CN114624217A

CN114624217A - System for detecting skin cancer lesion degree based on multispectral fluorescence lifetime

Info

Publication number: CN114624217A
Application number: CN202210237620.1A
Authority: CN
Inventors: 张琦; 周冬秀; 童宣科
Original assignee: Guilin University of Electronic Technology
Current assignee: Guilin University of Electronic Technology
Priority date: 2022-03-11
Filing date: 2022-03-11
Publication date: 2022-06-14

Abstract

The invention provides a system for detecting the pathological change degree of skin cancer based on multispectral fluorescence lifetime. The method is characterized in that: the device comprises a light source (1), a filter set (2), human skin (3), a collimator (4), a filter (5), a single photon detector (6), a counter (7), an optical power meter (8), a shaping circuit (9), a signal processor (10), a TCSPC system (11) and a PC (personal computer) terminal (12). The invention can be used for analyzing the fluorescence lifetime of skin cancer at different positions on the surface of a human body, and can be widely used in the fields of diagnosis and detection of skin cancer or other diseases and the like.

Description

System for detecting skin cancer lesion degree based on multispectral fluorescence lifetime

(I) technical field

The invention relates to a system for detecting the pathological change degree of skin cancer based on multispectral fluorescence lifetime, which detects the health state of skin by using multispectral fluorescence intensity and detects the pathological change degree of the skin by using the fluorescence lifetime, can be widely used for detecting skin cancer or other cancers, and belongs to the technical field of medical diagnosis of skin.

(II) background of the invention

Approximately 90% of skin cancers are caused by long-term exposure to ultraviolet light. Human skin is mainly composed of three layers, namely, a horny layer, an epidermal layer and a dermal layer from outside to inside. Skin cancer is largely classified into melanoma, basal cell carcinoma and squamous cell carcinoma. Melanoma is the most lethal of all skin cancers. Melanocytes are cells located in the lower part of the epidermis. When melanocytes are grouped together, they form atypical nevi, and the site of aggregation is more likely to be mutated into the skin area of malignant melanoma than other sites. Basal Cell Carcinoma (BCC) is the most abundant of the malignant skin tumors, and is a relatively common malignant skin cancer, BCC is relatively destructive, and has a low probability of metastasis, so most of BCC appears benign. At present, two traditional methods for early diagnosis of skin cancer are mainly used, firstly, a dermatologist diagnoses melanoma through visual clues of melanoma, and uses the ABCD rule to characterize the melanoma, but the detection accuracy is not stable due to the unobvious surface characteristics of the early stage of the melanoma. Second, dermoscopy, which is a non-invasive method that is manually examined by a physician, depends largely on the physician's experience. Dermoscopy improves melanoma detection but still allows final diagnosis only by pathological observation after removal of diseased tissue. Pathological biopsies can increase patient discomfort and pain. Therefore, it is important to develop a rapid, non-invasive, inexpensive and easy-to-diagnose method to accommodate the increase in skin cancer patients.

Visual inspection by means of a dermatoscope is now the most widespread technique for detecting skin cancer. The introduction of the ABCD rule provided a significant improvement in early detection of melanoma, but the size of melanoma may affect the outcome when this rule was used. When the melanoma is larger than 6mm, the ABCD standard is more obvious; when the melanoma is less than 6mm, at this stage, the shape, border and color may be relatively regular, but it is already melanoma. In very small melanomas, their characteristics can be enhanced by dermatoscopy. Many studies have shown that the skin mirror is also an effective auxiliary diagnostic tool for early diagnosis of malignant melanoma, and can significantly improve the accuracy of diagnosis. When the skin mirror is used for identifying benign and malignant melanocyte originated skin lesions, common judgment methods mainly comprise a pattern analysis method, an ABCD rule, a Menzies11 evaluation method, a 7 evaluation method, a revised pattern analysis method, a 3 evaluation method and the like, wherein the pattern analysis method has the greatest advantage. Malignant melanoma is expressed under a skin mirror in many ways, such as blue white veils, atypical pigmentation nets, irregular dots and spheres, and various changes in color tone, but each type of pathology and melanoma occurring in different parts have their own characteristic changes with early diagnostic significance.

