JP6762703B2 - Method for discriminating tumor site, tumor site discriminating device - Google Patents

Description

The present invention relates to a method for discriminating a tumor site and a device for discriminating a tumor site.

In recent years, Japan has entered a rapidly aging society, and the number of cancer patients is on the rise. For example, lymph node metastasis is one of the important prognostic factors in gastrointestinal cancer such as gastric cancer and colorectal cancer, and breast cancer, and it is not possible to accurately diagnose the presence or absence of lymph node metastasis in deciding how to treat a patient. It's very important. In particular, the lymph node that cancer cells away from the primary lesion first reach is called the sentinel lymph node. Since the number of cancer cells that reach the sentinel lymph node is initially limited, if the sentinel lymph node is examined and the cancer cells have very few metastases, it can be considered that they have hardly metastasized to the subsequent lymph nodes. Has been done.

As one method of intraoperative rapid diagnosis, for example, as disclosed in Patent Document 1, fluorescence observation is performed using 5-aminolevulinic acid (5-ALA) for cancer detection in a wide range including the gastrointestinal region. The method is applied. 5-ALA is a type of amino acid that is also present in the living body, is water-soluble, and can be administered orally or topically. When 5-ALA is administered from outside the body, it is rapidly metabolized to heme in normal cells, but in cancer cells, the metabolite protoporphyrin IX (PpIX) is selectively accumulated due to the difference in the activity of metabolic enzymes. Here, since heme does not recognize fluorescence, and PpIX is a fluorescent substance, it is possible to distinguish between a tumor site and a non-tumor site by detecting this light.

International Publication No. 2013/002350

However, in the human body, lymph nodes are surrounded by connective tissue, and these connective tissues (fat, collagen, etc.) are strong in the blue to green wavelength range under blue excitation light. It emits autofluorescence. The wavelength band of this autofluorescence has an overlap with the wavelength band of the fluorescence derived from PpIX.

FIG. 1 is a drawing showing the fluorescence spectra of collagen and FAD (Flavin Adenine Dinucleotide), which are typical substances having autofluorescence contained in a human body. Collagen has a fluorescence peak near 500 nm. Further, FAD has a fluorescence peak near 530 nm. It is confirmed that all the substances have a certain degree of fluorescence intensity in the vicinity of the wavelength of 635 nm, although they are lower than the fluorescence intensity at the peak wavelength.

FIG. 2 is a drawing showing an example of the fluorescence spectrum of a sample containing the autofluorescent substance such as collagen and FAD and PpIX. In FIG. 2, (a) shows the fluorescence spectrum of PpIX alone, (b) shows the fluorescence spectrum derived from the autofluorescent substance, and (c) shows the fluorescence spectrum actually measured. However, in FIG. 2, for convenience of explanation, the curves of (a), (b), and (c) are superposed on the same drawing with values standardized by different standards. Therefore, it should be noted that in FIG. 2, the magnitude relation of the values on the vertical axis of the graph does not always match between the curves of (a), (b), and (c).

As shown in FIG. 2A, the fluorescence spectrum of PpIX has a peak value near a wavelength of 635 nm. Therefore, when the sample does not contain an autofluorescent substance, the light (fluorescence) emitted from the sample is detected by spectroscopically detecting the light having a wavelength of around 635 nm, and the intensity distribution thereof is examined. It is possible to discriminate a site having a high fluorescence intensity, more specifically, a site showing a fluorescence intensity exceeding a predetermined threshold, as a tumor site.

However, since the sample actually contains an autofluorescent substance, among the light emitted from the sample, when the light having a wavelength of around 635 nm is separated and detected, the fluorescence derived from the autofluorescent substance shown in FIG. 2B is generated. Can be superposed. Therefore, the fluorescence intensity in the vicinity of the wavelength of 635 nm is recognized even in the portion where PpIX is not accumulated. The intensity of fluorescence derived from an autofluorescent substance varies depending on the site of the tissue extracted as a sample, and also has individual differences. Therefore, if a method of discriminating a site where the fluorescence intensity near a wavelength of 635 nm exceeds a predetermined threshold value as a tumor site is adopted, in some cases, even though PpIX is not accumulated, fluorescence derived from an autofluorescent substance is used. Due to its high strength, non-tumor sites may be mistakenly identified as tumor sites.

That is, in order to identify a tumor site, particularly a tumor site covered with connective tissue such as lymph nodes, in clinical application, a method for eliminating autofluorescence due to endogenous tissue is required. An object of the present invention is to realize a method and an apparatus capable of more accurately discriminating a tumor site from a sample containing connective tissue containing a substance having autofluorescence such as collagen and FAD.

なお、式（１）において、Ａ_c、及びＡ_i（ｉ＝１，２，…，ｎ）はいずれも係数である。 The present invention irradiates porphyrins accumulated in a suspected tumor site of a sample containing an autofluorescent substance having one or more types n with excitation light, and detects the fluorescence emitted by the porphyrins after excitation. It is a method of determining
The step (a) of irradiating the sample with the excitation light to excite the porphyrins,
Of the fluorescence emitted by the sample, the intensity distribution of the light L _c received through the filter F _c that transmits light in the wavelength band λ _c including a part of the fluorescence wavelength band emitted by the porphyrins corresponds to. Step (b) of acquiring image data I _c and
Of the fluorescence emitted by the sample, the filter F _si that transmits light in the wavelength band λ _si (i = 1, 2, ..., N) including a part of the wavelength band of fluorescence emitted by the n types of autofluorescent substances. Image data I _si (i = 1, 2, ..., N) corresponding to the intensity distribution of the light L _si (i = 1, 2, ..., N) received via (i = 1, 2, ..., N) The step (c) for acquiring n) and
Fluorescence emitted by the sample according to the following formula (1) based on the image data I _c and the image data I _si (i = 1, 2, ..., N) obtained in the steps (b) and (c). Among these, it is characterized by having a step (d) of generating image data I _p corresponding to the intensity distribution of fluorescence emitted by the porphyrins.

In Eq. (1), A _c and A _i (i = 1, 2, ..., N) are both coefficients.

The shape of the spectrum itself of the fluorescence emitted by various autofluorescent substances is determined depending on the type of the autofluorescent substance. In addition, the shape of the fluorescence spectrum itself emitted from the porphyrins is also uniquely determined according to the porphyrins. That is, the spectrum of fluorescence emitted when the suspected tumor site is irradiated with excitation light depends on the type of autofluorescent substance contained in the suspected tumor site and the content ratio of each autofluorescent substance and porphyrins. Become.

This method presupposes that the autofluorescent substance contained in the sample is a known autofluorescent substance. That is, with respect to the n types of autofluorescent substances that are assumed to be contained in the suspected tumor site, information on typical fluorescence spectra is recognized in advance. Then, as described above, the information on the spectrum of the fluorescence emitted from the porphyrins is also recognized in advance.

As described above, the wavelength band of fluorescence emitted from porphyrins may overlap with the wavelength band of fluorescence emitted from various autofluorescent substances. On the other hand, as described above, the spectrum of fluorescence derived from various autofluorescent substances and the spectrum of fluorescence emitted from porphyrins each show waveforms derived from the substance. Therefore, the sample is irradiated with excitation light, and among the fluorescence emitted from the sample, light of a plurality of specific wavelength bands is received, and the intensity distribution of the light in each wavelength band is obtained to obtain autofluorescence. It is possible to specify the ratio of the fluorescence derived from the substance and the fluorescence derived from the porphyrins, respectively.

