Tumor Organoid Maturity Detection System and Detection Method Based on Infrared Spectroscopy

Abstract

The present invention discloses a tumor organoid maturity detection system and a detection method based on infrared spectroscopy technology. The detection system includes a detection chamber, an infrared light source, an infrared spectroscopy sensor, a fixed bracket, and a displacement platform for placing an organoid cell culture plate. The center of the displacement platform is a through hole for spectral transmission and alignment of the organoid sample position. The infrared spectroscopy sensor is located below the displacement platform and is collinear with the infrared light source. The infrared spectroscopy sensor is connected to an embedded control module, which converts the collected optical signal into an electrical signal and inputs it into the embedded system. The present invention solves the problem of difficult real-time monitoring and quantification in the existing organoid inspection, and has the functions of automatic positioning, ensuring the cultivation environment, detecting and identifying the mature stage of organoid cultivation, monitoring the differentiation of organoids, and quantifying the growth progress.

Description

Technical Field

[0001] The present invention belongs to the technical field of detection equipment, and particularly relates to a tumor organoid maturity detection system and detection method based on infrared spectroscopy. Background Art

[0002] Human tumors are relatively complex, and different tumor patients may have significant differences in the same clinical treatment. Tumor organoids can reproduce the original characteristics of in-vivo tumors in vitro and have broad prospects in drug research, predicting patients' responses to treatment, and providing personalized medical treatment plans for patients. Therefore, it is particularly important to distinguish the types and monitor the growth status during organoid cultivation. The traditional method of cultivating organoids generally uses a microscope to observe their growth conditions, and judges whether they are mature according to their morphology and size for subculture and pathological detection. However, the disadvantages of this observation method are very obvious. It cannot clearly and quantitatively judge its cultivation and growth conditions, which will have a greater impact on experimental control variables, quantitative analysis, etc. Commonly used cell culture plates generally have 6 to 96 wells, and it is difficult to ensure the consistency of tumor organoid formation in each well, and there is a lack of a mature experimental protocol. It is time-consuming and laborious during the detection process. Manual operation is required to obtain organoid images, and the environmental control during the image acquisition process is not in place, resulting in limited repeatability and consistency of the results.

[0003] For the analysis of tumor cells and tissues, the application of infrared spectroscopy analysis technology has received increasing attention. Compared with the infrared spectrum of normal tissues, the infrared spectrum of tumor tissues has very significant differences in terms of peak shape, peak intensity, absorption frequency, etc. The detection is fast, non-destructive, and does not require reagents. The sample consumption is small, the applicable range is wide, and the structural damage is small. It can study both the surface structure and the overall structure.

[0004] At present, the methods for detecting the mature stage of organoid cultivation using a microscope or infrared spectroscopy detection technology are not yet systematic enough. Samples still need to be taken out and sampled in a microscope or spectroscopic equipment, which is greatly affected by the environment, and there is a lack of functions for monitoring the differentiation and quantification of organoids, and further improvement is needed. Summary of the Invention

[0005] Technical problems to be solved: In view of the above technical problems, the present invention provides a tumor organoid maturity detection system and detection method based on infrared spectroscopy, which can effectively solve the deficiencies of the above methods, such as lack of systematization, great influence by the environment, and lack of functions for monitoring the differentiation and quantification of organoids.

[0006] Technical solution: In the first aspect, the present invention provides a tumor organoid maturity detection system based on infrared spectroscopy, which includes a detection chamber, an infrared light source, a fixing bracket, an infrared spectroscopy sensor, a displacement platform, and an embedded control module. A square through hole is provided in the middle of the displacement platform. The fixing bracket is a U-shaped bracket and includes a first side rod, a bottom rod, and a second side rod. The fixing bracket is arranged in the detection chamber. The bottom rod is connected to the side wall of the detection chamber. The infrared light source is arranged on the lower surface of the first side rod. The infrared spectroscopy sensor is arranged on the upper surface of the second side rod. An X-axis transmission shaft and a Y-axis transmission shaft are arranged inside the bottom rod to control the left-right or front-back movement of the displacement platform respectively. One end of the X-axis transmission shaft and one end of the Y-axis transmission shaft are both connected to the displacement platform. The infrared spectroscopy sensor is arranged below the displacement platform and is collimated with the infrared light source. An infrared focusing coupling lens is arranged between the displacement platform and the infrared spectroscopy sensor. The embedded control module is arranged in the detection chamber and is connected to the infrared spectroscopy sensor.