Skin lesions are accurately judged through a skin mirror image to reduce discomfort of a patient caused by skin biopsy, and a skin mirror image processing method, a device and equipment are disclosed by Dongsuchi group limited company in 2018 (Chinese patent: CN 201810772239.9). The device receives a skin image to be processed, performs feature extraction on the skin image, and classifies the skin image according to feature vectors. The method needs to establish two classification models, and the algorithm is complex, takes long time and cannot detect skin cancer lesions quickly.

Optical diagnostic techniques enable physicians to gather information about tissue physiological parameters without interfering with the dynamics of biological processes associated with disease. Ningbo engineering college of 2020 discloses a skin cancer classification method based on optical intensity and gradient of OCT imaging images (Chinese patent: CN 202011050703.7). And carrying out noise reduction on the acquired skin image, and then extracting optical intensity characteristic parameters to distinguish melanoma and melanin nevus. But cannot penetrate deep into the skin tissue to detect skin cancer lesions and also cannot detect non-melanoma skin cancers.

The Shanghai transportation university in 2019 discloses a method for preparing a biomarker for preparing a product for detecting skin cancer by skin autofluorescence and application of the biomarker (Chinese patent: CN 201910008085.0). The patent judges the type and period of skin cancer by the light intensity of autofluorescence, but the light intensity of autofluorescence is affected to some extent by the excitation light, and the method requires the use of biological agents to increase the autofluorescence of skin, and the sensitivity and automation are low.

University of Reddish 2015 discloses a device combining autofluorescence lifetime imaging and fluorescence spectroscopy for early diagnosis of cancer (Chinese patent: CN 201510291576.2). The device receives fluorescence life information and fluorescence spectrum information by using different detectors, processes the autofluorescence life information and the fluorescence spectrum information by using a computer system, and fits a fluorescence life curve through fluorescence life software, so that a fitting method with high goodness of fit needs to be selected to reflect the authenticity of sample data.

2017 university of east China discloses a time-resolved fluorescence measurement system based on few-channel TCSPC and multiple detectors (Chinese patent: CN 201710670382.2). The system improves the efficiency of the single photon counter through a plurality of detectors, and can rapidly obtain the fluorescence spectrum life of different wavelengths. But has not been applied in the field of detecting skin cancer lesions.

The first people hospital in Shanghai city in 2021 disclosed a skin cancer screening system based on infrared imaging (Chinese patent: CN 202110776658.1). The system utilizes infrared light to collect a whole body image of a human body, then determines a suspected lesion area through a heat map processing module, screens and processes a skin image, and finally transmits the image to a convolutional neural network to train so as to obtain a diagnosis result. However, the spectrum range of the device is infrared light, and the multispectral detection of different types of skin cancer cannot be realized.

However, the above schemes cannot simultaneously realize qualitative and quantitative measurement of skin cancer, and therefore, the invention provides a system for detecting the pathological change degree of skin cancer based on multispectral fluorescence lifetime. Because the penetration depth of different spectrums to the skin is different, the system can detect skin cancer on different depths and irregular skin surfaces through multiple spectrums, and can also measure the fluorescence intensity to carry out qualitative analysis on the state of the skin, and whether the skin cancer exists or not is determined; and the TCSPC technology is used for detecting the fluorescence lifetime of the skin, so that the pathological change degree of the skin cancer can be quantitatively analyzed, and a more reliable, more accurate and non-invasive detection mode is provided for the diagnosis and analysis of the skin cancer.

Disclosure of the invention

The invention aims to provide a system for detecting the pathological change degree of skin cancer based on multispectral fluorescence lifetime, which is simple to operate, is used for diagnosing, analyzing and early classifying the skin cancer and has non-invasive detection.