That is, according to this method, the coefficients A _c and A _i (i = 1, 2, ..., N) are set for the fluorescence emitted from the sample containing n kinds of autofluorescent substances and porphyrins. Therefore, the fluorescence component derived from porphyrins can be extracted from the fluorescence emitted from the sample only by performing the calculation based on the above equation (1).

In the step (d), the image data I _p corresponding to the intensity distribution of the fluorescence emitted by the porphyrins among the fluorescence emitted by the sample may be generated by the following formula (2). However, in equation (2), A ₀ is a coefficient.

The method includes a light source unit for emitting excitation light, a light receiving portion for receiving the light emitted from the sample, the filter F _c which transmits light of a wavelength band lambda _c, and a wavelength band _{λ si (i = 1,2, ...} , This is done using the filter F _si (i = 1, 2, ..., N) that transmits the light of n). Therefore, it is conceivable that some light receiving units provided in the device may output a signal indicating that the light has been received even though the light is not received. In particular, in this case, the light receiving unit recognizes that it has received the same level of light (offset component) regardless of the wavelength band. Therefore, in the above equation (2), the constant term A ₀ irrelevant to the wavelength band is predetermined. With this configuration, this offset component can be canceled out, so that the fluorescent component derived from porphyrins contained in the fluorescence can be strictly extracted. However, it is possible to extract the fluorescence component derived from porphyrins contained in the fluorescence within the range necessary for discriminating the tumor site by an arithmetic expression that does not consider the offset component as in the above equation (1). is there.

In addition to the above method, it may have a step (e) of discriminating between a tumor site and a non-tumor site based on the image data I _p obtained in the step (d). The image data I _p is data based on the fluorescence intensity derived from porphyrins emitted from the sample. Therefore, for example, in this image data _Ip , it can be determined that the tumor site exists in the region on the sample corresponding to the region showing the light intensity exceeding a predetermined threshold value.

The filter F _c used in the step (b) and the filter F _si (i = 1, 2, ..., N) used in the step (c) determine the intensity of fluorescence emitted by the porphyrins. It may be composed of a combination of filters that minimizes the standard estimation error at the time of acquisition.

The term "porphyrins" as used herein refers to porphyrins having a substituent on the porphyrin ring, and as a kind of porphyrins, for example, PpIX and protoporphyrins such as photo-protoporphyrin (PPp) produced from PpIX There is a kind.

The porphyrins can also be protoporphyrins. In particular, when the protoporphyrins are protoporphyrin IX and the number of types of autofluorescent substances is 2.
The wavelength band λ _c of the light transmitted by the filter F _c is a wavelength band including 635 nm.
The wavelength band λ _s1 of the light transmitted by the filter F _s1 is a wavelength band having a center wavelength of 520 nm or more and 550 nm or less.
The wavelength band λ _s2 of the light transmitted by the filter F _s2 may be a wavelength band having a center wavelength of 730 nm or more and 760 nm or less.

By such a method, the fluorescence derived from protoporphyrin IX (PpIX) having a spectral peak near 635 nm and the fluorescence derived from an autofluorescent substance containing collagen and FAD are superposed and emitted from the sample. Even in this case, image data I _p corresponding to the fluorescence intensity derived from PpIX can be generated. By discriminating between the tumor site and the non-tumor site based on the image data _Ip , it is possible to discriminate the tumor site more accurately than before.

Furthermore, in the above method,
The filter F _c has a full width at half maximum of 10 nm or more and 30 nm or less.
The filter F _s1 has a full width at half maximum of 30 nm or more and 50 nm or less.
The filter F _s2 may have a full width at half maximum of 180 nm or more and 220 nm or less.

なお、式（１）において、Ａ_c、及びＡ_i（ｉ＝１，２，…，ｎ）はいずれも係数である。 Further, the present invention irradiates porphyrins accumulated in a suspected tumor site of a sample containing an autofluorescent substance having one or more types n with excitation light, and detects the fluorescence emitted by the porphyrins after excitation. A device for discriminating tumor sites
The light source unit that emits the excitation light and
A light receiving unit that receives the fluorescence emitted from the sample and
A filter F _c that transmits light in the wavelength band λ _c including a part of the wavelength band of fluorescence emitted by the porphyrins, and
Filter F _si (i = 1, 2, ..., N) that transmits light in the wavelength band λ _si (i = 1, 2, ..., N) including a part of the wavelength band of fluorescence emitted by the n types of autofluorescent substances. n) and
It has an arithmetic processing unit that performs arithmetic processing based on the intensity of light received by the light receiving unit.
The light receiving part is
Of the fluorescence emitted by the sample, the light L _c is received through the filter F _c , and the image data I _c corresponding to the intensity distribution of the light L _c is output to the arithmetic processing unit.
Wherein among the fluorescence which said sample is emitted filter _{F si (i = 1,2, ...} , n) light _{L si (i = 1,2, ...} , n) through the receiving and the light L _si (i The image data I _si (i = 1, 2, ..., N) corresponding to the intensity distribution of = 1, 2, ..., N) is output to the arithmetic processing unit.
Based on the image data I _c and the image data I _si (i = 1, 2, ..., N), the arithmetic processing unit emits the porphyrins among the light emitted by the sample according to the following formula (1). It is characterized in that image data I _p corresponding to the fluorescence intensity distribution is generated.

In Eq. (1), A _c and A _i (i = 1, 2, ..., N) are both coefficients.

According to the above apparatus, even if the fluorescence emitted from the sample is superposed with the fluorescence derived from porphyrins and the fluorescence derived from the autofluorescent substance, the fluorescence emitted by the porphyrins is extracted in the arithmetic processing unit. Then, in the arithmetic processing unit, image data I _p corresponding to the distribution of the fluorescence intensity emitted by the porphyrins is generated. Therefore, since the tumor site and the non-tumor site can be discriminated based on this image data _Ip , the tumor site can be discriminated more accurately than before.

The tumor site discriminating device may include a storage unit that stores the values of the coefficients _Ac and A _i (i = 1, 2, ..., N).

Based on the image data I _c and the image data I _si (i = 1, 2, ..., N), the arithmetic processing unit emits the porphyrins among the light emitted by the sample according to the following formula (2). Image data I _p corresponding to the fluorescence intensity distribution may be generated. In this case, the tumor site discriminating device may include a storage unit that stores the values of the coefficients _Ac , A _i (i = 1, 2, ..., N) and A _0. Absent.

The porphyrins can also be protoporphyrins. In particular, when the protoporphyrins are protoporphyrin IX and the number of types of autofluorescent substances is 2.
The wavelength band λ _c of the light transmitted by the filter F _c is a wavelength band including 635 nm.
The wavelength band λ _s1 of the light transmitted by the filter F _s1 is a wavelength band having a center wavelength of 520 nm or more and 550 nm or less.
The wavelength band λ _s2 of the light transmitted by the filter F _s2 may be a wavelength band having a center wavelength of 730 nm or more and 760 nm or less.

In addition
The filter F _c has a full width at half maximum of 10 nm or more and 30 nm or less.
The filter F _s1 has a full width at half maximum of 30 nm or more and 50 nm or less.
The filter F _s2 may have a full width at half maximum of 180 nm or more and 220 nm or less.