[0007] Preferably, the embedded control module includes a motor controller, a spectroscopy sensor drive circuit, a signal amplification circuit, an analog-to-digital converter, an embedded system, an X-axis motor, and a Y-axis motor. The input ends of the X-axis motor and the Y-axis motor are both connected to the output end of the motor controller. The X-axis motor provides power for the X-axis transmission shaft. The Y-axis motor provides power for the Y-axis transmission shaft. The output end of the spectroscopy sensor drive circuit is connected to the input end of the spectroscopy sensor. The input ends of the signal amplification circuit and the analog-to-digital converter are both connected to the output end of the infrared spectroscopy sensor. The output ends of the signal amplification circuit and the analog-to-digital converter are both connected to the embedded system.

[0008] Preferably, the infrared light source is a tungsten lamp, a halogen lamp, or an infrared LED.

[0009] Preferably, the material of the infrared focusing coupling lens is CaF 2 , silicon, ZnSe, MgF 2 , ZnS, sapphire, or quartz.

[0010] Preferably, the X-axis motor and the Y-axis motor are stepper motors or servo motors.

[0011] Preferably, the detection system further includes a constant temperature system, which includes a temperature sensor, a heating module, and a semiconductor refrigeration module. The heating module and the semiconductor refrigeration module are both connected to the temperature sensor. The constant temperature system is arranged inside the detection chamber, and the constant temperature system keeps the temperature inside the detection chamber at a temperature suitable for the growth of organoids, which is 36 - 38 °C.

[0012] Preferably, the detection chamber is provided with an openable upper cover. A display screen and a buzzer are arranged on the outer side wall of the detection chamber. The display screen and the buzzer are both connected to the embedded system.

[0013] In a second aspect, the present invention provides a method for detecting the maturity of tumor organoids using the detection system described in the first aspect, comprising the following steps:

[0014] S1. According to the well number specification of the cell culture plate, the X-axis motor and the Y-axis motor control the displacement platform to translate the center of the well at the first row and the first column of the cell culture plate to the collimation position of the infrared light source and the infrared spectrum sensor;

[0015] S2. The embedded control module automatically turns on the infrared light source, and the infrared spectrum sensor collects spectral signals, which are input into the embedded system through the signal amplification circuit and the analog-to-digital converter;

[0016] S3. The embedded system of the embedded control module calls the established machine learning classification model, and calculates and predicts the type of tumor cells of the currently detected organoids through the infrared spectrum data;

[0017] S4. According to the type of tumor cells of the organoids, the corresponding deep learning model is called, and the infrared spectrum data is used again to predict the maturity level from 1 to 5;

[0018] S5. The maturity of the organoids is used to display the culture progress in real time on the display screen, and at the same time, the remaining time until the culture is expected to be completed is displayed. When the maturity reaches level 5, the system buzzer is triggered for prompt;

[0019] S6. The embedded control module controls the X-axis motor and the Y-axis motor to move the displacement platform to translate the center of the well at the first row and the second column of the cell culture plate to the collimation position of the infrared light source and the infrared spectrum sensor;

[0020] S7. Repeat steps S2 - S5. After the displacement platform is translated to align the center of the next well, detection is carried out until the detection of all the well positions on the cell culture plate is completed. The organoids in the cell culture plate are detected once every 1 - 12 hours according to requirements.

[0021] Preferably, the types of tumor cells of the organoids in step S3 include intestine, kidney, liver, pancreas, stomach, prostate, breast.

[0022] Preferably, the backbone network of the deep learning model in step S4 includes a fully connected neural network and a one-dimensional convolutional neural network. Each layer in the one-dimensional convolutional neural network is defined as:

[0023] ,

[0024] The weight coefficient w and the bias parameter b are expressed as:

[0025] ,

[0026] In the formula, is the input to the k-th neuron in the l-th layer of the neural network, is the bias of the k-th neuron in the l-th layer, is the output of the i-th neuron in the (l - 1)-th layer, is the weight from the i-th neuron in the (l - 1)-th layer to the k-th neuron in the l-th layer. E is the mean square error, ε is the learning rate, t is the sequence number before the update of the weights and biases in the gradient descent calculation, t + 1 is the sequence number after the update, and N l-1 is the number of neurons in the (l - 1)-th layer.

[0027] Preferably, the maturity level in step S5 is uploaded to the cloud platform and synchronized to the PC side or the mobile phone side; when the maturity level reaches level 5, information prompts are given on the PC side and the mobile phone side.