The multispectral fluorescence life-span skin cancer lesion degree detection system is composed of a light source (1), a filter set (2), human skin (3), a collimator (4), a filter (5), a single-photon detector (6), a counter (7), an optical power meter (8), a shaping circuit (9), a signal processor (10), a TCSPC system (11) and a PC (personal computer) terminal (12).

The purpose of the invention is realized as follows: in the system, a light source (1) is composed of a plurality of Laser Diodes (LD) with different wavelengths, the light source emits light with the wavelength ranging from visible light to infrared light, and simultaneously, the light source also emits a synchronous pulse signal as a signal for starting counting by a TCSPC system (11), and the initial light intensity of the light source is recorded by a signal processor (10). The spectrum in a specific wavelength range can be obtained through the optical filter group (2), the light source irradiates on the skin (3) of a human body to excite an endogenous fluorophore in the skin, the fluorescence is changed into a parallel light beam through the collimator (4), the light of the residual light source is filtered out through the optical filter (5), the fluorescence is divided into two paths after coming out of the single-photon detector (6), one path enters the optical power calculation system, and the other path enters the shaping circuit (9). A counter (7) in the optical power calculation system can store the number of photons detected by the single photon detector, and meanwhile, an optical power meter (8) can detect the optical power at the moment, namely the light intensity can also be measured. With the continuous accumulation of the detection times, a curve graph of the light intensity of the skin fluorescence along with the time is obtained. The single photon detector converts the signal into a transistor-transistor logic (TTL) level signal through a shaping circuit (9) after detecting a photon in a period, and the signal is a signal pulse for finishing counting by a TCSPC system (11). Repeated measurements will result in a photon profile corresponding to the fluorescence lifetime. The final data is transmitted to the PC end (12) for display through a wired or wireless transmission mode.

The illumination light source (1) in the system consists of N lasers or laser diodes with different wavelengths, and N switches are controlled. N laser diodes are lighted at different time, and simultaneously can be used as a synchronous pulse signal for starting counting of a TCSPC system (11), and a light source disc is fixed at a certain filter set (2) through motor control, so that conversion of different wavelengths can be realized.

The light source irradiates on the skin (3) of a human body to excite autofluorescence, the autofluorescence is changed into parallel light beams through the collimator (4), the filter (5) can filter out redundant laser, and the fluorescence enters the single photon detector (6) to convert optical signals into measurable electric signals. The single photon detectors may be avalanche diodes or photomultiplier tubes.

The electric signal is divided into two paths after coming out of the single photon detector, one path enters the optical power calculation system, a photon is detected in each period, the counter (7) is added, and the time of detecting the first photon is recorded as t₁Then, the optical power meter (8) measures the optical power at the moment, namely the optical intensity is also measurable, the initial optical intensity of the light source is recorded by the signal processor (10), and a curve graph of the change of the optical intensity of the skin fluorescence along with the time is obtained along with the continuous increase of the measuring period. The other path enters a shaping circuit (9) which shapes the electric signal into a transistor-transistor logic (TTL) level signal which is a pulse signal for finishing counting by the TCSPC system (11). Repeating the steps to obtain the photon distribution diagram of the corresponding fluorescence lifetime. The final data is transmitted to the PC end (12) for display through a wired or wireless transmission mode.

The TCSPC system consists of a time amplitude conversion circuit (TAC), an analog-digital conversion circuit (ADC), a memory and an adder. After the TTL level signal is input to the TCSPC system, the time-to-amplitude converter TAC converts the time interval from the time a detector detects one photon to the next laser pulse into a voltage. The a/D converter ADC converts the TAC output voltage into a number as an address of the memory.

The fluorescence of the invention is skin autofluorescence, and a plurality of endogenous fluorophores exist in a human body, mainly including proteins such as collagen, elastin, keratin and the like, as well as melanin and hemoglobin. Under the excitation of specific wavelength, the skin can generate autofluorescence, such as collagen, elastin and cross-linked substances thereof emit fluorescence under the excitation of 320-400 nm, but due to skin lesion, in vivo fluorophores are reduced or increased, so that the health state of the skin can be judged by measuring the fluorescence intensity, and the skin lesion degree can be detected by analyzing the fluorescence lifetime, thereby providing a noninvasive and accurate method for the diagnosis and detection of skin cancer.