In addition, the device for discriminating the tumor site is
Equipped with an information input section
The storage unit contains information on the filter F _si (i = 1, 2, ..., N), the coefficients A _c , and A _i (i = 1) according to the type of the autofluorescent substance contained in the sample. , 2, ..., n) stores information about the combination of values,
When the information input unit gives information about the type of autofluorescent substance contained in the suspected tumor site, the arithmetic processing unit receives information on the filter F _si (i = 1, 2, ..., N) from the storage unit. _Is read to specify the filter F _si (i = 1, 2, ..., N), and the values of the coefficients _Ac and A _i (i = 1, 2, ..., N) are read from the storage unit. Therefore, the calculation of the above equation (1) may be performed.

It may be possible to determine the type of autofluorescent substance contained in the sample from the analysis result of the fluorescence spectrum received by irradiating the sample with light in advance and the characteristics of the part of the sample in the human body. In the storage unit of the above device, a coefficient _Ac and A _i (i = 1, 2, ..., N) according to the type of the autofluorescent substance contained in the sample, and a filter F _si (i = 1, 2 _,, n) are stored. ..., N) information is stored. Therefore, the image data _{I si (i = 1,2, ...} , n) by using the optimum filter corresponding to the specimen can be generated, also, the image data _{I si (i = 1,2, ...} , Image data I _p corresponding to the distribution of fluorescence intensity emitted by porphyrins can be generated based on n) and the optimum coefficients A _c and A _i (i = 1, 2, ..., N) according to the sample. it can.

When the arithmetic processing unit generates the image data I _p by the above equation (2), the storage unit corresponds to the type of the autofluorescent substance contained in the sample, and the filter F _si (i = 1). , 2, ..., N) and information on the combination of the coefficients _Ac , A _i (i = 1, 2, ..., N) and A ₀ may be stored in memory. Absent. At this time, when the information input unit gives information about the type of autofluorescent substance contained in the suspected tumor site, the arithmetic processing unit receives the filter F _si (i = 1, 2, ..., N) from the storage unit. ) _Is read out to identify the filter F _si (i = 1, 2, ..., N), and the coefficients _Ac , A _i (i = 1, 2, ..., N) and The value of A ₀ may be read out and the calculation of the above equation (2) may be performed.

Further, the tumor site discriminating device is placed on the optical path between the sample and the light receiving portion by any one of the filter F _c and the filter F _si (i = 1, 2, ..., N). It may be provided with a filter switching unit for switching and installing the filter.

According to the method for discriminating a tumor site and the device for discriminating a tumor site of the present invention, the influence of fluorescence derived from an autofluorescent substance can be largely eliminated, so that the tumor site can be discriminated more accurately than before.

It is a figure which shows the autofluorescence spectrum of collagen and FAD. It is a figure which shows an example of the fluorescence spectrum of the sample containing an autofluorescent substance and PpIX. It is a figure which shows an example of the appearance of the tumor site discriminator typically. It is a block diagram which shows typically an example of the internal structure of the tumor site discriminating apparatus. It is a figure which shows the fluorescence spectrum of PpIX. Central wavelength lambda _c of the filter F _c, is the central wavelength lambda _s1 filter F _s1, and the size distribution of the error ε with respect to the center wavelength lambda _s2 filter F _s2 as a graph. FWHM FWHM _c of the filter F _c, is the full width at half maximum FWHM _s1 filter F _s1, and the size distribution of the error ε for the full width at half maximum FWHM _s2 filter F _s2 as a graph. It is a graph of the relationship between the fluorescence intensity I _{pr (j_k)} estimated by the calculation and the actual fluorescence intensity I _{p (j_k)} . This is an example of image data when a sample in which PpIX, FAD, and collagen are arranged in a row is irradiated with light from a light source unit. It is a block diagram which shows typically an example of the internal structure of another embodiment of a tumor site discriminating apparatus.

[Device configuration]
The configuration of the tumor site discriminating device will be described with reference to the drawings. In each drawing, the dimensional ratio in the drawing and the actual dimensional ratio do not always match.

FIG. 3 is a drawing schematically showing an example of the appearance of the tumor site discriminating device. Further, FIG. 4 is a block diagram schematically showing an example of the internal configuration of the tumor site discriminating device.

As shown in FIG. 3, the tumor site determination device 1 (hereinafter, may be appropriately referred to as “device 1”) includes a holder mounting port 11 and a display unit 12. The holder mounting port 11 is a mechanism for mounting the sample holder 10 containing the sample 2 (not shown in FIG. 3, see FIG. 4). In addition, the display unit 12 corresponds to a monitor that displays the result of discriminating between the tumor site and the non-tumor site by the tumor site discriminating device 1.

Although the configuration in which the display unit 12 is provided on the main body of the tumor site discrimination device 1 is shown here, the main body of the device 1 does not have the display unit 12 and displays the discrimination result on another monitor. You may adopt the structure to make it. Further, here, an example is illustrated in which the tumor site determining device 1 is configured to determine whether or not the sample 2 contains a tumor site by inserting the sample holder 10 containing the sample 2. However, it is not limited to this configuration.

Specimen 2 is a sample (for example, a living tissue including a sentinel lymph node) for which a tumor site is discriminated. Predetermined measures have been taken for this sample 2 so that porphyrins are accumulated in the tumor site if it is contained. As an example, this biological tissue can be assumed to have been excised after administration of 5-ALA to the human body.

As shown in FIG. 4, the device 1 includes a light source unit 21, a filter 22, a dichroic mirror 23, an objective lens 24, a filter switching unit 25, a light receiving unit 26, an arithmetic processing unit 27, and a storage unit 28. Note that FIG. 4 assumes a configuration in which the device 1 includes a display unit 12 in accordance with FIG.

The light source unit 21 can be composed of, for example, a lamp such as a mercury lamp, a xenon lamp, a halogen lamp, or a metal halide lamp, as well as a light emitting diode element, a laser diode element, or the like. The filter 22 has a function of selectively transmitting light having a specific wavelength among the light emitted from the light source unit 21, and is composed of, for example, a dielectric multilayer film. Here, the filter 22 will be described as having a function of selectively transmitting light having a wavelength of 405 nm, but in the present embodiment, a function of selectively transmitting light in a specific wavelength band of 385 nm or more and 425 nm or less is provided. You just have to have it. More generally, the filter 22 has a function of selectively transmitting light in the wavelength band necessary for exciting the porphyrins accumulated when the sample 2 contains a tumor site. I just need to be there.

The dichroic mirror 23 has a function of reflecting light in a predetermined wavelength band and transmitting light in another predetermined wavelength band, and can be formed of, for example, a dielectric multilayer film. Here, the dichroic mirror 23 will be described as having a function of reflecting light having a wavelength of 405 nm and transmitting light having a wavelength of 580 nm or more. The dichroic mirror 23 reflects light having a wavelength selected by the filter 22, and is emitted from the sample 2 and included in the filter switching unit 25 described later. The filter F _c and the filter F _si (i = 1, 2). It suffices to have a function of transmitting light having a wavelength selected by, ..., N).

The excitation light 31 emitted from the light source unit 21 and transmitted through the filter 22 and having a wavelength of around 405 nm is reflected by the dichroic mirror 23 and guided to the objective lens 24. Then, the light that has passed through the objective lens 24 passes through the holder 10 and irradiates the sample 2. When PpIX (which is also an example of protoporphyrins), which is an example of porphyrins, is accumulated in the sample 2, PpIX is excited by the excitation light 31 having a wavelength of 405 nm and emits fluorescence 32. The step of irradiating the sample 2 with the excitation light 31 to excite the PpIX accumulated in the sample 2 corresponds to the step (a).