[0028] Beneficial effects: The present invention can perform non-destructive in-situ detection on tumor organoids, integrating functions of positioning, spectral acquisition, data processing, and algorithm prediction, automatically identifying and classifying organoids, and directly quantifying the maturity level into levels 1 - 5; improving the efficiency of organoid culture, avoiding manual sampling and verification, and enabling the cultivation progress to be tracked and predicted, realizing digitalization of organoid culture. Brief Description of the Drawings

[0029] Figure 1 is a schematic diagram of the internal structure of the detection device of the present invention;

[0030] Figure 2 is a flowchart of the detection method of the present invention;

[0031] Figure 3 is the detection spectrogram of the intestinal cancer organoid sample in three periods of the initial stage, middle stage, and mature stage in an embodiment;

[0032] Reference numerals in the figures: 101, detection chamber; 102, infrared light source; 103, fixed bracket; 104, infrared spectrum sensor; 105, displacement platform; 106, embedded control module; 107, constant temperature system. Detailed Description of the Embodiments

[0033] The present invention will be described in detail below with reference to the drawings and specific embodiments: Embodiment 1

[0034] As Figure 1As shown in the figure, the tumor organoid maturity detection system based on infrared spectroscopy includes a detection chamber 101, an infrared light source 102, a fixed bracket 103, an infrared spectroscopy sensor 104, a displacement platform 105 for placing an organoid cell culture plate, and an embedded control module 106. The displacement platform 105 is a transparent plate or a support plate with a square through hole in the middle, which is used for spectral transmission and aligning the position of the organoid sample. The fixed bracket 103 is a U-shaped bracket and includes a first side rod, a bottom rod, and a second side rod. The fixed bracket 103 is arranged in the detection chamber 101, the bottom rod is connected to the side wall of the detection chamber 101, the infrared light source 102 is arranged on the lower surface of the first side rod, the infrared spectroscopy sensor 104 is arranged on the upper surface of the second side rod. The bottom rod internally is provided with an X-direction transmission shaft and a Y-direction transmission shaft. The X-direction transmission shaft controls the left and right movement of the displacement platform 105, and the Y-direction transmission shaft controls the front and back movement of the displacement platform 105. One end of the X-direction transmission shaft and one end of the Y-direction transmission shaft are both connected to the displacement platform 105. The infrared spectroscopy sensor 104 is arranged below the displacement platform 105 and is collimated with the infrared light source 102. An infrared focusing coupling lens is arranged between the displacement platform 105 and the infrared spectroscopy sensor 104. The embedded control module 106 is arranged in the detection chamber 101 and is connected to the infrared spectroscopy sensor 104. The embedded control module 106 includes an X-axis motor, a Y-axis motor, a motor controller, a spectroscopy sensor drive circuit, a signal amplification circuit, an analog-to-digital converter, and an embedded system. The X-axis motor and the Y-axis motor are stepper motors. The input ends of the X-axis motor and the Y-axis motor are both connected to the output end of the motor controller. The X-axis motor provides power for the X-direction transmission shaft, and the Y-axis motor provides power for the Y-direction transmission shaft. The input ends of the signal amplification circuit and the analog-to-digital converter are both connected to the infrared spectroscopy sensor 104, and the output ends of the signal amplification circuit and the analog-to-digital converter are both connected to the embedded system. The above-mentioned infrared light source 102 is a tungsten lamp, a halogen lamp, or an infrared LED. The material of the above-mentioned infrared coupling lens is CaF 2 , silicon, ZnSe, MgF 2 , ZnS, sapphire, or quartz. The above-mentioned detection chamber 101 is provided with an openable upper cover for easy operation. The outer side wall of the detection chamber 101 is provided with a display screen and a buzzer. Both the display screen and the buzzer are connected to the embedded system. The display screen is used to display the culture progress in real time, and the buzzer plays a prompting role.

