and C value are along with biological species is different, sex is different, age different value is different;

4) step 2 is utilized) T that obtains ₂relaxation time spectrum, calculates detection bone density according to formula 4, completes Bone mineral density.

8. Bone mineral density method as claimed in claim 7, is characterized in that, step 2) spin echo data recorded is carried out mathematical inversion process, obtain T ₂relaxation time spectrum; Described mathematical inversion is conversion inversion algorithm, specifically comprises the steps:

21) measure the superposition that spin echo signal y (t) obtained is a series of single hole spin echo signal, through type 11 represents; Corresponding T is obtained by solving formula 11 ₂relaxation time spectrum; Formula 11 is:

$y\left(t\right)={\mathrm{\Σ}}_i{f}_i\·\exp (-\frac{t}{T_{2i}})+\mathrm{\ϵ}\left(t\right),t=n\·\mathrm{\τ}$ (formula 11)

In formula 11, y (t) is spin echo signal; f _ibe the share of the i-th class hole shared by total pore space; T _2ibe the T of the i-th class hole ₂relaxation time; τ is the echo sounding time; ε (t) is random noise sequences;

22) by establishing target functional expression 12, formula 11 is solved:

${\mathrm{\χ}}^2={\mathrm{\Σ}}_{i=1}^n{\[y_i-{\mathrm{\Σ}}_{j=1}^m(f_j\·m_{ij})\]}^2+\mathrm{\λ}\·{\mathrm{\Σ}}_{j=1}^m{f}_j=\left|\right|y-Mf|{|}^2+\mathrm{\λ}\·|\left|f\right|{|}^2$ (formula 12)

In formula 12, M=[m _ij]=[exp (-t _i/ T _2j)], m _ij=exp (-t _i/ T _2j); λ is smoothing factor; F=(f ₁, f ₂..., f _m) ^tfor amplitude; χ ²for corresponding object function; y _ifor the n chosen ties up the magnetic resonance signal that in time domain, the i-th dimension is corresponding; f _jcorresponding hole share shared in total pore space is tieed up for m ties up jth in T2 spatial domain; Y is spin echo signal;

23) to step 22) in amplitude f kth (k=1,2 ..., m) individual component asks extreme value, and makes it equal 0, obtains formula 13:

(M ^tm) f+ λ I _{m × m}f=M ^ty (formula 13)

In formula 13, I _{m × m}for m × m rank unit matrix; M=[m _ij]=[exp (-t _i/ T _2j)], m _ij=exp (-t _i/ T _2j); λ is smoothing factor; F=(f ₁, f ₂..., f _m) ^tfor amplitude; Y is spin echo signal;

24) linear change is done to formula 13, make f=M ^tc substitutes into formula 13, setting one value, and through type 14 obtains the least square solution of equation 11:

$\hat{f}=M^T\·{(M\·M^T+{\mathrm{\λI}}_{n\×n})}^{-1}\·y$ (formula 14)

In formula 14, for n ties up the solution of range value; The number of n is that cloth is counted, a corresponding n T ₂relaxation time values.

9. Bone mineral density method as claimed in claim 7, is characterized in that, step 3) described graticule formula carries out linear fit especially by method of least square and obtains.

10. Bone mineral density method as claimed in claim 7, it is characterized in that, described Bone mineral density method is method for fast measuring, measures required time and is less than 0.5 minute.