The fluorescence 32 passes through the holder 10 and travels in the direction opposite to the excitation light, and is guided to the objective lens 24. Then, the fluorescence 32 passes through the dichroic mirror 23 and heads for the filter switching unit 25.

In the present embodiment, the filter switching unit 25 can switch and arrange three filters, the filter F _c , the filter F _s1 , and the filter F _s2 , on the optical path between the sample 2 and the light receiving unit 26. Each filter (F _c , F _s1 , F _s2 ) has a function of selectively transmitting light in a predetermined wavelength band, and each of them can be composed of an optical filter such as a bandpass filter. The number of filters is not limited to 3, and a configuration in which 4 or more filters can be installed while switching on the optical path may be used. This point will be described later.

In the present embodiment, the filter F _c has a function of transmitting light in the wavelength band λ _c including the main peak wavelength of fluorescence emitted by porphyrins. For example, when PpIX, which is an example of porphyrins, is used as the detection target, the peak wavelength of fluorescence derived from PpIX is near 635 nm as shown in FIG. 5, so light in the wavelength band λ _c including this 635 nm is selected. A filter F _c designed to be transparent is used. As an example of the filter F _c, a filter having a center wavelength of 634 nm and a full width at half maximum of 20 nm can be used. In this case, the wavelength band λ _c is 624 nm or more and 644 nm or less. The method for determining the filter F _c will be described later.

In the present embodiment, the center wavelength of the transmitted light of the filter F _s1 is shorter than that of the filter F _c . As an example of the filter F _s1, a filter having a center wavelength of 536 nm and a full width at half maximum of 40 nm can be used. In this case, the wavelength band λ _s1 is 516 nm or more and 556 nm or less. The method for determining the filter F _s1 will be described later.

In the present embodiment, the center wavelength of the transmitted light of the filter F _s2 is on the longer wavelength side than the filter F _c . As an example of the filter F _s2, a filter having a center wavelength of 745 nm and a full width at half maximum of 204 nm can be used. In this case, the wavelength band λ _s1 is 643 nm or more and 847 nm or less. The method for determining the filter F _s2 will be described later.

Receiving unit 26 receives a filter that is set in the filter switching unit 25 of each fluorescence 32 that passes through the _{(F c, F s1, F} s2). The light receiving unit 26 outputs the fluorescence intensity according to the position on the sample 2 to the arithmetic processing unit 27. More specifically, the light receiving unit 26 receives the fluorescence 32 (hereinafter, referred to as “fluorescence L _c ”) received through the filter F _c, and the fluorescence L _c corresponding to the position on the sample 2 is received. The image data I _c corresponding to the intensity distribution is output to the arithmetic processing unit 27. The step in which the light receiving unit 26 outputs the image data I _c to the arithmetic processing unit 27 corresponds to the step (b).

Similarly, the light receiving unit 26 receives the fluorescence 32 (hereinafter, referred to as “fluorescence L _s1 ”) received through the filter F _s1 and _adjusts the intensity distribution of the fluorescence L _s1 according to the position on the sample 2. The corresponding image data _Is1 is output to the arithmetic processing unit 27. Similarly, the light receiving unit 26 receives the fluorescence 32 (hereinafter, referred to as “fluorescence L _s2 ”) received through the filter F _s2, and _adjusts the intensity distribution of the fluorescence L _s2 according to the position on the sample 2. The corresponding image data _Is2 is output to the arithmetic processing unit 27. The step in which the light receiving unit 26 outputs the image data _Is1 and _Is2 to the arithmetic processing unit 27 corresponds to the step (c).

Information about the coefficients _Ac , A ₁ , A ₂ , and A ₀ is stored in the storage unit 28 in advance. Arithmetic processing unit 27 from the storage unit 28, the coefficient _{A c, A 1, A 2} , and reads the information about the A _0, the image data I _c given from the light receiving unit 26, the I _s1, and I _s2 Based on this, the image data I _p is derived by performing the calculation of the following equation (3).

The constants _{A c, A 1, A 2} , and A ₀ is the fluorescence having intensity of the fluorescence L _c having a component of the wavelength band lambda _c of the fluorescence 23, the component of the wavelength band lambda _s1 of the fluorescence 23 Given the intensity of L _{s1 and} the intensity of fluorescence L _s2 having a component of the wavelength band λ _s2 of the fluorescence 23, the calculation of the above equation (3) is performed to derive the porphyrins contained in the fluorescence 23. It is predetermined so that the fluorescence intensity can be calculated. Therefore, when the arithmetic processing unit 27 performs the arithmetic of the above equation (3), the image data I _p corresponding to the intensity distribution of the fluorescence emitted by the porphyrins contained in the sample 2 is generated. The step in which the arithmetic processing unit 27 generates the image data I _p corresponds to the step (d).

The method for determining the coefficients A _c , A ₁ , A ₂ , and A ₀ will be described later.

This image data I _p is output to the display unit 12. As a result, the fluorescence intensity derived from porphyrins according to the position of the sample 2 is displayed as image information. For example, by checking this displayed image, it is possible to determine whether or not a tumor site is present in the sample 2, and if the tumor site is present in the sample 2, the tumor in the sample 2 is present. The location of the site can be identified. This determination process corresponds to the step (e).

Further, the arithmetic processing unit 27 may even determine whether or not a tumor site is present in the sample 2 based on the image data _Ip . For example, the arithmetic processing unit 27 determines whether or not the fluorescence intensity for each position exceeds a predetermined threshold value based on the image data _Ip . Then, the arithmetic processing unit 27 determines that the portion where the fluorescence intensity exceeds the predetermined threshold value is the tumor site, and the portion where the fluorescence intensity is equal to or lower than the predetermined threshold value is the non-tumor site. This determination process corresponds to the step (e).

The display unit 12 displays, for example, a mark indicating that the tumor site is a tumor site or colored image data at a predetermined position on the image of the sample 2 based on the coordinate information of the tumor site sent from the arithmetic processing unit 27. It doesn't matter if you do. Further, when the region determined to be the tumor site by the arithmetic processing unit 27 does not exist, the display unit 12 may display the information to that effect.

By visually checking the display unit 12, the inspector can easily recognize whether or not the tumor site is present in the sample 2, and if the tumor site is present, the location of the tumor site. Further, for example, the device 1 is provided with an operation button, and when the holder 10 containing the sample 2 is attached to the device 1 and the operation button is pressed, the excitation light is emitted from the light source unit 21. Therefore, the tumor site discrimination process of the sample 2 can be automatically performed. As a result, the variation in the judgment result depending on the skill of the inspector is eliminated, and the judgment by the pathologist becomes unnecessary.

The image data I _p generated by the above process is a light intensity distribution belonging to the wavelength band λ _c including the peak wavelength of the fluorescence spectrum of porphyrins among the fluorescence 32 emitted from the sample 2, and is derived from an autofluorescent substance. This is the information in which the fluorescence of is excluded. Therefore, according to the tumor site discriminating device 1, since the tumor site and the non-tumor site can be discriminated based on this image data _Ip , fluorescence derived from an autofluorescent substance is included as compared with the conventional case. The risk of misidentification due to the above can be greatly reduced.