[0035] The detection system further includes a constant temperature system 107. The constant temperature system 107 includes a temperature sensor, a heating module, and a semiconductor refrigeration module. Both the heating module and the semiconductor refrigeration module are connected to the temperature sensor. The constant temperature system 107 is arranged inside the detection chamber 101. The constant temperature system 107 can keep the temperature inside the detection chamber 101 at a temperature suitable for the growth of organoids (36 - 38 °C). Example 2

[0036] As Figure 2 shown, the method for detecting the maturity of tumor organoids using the detection system described in Example 1 includes the following steps:

[0037] S1. According to the specifications of the number of wells in the cell culture plate (6-, 12-, 24-, 48- or 96-well cell culture plate), the X-axis motor and the Y-axis motor control the displacement platform 105 to translate the center of the well at the first row and the first column of the cell culture plate to the collimation position of the infrared light source 102 and the infrared spectroscopy sensor 104;

[0038] S2. The field-effect transistor of the embedded system in the embedded control module 106 automatically turns on the infrared light source 102, and the infrared spectroscopy sensor 104 collects spectral signals, which are input into the embedded system through a signal amplification circuit and an analog-to-digital converter;

[0039] S3. The embedded system of the embedded control module 106 calls the established machine learning classification model, and calculates and predicts the type of tumor cells of the currently detected organoid through infrared spectral data. The types of tumor cells of the organoid include intestine, kidney, liver, pancreas, stomach, prostate, and breast;

[0040] S4. According to the type of tumor cells of the organoid, the corresponding deep learning model is called, and the infrared spectral data is used again for the prediction of maturity levels 1-5. The backbone network of the deep learning model includes a fully connected neural network and a one-dimensional convolutional neural network. Each layer in the one-dimensional convolutional neural network is defined as:

[0041]

[0042] The weight coefficient w and the bias parameter b are expressed as:

[0043]

[0044] In the formula, is the input of the kth neuron in the lth layer of the neural network, is the bias of the kth neuron in the lth layer, is the output of the ith neuron in the (l-1)th layer, is the weight from the ith neuron in the (l-1)th layer to the kth neuron in the lth layer, E is the mean square error, ε is the learning rate, t is the serial number before the update of the weight and the bias in the gradient descent calculation, t + 1 is the serial number after the update, and N l-1 is the number of neurons in the (l-1)th layer;

[0045] S5. Real-time display the culture progress of the organoid maturity through the display screen, and at the same time display the remaining time until the culture is expected to be completed. When the maturity reaches level 5, trigger the system buzzer to prompt, upload the maturity to the cloud platform, and synchronize it to the PC side or the mobile phone side; when the maturity reaches level 5, give information prompts on the PC side and the mobile phone side;

[0046] S6. The embedded control module 106 controls the X-axis motor and the Y-axis motor to move the displacement platform 105 so that the center of the hole in the second column of the first row of the cell culture plate is translated to the collimation position of the infrared light source 102 and the infrared spectrum sensor 104;

[0047] S7. Repeat steps S2 - S5. After the displacement platform 105 is translated to align the center of the next hole position, perform the detection until all the hole positions of the cell culture plate are completed. Detect the organoids in the cell culture plate every 2 hours according to requirements.

[0048] Figure 3 The measurement spectrogram of the colorectal cancer organoid samples with known maturity level 1 - 5 labels by using the detection method of Embodiment 2 is shown. It can be seen from the figure that the organoid samples with different maturities have certain characteristics, but it is difficult to distinguish them by traditional analysis methods. Through the deep learning model in step S4, the maturity can be accurately predicted.