(IV) description of the drawings

FIG. 1 is a schematic diagram of a system for detecting the extent of skin cancer lesion based on multi-spectral fluorescence lifetime using TCSPC to detect multi-spectral fluorescence lifetime and fluorescence intensity observed with a fluorescence microscope. The device comprises a light source (1), a filter set (2), human skin (3), a collimator (4), a filter (5), a single photon detector (6), a counter (7), an optical power meter (8), a shaping circuit (9), a signal processor (10), a TCSPC system (11) and a PC (personal computer) terminal (12).

FIG. 2 is an example of a system for detecting the extent of skin cancer lesions based on multi-spectral fluorescence lifetime.

FIG. 3 is a graph showing the fluorescence intensity of melanoma.

FIG. 4 is a schematic of the fluorescence lifetime of melanoma.

(V) detailed description of the preferred embodiments

FIG. 2 illustrates an embodiment of a system for detecting the extent of skin cancer lesions based on multi-spectral fluorescence lifetime. The system comprises a light source (1), a filter set (2), human skin (3), a collimator (4), a filter (5), a single photon detector (6), a counter (7), an optical power meter (8), a shaping circuit (9), a signal processor (10), a TCSPC system (11) and a PC (personal computer) terminal (12).

The multispectral light source (1) is composed of laser diodes with the wavelength of 400nm, 500nm, 600nm, 700nm, 800nm and 900 nm. The laser diode can modulate the intensity of output light thereof through current and has larger energy density. The light source is controlled by a motor to be fixed at a certain position of the short-wave-pass filter set (2), and the filter set consists of lenses with wavelengths of 425nm, 525nm, 625nm, 725nm, 825nm and 925 nm. Light of a certain wave band irradiates on human skin (3) to excite endogenous fluorescence of the skin, laser reflected by the skin is filtered by an optical filter (5), the fluorescence enters a single-photon avalanche diode (6) to generate an avalanche multiplication effect to amplify an optical electrical signal, the electrical signal is divided into two paths, one path is input into an optical power calculation system, and the other path is input into a shaping circuit (9). In an optical power calculating system, a single photon avalanche diode detects one photon in each cycle, a counter (7) is incremented, and the first photon is detectedTime of t₁And the optical power, namely the light intensity, measured by the optical power meter (8) at the moment is also measured. Repeating the above steps, a graph of the intensity of the skin fluorescence with time (as shown in fig. 3) is obtained. The shaping circuit (9) shapes the output electrical signal into a transistor-transistor logic (TTL) level signal, which is a pulse signal for the TCSPC system (11) to end counting. After repeated many times, a photon profile of the corresponding fluorescence lifetime is obtained (as shown in FIG. 4). Finally, the data of the light intensity and the fluorescence lifetime are transmitted to a PC (12) in a wireless way for display.

As the accumulation of lesion melanin proceeds, melanoma absorbs more and more light, the intensity of reflected fluorescence becomes weaker and the fluorescence lifetime shifts to a shorter lifetime as compared to normal skin. Basal cell carcinomas have slightly longer fluorescence lifetimes compared to melanomas, but they all have shorter fluorescence lifetimes compared to normal skin. And the emission spectrum of Malignant Melanoma (MM) lesions is very similar to that of healthy skin, with no obvious spectral shape change.