The order of the filters (F _c , F _s1 , F _s2 ) arranged on the optical path may be any order.

The dichroic mirror 23 is provided for the purpose of partially sharing the optical paths of the excitation light 31 and the fluorescence 32 in order to reduce the size of the device 1. That is, the dichroic mirror 23 is not always necessary if the sample 2 is irradiated with the excitation light emitted from the light source unit 21 and the light emitted from the sample 2 is received by the light receiving unit 26. Further, the filter 22 may be integrated with the light source unit 21. It goes without saying that the configuration of the device 1 shown in FIG. 4 is only an example, and various design changes can be made as long as the configuration realizes the same function.

[Determination of filter conditions]
Hereinafter, an example of a method for determining the three filters, the filter F _c , the filter F _s1 , and the filter F _s2 , will be described. Here, for convenience of explanation, the case where the assumed autofluorescent substances are two types of collagen and FAD will be described, but it will be described later that the same discussion is possible in the case of three or more types.

(Step to determine the center wavelength)
The most preferable filters F _c , filter F _s1 , and filter F _s2 are calculated based on the combined fluorescence of porphyrin-derived fluorescence and autofluorescent substance-derived fluorescence received through each filter. It is a filter that can isolate only fluorescence derived from porphyrins. The following simulation was performed to set the conditions of the filter F _c , the filter F _s1 , and the filter F _s2 that can realize this.

As described above, the fluorescence 32 emitted from the sample 2 is a combination of the fluorescence derived from the porphyrins and the fluorescence derived from the autofluorescent substance, and the degree of this synthesis is the porphyrins contained in the sample 2 and the autofluorescence. It is affected by the amount of fluorescent material. Therefore, it is considered that the fluorescence 32 irradiated from the sample 2 is roughly represented by the linear sum of the fluorescence derived from porphyrins and the fluorescence derived from the autofluorescent substance.

Therefore, a test spectrum was prepared by synthesizing the fluorescence spectrum of PpIX (see FIG. 5) as an example of porphyrins and the fluorescence spectrum of collagen and FAD (see FIG. 1) as an example of an autofluorescent substance at a predetermined ratio. Here, by changing the amount of PpIX, the amount of collagen, and the amount of FAD in three steps (0, w1, w2) (where w2> w1), a total of 27 types of reference test spectra TSb _{1 to} TSb ₂₇ It was created. Here, the case where the content is 0 is included in one of the three stages. That is, the test spectrum simulates a fluorescence spectrum that includes, for example, PpIX and collagen and does not contain FAD.

However, in reality, it is assumed that the fluorescence 32 received by the light receiving unit 26 does not appear as a perfect linear sum of the combined spectra of each fluorescence and contains a certain disturbance. Therefore, in order to simulate this disturbance, Gaussian noise having a standard deviation of 5% of the maximum concentration is superimposed on each of the 27 types of reference test spectra TSb _{1 to} TSb ₂₇ by random numbers in 100 ways. It was. That is, 2700 kinds of test spectra were generated. For example, test spectra TS _{1_1 to} TS _{1_100} were generated by superimposing 100 types of noise on the reference test spectrum TSb ₁ . In addition, test spectra TS _{2_1 to} TS _{2_100} were generated by superimposing 100 types of noise on the reference test spectrum TSb ₂ . Similarly, test spectra TS _{j_1 to} TS _{j_100} (j = 1, 2, ... 27) in which 100 types of noise are superimposed on all reference test spectra TSb _j (j = 1, 2, ... 27) are generated. did.

Then, the central wavelength lambda _c of the filter F _c, the center wavelength lambda _s1 filter F _s1, and the center wavelength lambda _s2 filter F _s2, were selected by the following procedure.

The test filter TF _c as filter F _c were 100 type setting. Specifically, 100 types of test filters TF _{c_1 to} TF _{c_100} are set by fixing the full width at half maximum at a predetermined value (5 nm as an example) and changing the center wavelength by 1 nm in the wavelength band of 620 nm or more and 720 nm or less. did. As shown in FIG. 5, the wavelength range of 620 nm or more and 720 nm or less is set from a range in which the fluorescence spectrum of PpIX is considered to be confirmed.

Also, a test filter TF _s1 of filter F _s1 and 110 type setting. Specifically, 110 types of test filters TF _{s1_1 to} TF _{s1_110} are set by fixing the full width at half maximum at a predetermined value (5 nm as an example) and changing the center wavelength by 1 nm in the wavelength band of 520 nm or more and 630 nm or less. did. The wavelength range of 520nm or more 630nm or less, than the range of the center wavelength of the test filter _{TF c_1} ~TF c_100 a shorter wavelength side, and the wavelength band for a part of the test filter _{TF c_1} ~TF c_100 It is set by overlapping.

Also, a test filter TF _s2 for the filter F _s2 is 220 type setting. Specifically, 220 types of test filters TF _{s2_1 to} TF _{s2_220} are set by fixing the full width at half maximum at a predetermined value (5 nm as an example) and changing the center wavelength by 1 nm in the wavelength band of 640 nm or more and 860 nm or less. did. The wavelength range of 640nm or more 860nm or less, than the range of the center wavelength of the test filter _{TF c_1} ~TF c_100 a long wavelength side, and the wavelength band for a part of the test filter _{TF c_1} ~TF c_100 It is set by overlapping.

First, the light intensities I _{(1_1) (c_1)} when one test spectrum TS _{1_1} is received through the test filter TF _{c_1} are extracted. Similarly, one of the test spectrum TS _{1_1,} light intensity when received through the test filter _{TF s1_1 I (1_1) (s1_1} ), and the light intensity I when a received through the test filter _TF s2_1 _{(1_1)} Extract _{(s2_1)} .

Here, the light intensity I _{p1_1} derived from PpIX included in the test spectrum TS _{1_1} is a _linearity of the _values of the light intensities I _{(1_1) (c_1)} , I _{(1_1) (s1_1)} , and I _{(1_1) (s2_1).} Assuming that it can be expressed by sum, it can be expressed by the following equation (4).

Then, another test spectrum TS _{1_2,} the light intensity I _{(1_2)} when received through the test filter TF _{c_1 (c_1),} the light intensity I _{(1_2)} when received through the test filter TF _{S1_1 (} The light intensities I _{(1_2) (s2_1)} when the light is received through the _{s1_1)} and the test filter TF _{s2_1} are extracted. In this case as well, if the appropriate coefficients A _c , A ₁ , A ₂ , and A ₀ are set, the light intensity I _{p1_2} derived from PpIX included in the test spectrum TS _{1_2} can be expressed by the following equation (5). ..

Hereinafter, based on the same discussion, the light intensity I when light is received through the test filter TF _{c_1} for 2700 types of test spectra TS _{j_k} (j = 1,2, ... 27, k = 1,2, ..., 100). the _{(j_k) (c_1),} the light intensity when the received light through the test filter _{TF s1_1 I (j_k) (s1_1} ), and the light intensity I when a received through the test filter _TF s2_1 _{(j_k) (s2_1)} Extract. As a result, the light intensity I _p derived from PpIX contained in the test spectrum TS _{j_k} (j = 1, 2, ... 27, k = 1, 2, ..., 100) can be expressed by the following equation (6).