[0049] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. Tumor organoid maturity detection system based on infrared spectroscopy, characterized in that: It includes a detection chamber (101), an infrared light source (102), a fixed bracket (103), an infrared spectroscopy sensor (104), a displacement platform (105) and an embedded control module (106). A square through hole is provided in the middle of the displacement platform (105). The fixed bracket (103) is a U-shaped bracket and includes a first side rod, a bottom rod and a second side rod. The fixed bracket (103) is arranged in the detection chamber (101), the bottom rod is connected to the side wall of the detection chamber (101), the infrared light source (102) is arranged on the lower surface of the first side rod, the infrared spectroscopy sensor (104) is arranged on the upper surface of the second side rod. An X-axis transmission shaft and a Y-axis transmission shaft are arranged inside the bottom rod to respectively control the left-right or front-back movement of the displacement platform (105). One end of the X-axis transmission shaft and one end of the Y-axis transmission shaft are both connected to the displacement platform (105). The infrared spectroscopy sensor (104) is arranged below the displacement platform (105) and is collimated with the infrared light source (102). An infrared focusing coupling lens is arranged between the displacement platform (105) and the infrared spectroscopy sensor (104). The embedded control module (106) is arranged in the detection chamber (101) and is connected to the infrared spectroscopy sensor (104); The embedded control module (106) includes a motor controller, a spectroscopy sensor drive circuit, a signal amplification circuit, an analog-to-digital converter, an embedded system, an X-axis motor and a Y-axis motor. The input ends of the X-axis motor and the Y-axis motor are both connected to the output end of the motor controller. The X-axis motor provides power for the X-axis transmission shaft, and the Y-axis motor provides power for the Y-axis transmission shaft. The output end of the spectroscopy sensor drive circuit is connected to the input end of the spectroscopy sensor. The input ends of the signal amplification circuit and the analog-to-digital converter are both connected to the output end of the infrared spectroscopy sensor (104). The output ends of the signal amplification circuit and the analog-to-digital converter are both connected to the embedded system; The detection system further includes a constant temperature system (107). The constant temperature system (107) includes a temperature sensor, a heating module and a semiconductor refrigeration module. The heating module and the semiconductor refrigeration module are both connected to the temperature sensor. The constant temperature system (107) is arranged inside the detection chamber (101), and the constant temperature system (107) keeps the temperature inside the detection chamber (101) at a temperature suitable for the growth of organoids, which is 36 - 38 °C; The detection chamber (101) is provided with an openable upper cover. A display screen and a buzzer are arranged on the outer side wall of the detection chamber (101). The display screen and the buzzer are both connected to the embedded system.

2. The tumor organoid maturity detection system based on infrared spectroscopy according to claim 1, characterized in that: The infrared light source (102) is a tungsten lamp, a halogen lamp or an infrared LED.

3. The tumor organoid maturity detection system based on infrared spectroscopy according to claim 1, characterized in that: The material of the infrared focusing coupling lens is CaF 2 , silicon, ZnSe, MgF 2 , ZnS, sapphire or quartz.

4. A method for detecting the maturity of tumor organoids using the detection system according to any one of claims 1-3, characterized in that, it includes the following steps: S1. According to the well number specification of the cell culture plate, the X-axis motor and the Y-axis motor control the displacement platform (105) to translate the center of the well at the first row and the first column of the cell culture plate to the collimation position of the infrared light source (102) and the infrared spectrum sensor (104); S2. The embedded control module (106) automatically turns on the infrared light source (102), and the infrared spectrum sensor (104) collects spectral signals, which are input into the embedded system through the signal amplification circuit and the analog-to-digital converter; S3. The embedded system of the embedded control module (106) calls the established machine learning classification model, and calculates and predicts the type of tumor cells of the currently detected organoids through the infrared spectrum data. The types of tumor cells of the organoids include intestine, kidney, liver, pancreas, stomach, prostate, and breast; S4. According to the type of tumor cells of the organoids, call the corresponding deep learning model, and use the infrared spectrum data again to predict the maturity level of 1-5; S5. The maturity of the organoids is used to display the culture progress in real time on the display screen, and at the same time display the remaining time until the culture is expected to be completed. When the maturity reaches level 5, the system buzzer is triggered to give a prompt; S6. The embedded control module (106) controls the X-axis motor and the Y-axis motor to move the displacement platform (105) to translate the center of the well at the first row and the second column of the cell culture plate to the collimation position of the infrared light source (102) and the infrared spectrum sensor (104); S7. Repeat steps S2-S5. After the displacement platform (105) is translated to align the center of the next well for detection until all the wells of the cell culture plate are detected, the organoids in the cell culture plate are detected every 1-12 hours according to the requirements.

5. According to the detection method described in claim 4, characterized in that: The backbone network of the deep learning model in step S4 includes a fully connected neural network and a one-dimensional convolutional neural network. Each layer in the one-dimensional convolutional neural network is defined as: ， The weight coefficient w and the bias parameter b are expressed as: ， Wherein, is the input of the k-th neuron in the l-th neural network layer, is the bias of the k-th neuron in the l-th layer, is the output of the i-th neuron in the (l-1)-th layer, is the weight from the i-th neuron in the (l-1)-th layer to the k-th neuron in the l-th layer. E is the mean square error, ε is the learning rate, t is the sequence number before the update of the weights and biases in the gradient descent calculation, t+1 is the sequence number after the update, and N l-1 is the number of neurons in the (l-1)-th layer.

6. According to the detection method described in claim 4, characterized in that: The maturity in step S5 is uploaded to the cloud platform and synchronized to the PC side or the mobile phone side; when the maturity reaches level 5, information prompts are given on the PC side and the mobile phone side.

Images