Claims

1. A system for detecting the lesion degree of skin cancer based on multispectral fluorescence lifetime. The method is characterized in that: the device comprises a light source (1), a filter set (2), human skin (3), a collimator (4), a filter (5), a single photon detector (6), a counter (7), an optical power meter (8), a shaping circuit (9), a signal processor (10), a TCSPC system (11) and a PC (personal computer) terminal (12). The system is characterized in that a light source (1) consists of Laser Diodes (LD) with different wavelengths, and can simultaneously emit a synchronous pulse signal to enable a TCSPC system (11) to start counting when used as human skin excitation light, the initial intensity of the light source can be recorded by a signal processor (10), a spectrum in a specific wavelength range can be obtained through a filter set (2), the light source irradiates on human skin (3), autofluorescence of the skin is excited, the autofluorescence is changed into parallel beams through a collimator (4), the light of the excitation light source is filtered by a filter (5), the fluorescence enters a single-photon detector (6), the single-photon detector converts a fluorescence signal into an electric signal to be output, and the electric signal is divided into two paths: one path is input into the optical power calculation system, and the other path is input into the shaping circuit (9). A counter (7) in the optical power calculating system can be used for storing the number of photons detected by the single photon detector, and meanwhile, the optical power meter (8) can measure the power at the moment, namely, the light intensity can also be measured. With the continuous accumulation of the counting times, the optical power calculating system can obtain a curve graph of the light intensity of the skin fluorescence along with the time change. After the single photon detector detects a photon in one period, the signal is converted into a transistor-transistor logic (TTL) level signal through a shaping circuit (8), and then the signal is a signal pulse for finishing counting by the TCSPC system. Repeated for many times, the photon distribution diagram of the corresponding fluorescence lifetime can be obtained. The final data is transmitted to the PC end (12) for display through a wired or wireless transmission mode.

2. The system for detecting the extent of skin cancer lesion based on multispectral fluorescence lifetime as recited in claim 1, wherein: the light source (1) is formed by arranging N laser diodes with different wavelengths in a circular shape to form an optical disk, the Laser Diodes (LD) or lasers on the light source disk can be switched manually to realize the conversion of different wavelengths, then the motor controls the laser diodes to be fixed at a certain optical filter, and the optical filter group (2) is utilized to obtain the spectrum in a specific wavelength range.

3. The system for detecting the extent of skin cancer lesion based on multispectral fluorescence lifetime as recited in claim 1, wherein: the excitation light excites the autofluorescence of the human skin (3), and the autofluorescence is converted into parallel light beams through the collimator (4), and the residual excitation light is filtered by the optical filter (5). The single photon detector (6) converts the optical signal into a measurable electrical signal, and can be a photoelectric detector or a single photon avalanche diode.

4. Autofluorescence of human skin (3) according to claim 1, characterized by: when excited by a laser of a certain energy, the fluorophore in the skin absorbs the energy and transits from the ground state to an unstable state, and in the process of returning to the stable ground state, the fluorophore radiates a fluorescence photon. The fluorescence lifetime reflects the average time that the fluorescent substance stays in the excited state.

5. The optical power calculation system of claim 1, wherein: after a photon is detected in each period, a counter (7) in the optical power calculating system is added with one, meanwhile, an optical power meter (8) can measure the optical power at the moment, namely the light intensity can also be measured, and the time for detecting the first photon is t₁The initial light intensity of the light source is measured by a signal processor (10). After repeated for many times, a graph of the intensity of the skin fluorescence as a function of time is obtained.

6. The system for detecting the extent of skin cancer lesion based on multispectral fluorescence lifetime as recited in claim 1, wherein: after a photon is detected by the single photon detector (6), the photon is converted into a measurable electric signal, and the measurable electric signal is converted into a transistor-transistor logic (TTL) level signal through a shaping circuit (8), and the transistor-transistor logic (TTL) level signal is used as a pulse signal for finishing counting by the TCSPC system (9). After recording a large number of signal cycles, the number of photons and the time are plotted, and the fluorescence attenuation curve of the sample is obtained after smoothing treatment.

7. The system for detecting the extent of skin cancer lesion based on multispectral fluorescence lifetime as recited in claim 1, wherein: the TCSPC system (9) mainly comprises a time amplitude conversion circuit (TAC), an analog-digital conversion circuit (ADC), a memory and an adder. After the TTL level signal is input to the TCSPC system, the time-to-amplitude converter TAC converts the time interval from the time a detector detects one photon to the next laser pulse into a voltage. The a/D converter ADC converts the TAC output voltage into a number as an address of the memory.