By the way, the matrix I _{p on} the left side corresponds to the light intensity of PpIX included in each of the 2700 types of test spectra TS _{j_k} (j = 1, 2, ... 27, k = 1, 2, ..., 100). Each test spectral TS _jk is the reference test spectral _{TSb j (j = 1,2, ...} 27) are generated based on. Then, as described above, each reference test spectrum TSb _j is generated by changing the amount of PpIX, the amount of collagen, and the amount of FAD in three steps (0, w1, w2) and synthesizing them. .. Therefore, the mixed amount of PpIX contained in each reference test spectrum TSb _j can be recognized in advance. Therefore, the value of the fluorescence intensity derived from PpIX contained in each test spectrum TS _{j_k} is known.

Further, since the matrix I on the right side corresponds to the light intensity when the light is received through the test filters TF _{c_1} , TF _{s1_1} and TF _{s2_1} , this value is also known.

Therefore, the matrix A can be calculated based on the following equation (7). The matrix I ⁺ in the equation (6) is a pseudo inverse matrix of the matrix I.

From the above equation (7), the coefficients A _c , A ₁ , A ₂ , and A ₀ defined by the matrix A are obtained by the least squares method. The matrix A consisting of the coefficients A _c , A ₁ , A ₂ , and A ₀ obtained here is referred to as the matrix A'to indicate that the numerical values have already been determined.

Since the matrix A'is calculated by the least squares method, the fluorescence derived from PpIX contained in the actual test spectrum TS _{j_k} (j = 1,2, ... 27, k = 1,2, ..., 100) An error occurs between the intensity I _p and the estimated fluorescence intensity I _pr derived from PpIX calculated by the calculation using the matrix A'. The estimated fluorescence intensity I _pr is calculated by the following formula (8). As described above, the matrix I uses the light indicated by the test spectrum TS _{j_k} (j = 1, 2, ... 27, k = 1, 2, ..., 100) as the test filters TF _{c_1} , TF _{s1_1} and TF _{s1_1.} Corresponds to the light intensity when light is received via TF _{s2_1} .

Although it is described in the matrix format in the above equation (8), the estimated fluorescence intensity I _pr is _{calculated for each} test spectrum TS _{j_k} (j = 1, 2, ... 27, k = 1, 2, ..., 100). Since it is calculated, it is actually calculated in the format of I _{pr (j_k)} (j = 1, 2, ... 27, k = 1, 2, ..., 100). In addition, the fluorescence intensity I _p derived from PpIX actually contained in the test spectrum TS _{j_k} (j = 1, 2, ... 27, k = 1, 2, ..., 100) is also I _{p (j_k)} (j = 1 _). , 2, ... 27, k = 1, 2, ..., 100) can be expressed.

At this time, regarding the fluorescence intensity derived from PpIX included in the test spectrum TS _{j_k} , the error ε ₁ between the fluorescence intensity I _{pr (j_k)} estimated by calculation and the actual fluorescence intensity I _{p (j_k)} is calculated by the following equation (9). ).

In the equation (9), N is the number of types of the fluorescence intensity I _{pr (j_k)} estimated by the calculation and the actual fluorescence intensity I _{p (j_k)} , which is the number of types of the test spectrum. Correspond. As described above, in this example, 2700 kinds of test spectra TS _{j_k} (j = 1,2, ... 27, k = 1,2, ..., 100) are created, so that N = 2700.

The error ε ₁ obtained by the above equation (9) corresponds to the error when the filter set FS _{1 including} the combination of the test filter TF _{c_1} , the test filter TF _{s1_1} , and the test filter TF _{s2_1} is used. As described above, in this embodiment, set 100 types of test filter _{TF c_1} ~TF c_100 as a test filter TF _c, sets of 110 types as a test filter TF _s1 test filter _TF s1_1 ~TF s1_110, test filter 220 types of test filters TF _{s2_1 to} TF _{s2_220} are set as TF _s2 . That is, there are 100 × 110 × 220 = 2420000 combinations as the filter set FS.

The error ε _i (i = 1, 2, ..., 2420000) is calculated by performing the same calculation as above for these filter sets FS _i (i = 1, 2, ..., 2420000). It can be concluded that the filter set FS _i with the smallest error ε _i is the filter set with the smallest standard estimation error.

Figure 6 is the central wavelength lambda _c of the filter F _c, the center wavelength lambda _s1 filter F _s1, and the size distribution of the error ε with respect to the center wavelength lambda _s2 filter F _s2 as a graph. In FIG. 6, the error ε for each filter set is calculated by performing the same calculation as above when the full width at half maximum of each filter is set to 5 nm, when it is set to 10 nm, and when it is set to 20 nm. The error distribution is derived. In FIG. 6, "FWHM" means the full width at half maximum. In FIG. 6, the image is binarized so that the smaller the magnitude of the error ε, the whitish the image.

According to FIG. 6, the central wavelength lambda _c of the filter F _c and 634 nm, the central wavelength lambda _s1 filter F _s1 and 536 nm, in the case where the central wavelength lambda _s2 filter F _s2 and 745 nm, standard estimation error ε It was confirmed to be the minimum.

(Step to determine half width)
In the above step, all test spectra TS _{j_k} (j = 1, 2, ... 27, k = 1, 2, ..., 100) were received and obtained with the half width fixed at a predetermined value. The combination of the center wavelengths of the filter having the smallest error between the estimated fluorescence intensity I _{pr (j_k)} derived by calculation based on the fluorescence intensity and the actual fluorescence intensity I _{p (j_k)} was determined.

Next, the center wavelengths λ _c , λ _s1 , and λ _s2 of each determined filter are fixed, and a plurality of filter sets in which the full width at half maximum is changed are set. Then, similarly, the combination of the full width at half maximum that minimizes the standard estimation error ε is determined. For the filter F _c , the center wavelength λ _c was set to 634 nm, and the full width at half maximum was changed from 5 nm to 50 nm in 1 nm increments. For the filter F _s1 , the center wavelength λ _s1 was set to 536 nm, and the full width at half maximum was changed from 5 nm to 50 nm in 1 nm increments. For the filter F _s2 , the center wavelength λ _s2 was set to 745 nm, and the full width at half maximum was changed from 5 nm to 300 nm in 1 nm increments.

7, which the full width at half maximum FWHM _c of the filter F _c, the full width at half maximum FWHM _s1 filter F _s1, and the size distribution of the error ε for the full width at half maximum FWHM _s2 filter F _s2, were similarly graphed as FIG. 6 Is. The center wavelength of each filter F _c, F _s1, and F _s2, advance for a standard estimation error ε becomes minimum value is adopted, to be distributed in the area error in comparison with FIG. 6 is small Is confirmed. According to FIG. 7, the full width at half maximum FWHM _c of the filter F _c 20 nm, the full width at half maximum FWHM _s1 filter F _s1 40 nm, and the full width at half maximum FWHM _s2 filter F _s2 when the 204 nm, a standard estimation error minimum It was confirmed that

8, each filter F _c determined by the above two steps, F _s1, and when having received each test spectrum TS _jk through F _s2, estimated by calculating the fluorescence intensity I _{pr (jk)} This is a graph of the relationship between and the actual fluorescence intensity I _{p (j_k)} . The horizontal axis is the actual fluorescence intensity I _{p (j_k)} , and the vertical axis is the fluorescence intensity I _{pr (j_k)} estimated by calculation. According to FIG. 8, it is confirmed that the standard estimation error ε = 5.44 and is kept within an extremely small error range.

[Determination of constants in the measurement system]
The above steps, most estimation error less filter F _c, F _s1, and conditions of F _s2 is determined. However, in reality, using the device 1 as shown in FIG. 4, the fluorescence intensity I _p derived from porphyrins is estimated based on the intensity of the light actually received by the light receiving unit 26. Therefore, it is preferable to determine the coefficients _Ac , A ₁ , A ₂ , and A ₀ according to the actual configuration of the measurement system.

Therefore, in this step, actually the filter switching unit 25, the filter F _c having a filter condition determined in step, in terms of installed the F _s1, and F _s2, measurement system by performing experiments Determine the coefficients A _c , A ₁ , A ₂ , and A _{0 according} to.

Here, in consideration of the ease of mixing the fluorescent substances, a plurality of types of mixed solutions of PpIX and FAD were prepared with different mixing ratios. This prepared mixed solution corresponds to the test spectrum in the steps described above. Then, actually irradiated with excitation light 31 from the light source unit 21, the filter F _c, F _s1, and via the F _s2 for measuring the intensity of light received by the light receiving unit 26 for each mixed solution. Specifically, among the fluorescence 32 emitted from the m-th mixed solution (m = 1, 2, ..., N), the light intensity _{Im_c} and the filter F _s1 received by the light receiving unit 26 via the filter F _c. light intensity received by the light receiving unit 26 through the _I m_s1, via a filter F _s2 detects the light intensity I _{M_s2} received by the light receiving unit 26. Then, as in the equation (6), each value is substituted into the following equation (10).

After that, the matrix A can be calculated by calculating the pseudo-inverse matrix I ⁺ of the matrix I in the same manner as in the equation (7). The matrix A obtained in this step, that is, the coefficients A _c , A ₁ , A ₂ , and A ₀ are the coefficients according to this measurement system. This value is stored in the storage unit 28 of the device 1.

As an example, the concentration of FAD is changed in 5 types of 0 μM, 3 μM, 6 μM, 9 μM, and 12 μM, and the concentration of PpIX is changed in 5 types of 0 μM, 25 μM, 50 μM, 75 μM, and 100 μM, so that n = 25. Different types of mixed solutions were prepared. Then, when the constant was calculated by the above steps, it was determined that A _c = −7.82, A ₁ = 96.4, A ₂ = −26.9, and A ₀ = 0.00.

[Example]
The determined filter F _c , filter F _s1 , and filter F _s2 are set in the filter switching unit 25 of the apparatus 1, and the information regarding the determined coefficients A _c , A ₁ , A ₂ , and A ₀ is stored in the storage unit 28. The sample formed by arranging PpIX, FAD, and collagen in a row was irradiated with light from the light source unit 21 in the state of being stored in. An example of the image data at this time is shown in FIG.

FIG. 9A is image data shown as a comparison, and is image data in which the fluorescence 32 emitted from the sample is received through a filter having a central wavelength of 635 nm. 9 (b) is a fluorescent 32 emitted from the sample each filter F _c, F _s1, and via the F _s2 received, each of the light receiving data and the coefficient A _c, A _1, A _2, and A _This is image data generated by the arithmetic processing unit 27 based on ₀ . In FIG. 9A, all of PpIX, FAD, and collagen emit light, and it is difficult to distinguish between them. On the other hand, according to FIG. 9B, only the location where PpIX exists emits light, and an image of only PpIX can be obtained.

[Another Embodiment]
Hereinafter, another embodiment will be described.

<1> In the above embodiment, the case where the filters mounted on the filter replacement unit 25 are three types of the filter F _c , the filter F _s1 , and the filter F _s2 has been described. However, there may be four or more types of filters. In particular, the type of filter may be increased or decreased depending on the type of autofluorescent substance that is expected to be contained in sample 2.

In the above embodiment, the case where the autofluorescent substances contained in the sample 2 are collagen and FAD has been taken up and described. Collagen and FAD are typical examples of the autofluorescent substance contained in the sample 2 as the suspected tumor site, but it is also possible that other autofluorescent substances may be contained depending on the location. In such a case, by preparing a filter according to the expected number of types of autofluorescent substances, the accuracy of extracting the autofluorescent component derived from porphyrins from the received fluorescence 32 can be further improved.

FIG. 10 is a drawing schematically showing the internal configuration of the device 1 in another embodiment. This device 1 additionally includes an information input unit 29 with respect to the device 1 shown in FIG. Further, in the storage unit 28, the coefficients A _c , A _i (i = 1, 2, ..., N) and A ₀ according to the type n of the autofluorescent substance, and the filter F _si (i = 1, 2 _,, n) are stored. ..., The information of n) is stored in advance. For this information, those for which appropriate filters and coefficients have been obtained in advance by performing the same processing as the above-mentioned steps are used.

It may be possible to determine the type of autofluorescent substance contained in the sample 2 from the analysis result of the fluorescence spectrum received by irradiating the sample 2 with light in advance and the characteristics of the part of the sample 2 in the human body. .. According to the apparatus shown in FIG. 10, by giving the information of the autofluorescent substance contained in the sample 2 in advance from the information input unit 29, the arithmetic processing unit 27 gives the information of the coefficient and the filter according to the type of the autofluorescent substance. Is read from the storage unit 28. Then, the arithmetic processing unit 27 controls the filter switching unit 25 to set the corresponding filter. As a result, the arithmetic processing unit 27 can generate image data based on the fluorescence received by using the optimum filter according to the sample, and can obtain the optimum coefficient according to the image data and the sample 2. Based on this, it is possible to generate image data corresponding to the distribution of fluorescence intensity emitted by porphyrins.

<2> In the above embodiment, the wavelength band λ _c through which the filter F _c transmits light is assumed to include 635 nm, which is the main peak wavelength of the fluorescence spectrum of PpIX as an example of porphyrins. However, as shown in FIG. 5, the fluorescence spectrum of PpIX also has subpeaks in the vicinity of the wavelength of 700 nm. Therefore, the wavelength band λ _c may include the vicinity of 700 nm, which is a sub-peak wavelength, instead of the main peak wavelength of PpIX.

<3> Porphyrins accumulated at the tumor site may be substances other than protoporphyrin IX (PpIX), for example, emitted from photo-protoporphyrin (PPp) produced from PpIX, urporphyrin, or the like. It is also possible to adopt the above method when spectroscopically detecting the fluorescence to be obtained.

In this case, in particular, the filter 22 may have a configuration capable of selectively transmitting the excitation light 31 in the wavelength band required for exciting the porphyrins. Further, the filter F _c may have a function of transmitting light in the wavelength band λ _c including the peak wavelength of fluorescence emitted by the porphyrins.

<4> In the above embodiment, the coefficient A ₀ is stored in the storage unit 28, and the calculation processing unit 27 has been described as deriving the image data I _p by performing the calculation of the above equation (3). .. Here, the coefficient A ₀ is a value assuming an offset amount of the device system. However, this offset amount may not be considered at the time of calculation. In this case, the arithmetic processing unit 27 may derive the image data I _p by the arithmetic formula in which A ₀ = 0 in the above formula (3), that is, the following formula (11). The same applies to the other equations described above.

1: Tumor site determination device 2: Specimen 10: Specimen holder 11: Holder mounting port 12: Display unit 21: Light source unit 22: Filter 23: Dichroic mirror 24: Objective lens 25: Filter switching unit 26: Light receiving unit 27: Calculation Processing unit 28: Storage unit 29: Information input unit 31: Excitation light 32: Fluorescence

Claims (13)

To identify the tumor site that detects the fluorescence emitted by the porphyrins after excitation by irradiating the porphyrins accumulated in the suspected tumor site of the sample containing the autofluorescent substance having one or more types n with excitation light. Is the method of
The step (a) of irradiating the sample with the excitation light to excite the porphyrins,
Of the fluorescence emitted by the sample, the intensity distribution of the light L _c received through the filter F _c that transmits light in the wavelength band λ _c including a part of the fluorescence wavelength band emitted by the porphyrins corresponds to. Step (b) of acquiring image data I _c and
Of the fluorescence emitted by the sample, the filter F _si that transmits light in the wavelength band λ _si (i = 1, 2, ..., N) including a part of the wavelength band of fluorescence emitted by the n types of autofluorescent substances. Image data I _si (i = 1, 2, ..., N) corresponding to the intensity distribution of the light L _si (i = 1, 2, ..., N) received via (i = 1, 2, ..., N) The step (c) for acquiring n) and
Fluorescence emitted by the sample according to the following formula (1) based on the image data I _c and the image data I _si (i = 1, 2, ..., N) obtained in the steps (b) and (c). A method for discriminating a tumor site, which comprises a step (d) of generating image data I _p corresponding to the intensity distribution of fluorescence emitted by the porphyrins.

In Eq. (1), _Ac and A _i (i = 1, 2, ..., N) are both coefficients. The first aspect of the present invention is characterized in that the step (d) generates image data I _p corresponding to the intensity distribution of the fluorescence emitted by the porphyrins among the fluorescence emitted by the sample by the following equation (2). A method for determining the described tumor site.

In equation (2), A ₀ is a coefficient. The filter F _c used in the step (b) and the filter F _si (i = 1, 2, ..., N) used in the step (c) determine the intensity of fluorescence emitted by the porphyrins. The discrimination method for a tumor site according to claim 1 or 2, wherein the method is composed of a combination of filters that minimizes the standard estimation error at the time of acquisition. The porphyrins are protoporphyrin IX,
The number of types of autofluorescent substances is 2,
The wavelength band λ _c of the light transmitted by the filter F _c is a wavelength band including 635 nm.
The wavelength band λ _s1 of the light transmitted by the filter F _s1 is a wavelength band having a center wavelength of 520 nm or more and 550 nm or less.
The tumor site according to any one of claims 1 to 3, wherein the wavelength band λ _s2 of the light transmitted by the filter F _s2 is a wavelength band having a central wavelength of 730 nm or more and 760 nm or less. How to determine for. The filter F _c has a full width at half maximum of 10 nm or more and 30 nm or less.
The filter F _s1 has a full width at half maximum of 30 nm or more and 50 nm or less.
The method for discriminating a tumor site according to claim 5, wherein the filter F _s2 has a full width at half maximum of 180 nm or more and 220 nm or less. An apparatus for discriminating tumor sites that detects the fluorescence emitted by the porphyrins after excitation by irradiating the porphyrins accumulated in the suspected tumor site of a sample containing an autofluorescent substance having one or more types n with excitation light. There,
The light source unit that emits the excitation light and
A light receiving unit that receives the fluorescence emitted from the sample and
A filter F _c that transmits light in the wavelength band λ _c including a part of the wavelength band of fluorescence emitted by the porphyrins, and
Filter F _si (i = 1, 2, ..., N) that transmits light in the wavelength band λ _si (i = 1, 2, ..., N) including a part of the wavelength band of fluorescence emitted by the n types of autofluorescent substances. n) and
It has an arithmetic processing unit that performs arithmetic processing based on the intensity of light received by the light receiving unit.
The light receiving part is
Of the light emitted by the sample, the light L _c is received through the filter F _c , and the image data I _c corresponding to the intensity distribution of the light L _c is output to the arithmetic processing unit.
Wherein among the light which the specimen is emitted filter _{F si (i = 1,2, ...} , n) light _{L si (i = 1,2, ...} , n) through the receiving and the light L _si (i The image data I _si (i = 1, 2, ..., N) corresponding to the intensity distribution of = 1, 2, ..., N) is output to the arithmetic processing unit.
Based on the image data I _c and the image data I _si (i = 1, 2, ..., N), the arithmetic processing unit emits the porphyrins among the fluorescence emitted by the sample according to the following formula (1). A device for discriminating tumor sites, which is characterized by generating image data I _p corresponding to an intensity distribution of fluorescence.

In Eq. (1), _Ac and A _i (i = 1, 2, ..., N) are both coefficients. The sixth aspect of claim 6 is characterized in that the arithmetic processing unit generates image data I _p corresponding to the intensity distribution of the fluorescence emitted by the porphyrins among the fluorescence emitted by the sample by the following equation (2). Tumor site discrimination device.

In equation (2), A ₀ is a coefficient. The device for discriminating a tumor site according to claim 6, further comprising a storage unit that stores the values of the coefficients A _c and A _i (i = 1, 2, ..., N). The tumor site discriminating device according to claim 7, further comprising a storage unit that stores the values of the coefficients A _c , A _i (i = 1, 2, ..., N), and A _0. .. The porphyrins are protoporphyrin IX,
The number of types of autofluorescent substances is 2,
The wavelength band λ _c of the light transmitted by the filter F _c is a wavelength band including 635 nm.
The wavelength band λ _s1 of the light transmitted by the filter F _s1 is a wavelength band having a center wavelength of 520 nm or more and 550 nm or less.
The tumor site according to any one of claims 6 to 9, wherein the wavelength band λ _s2 of the light transmitted by the filter F _s2 is a wavelength band having a central wavelength of 730 nm or more and 760 nm or less. Discriminating device. The filter F _c has a full width at half maximum of 10 nm or more and 30 nm or less.
The filter F _s1 has a full width at half maximum of 30 nm or more and 50 nm or less.
The device for discriminating a tumor site according to claim 10, wherein the filter F _s2 has a full width at half maximum of 180 nm or more and 220 nm or less. Equipped with an information input section
The storage unit contains information on the filter F _si (i = 1, 2, ..., N) and the coefficients A _c and A _i (i = 1,) according to the type of the autofluorescent substance contained in the sample. It stores information about the combination of values 2, ..., N).
When the information input unit gives information about the type of autofluorescent substance contained in the suspected tumor site, the arithmetic processing unit receives information on the filter F _si (i = 1, 2, ..., N) from the storage unit. _Is read to specify the filter F _si (i = 1, 2, ..., N), and the values of the coefficients _Ac and A _i (i = 1, 2, ..., N) are read from the storage unit. The tumor site discriminating device according to claim 8, wherein the calculation of the above formula (1) is performed. Equipped with an information input section
The storage unit contains information on the filter F _si (i = 1, 2, ..., N) and the coefficients A _c , A _i (i = 1,) according to the type of the autofluorescent substance contained in the sample. It stores information about combinations of 2, ..., n), and A ₀ values.
When the information input unit gives information about the type of autofluorescent substance contained in the suspected tumor site, the arithmetic processing unit receives information on the filter F _si (i = 1, 2, ..., N) from the storage unit. _Is read to specify the filter F _si (i = 1, 2, ..., N), and the coefficients A _c , A _i (i = 1, 2, ..., N), and A ₀ The device for discriminating a tumor site according to claim 9, wherein the value is read out and the calculation of the above equation (2) is performed.

Images