Image data analysis method for radiology department

Abstract

The invention relates to the technical field of medicine, and discloses a radiology department image data analysis method, which comprises an interaction module, the interaction module is used as an interface between a system and a user, information input and output are realized, a patient provides the information through the interaction module, and the system analyzes the provided information to obtain a conclusion. The feature extraction module is used for extracting a conclusion and feeding back the conclusion to the interaction module, the interaction module is connected with a data collection module, and the data collection module is responsible for receiving information data provided by a user and transmitting the data information provided by the user to the feature extraction module. Contents in the clinical information module and the multi-modal data fusion module are substituted by using the comprehensive calculation module, and a judgment result is comprehensively calculated, so that the system can comprehensively analyze patient information, the problem of information loss caused by only depending on image data is avoided, and the accuracy and comprehensiveness of diagnosis are improved.

Description

Technical Field

[0001] The present invention relates to the field of medical technology, and in particular to a method for analyzing radiology image data. Background Art

[0002] Radiology imaging plays a vital role in modern medical diagnosis. Through various imaging technologies such as X-ray, CT, MRI, and ultrasound, doctors can observe the internal structure and pathological conditions of the human body, providing key evidence for disease diagnosis, treatment plan formulation, and disease monitoring.

[0003] At present, most radiology image analysis technologies have serious deficiencies in the use of auxiliary information. On the one hand, in the data processing link, auxiliary information is often not effectively integrated with image data. Most imaging diagnostic systems simply focus on the analysis of the image itself, and information such as medical history and symptoms exist independently in the medical record in the form of text, without establishing an organic connection with the image data at the data level. This results in the inability of these auxiliary information to provide timely and accurate guidance for image interpretation when analyzing images, making it impossible to play the collaborative diagnostic role between the two. On the other hand, in the construction of diagnostic models, existing technologies rarely incorporate auxiliary information into the model algorithm. Even if there are some attempts, they simply use auxiliary information as an additional factor without in-depth consideration of the complex relationship and mutual influence between auxiliary information and image features, thereby failing to fully tap the potential of auxiliary information to improve diagnostic accuracy.

[0004] This lack of utilization of auxiliary information has brought many negative impacts on radiology imaging diagnosis. First, the accuracy of diagnosis is greatly affected. Due to the lack of support from auxiliary information, relying solely on imaging data may lead to misdiagnosis or missed diagnosis. For example, some early mild lesions are atypical in imaging, and it is difficult for doctors to make a correct diagnosis without combining relevant symptoms or medical history. Secondly, the efficiency of diagnosis is reduced. Doctors need to frequently switch between interpreting imaging data and viewing auxiliary information, and manually integrate this information, which increases the time cost of diagnosis. Moreover, due to the lack of a unified integration and utilization model, different doctors use auxiliary information and imaging data in different ways, resulting in poor consistency in diagnostic results, which is not conducive to the standardization and standardization of clinical diagnosis and treatment. Therefore, there is an urgent need to improve existing technologies, strengthen the effective use of auxiliary information, and improve the quality and efficiency of radiology imaging data analysis. Summary of the invention

[0005] In view of the shortcomings of the existing technology, the present invention provides a method for analyzing radiology image data, which solves the problem that the existing radiology image analysis technology seriously underutilizes auxiliary information and is not effectively integrated with image data in the data processing link, which easily affects the subsequent judgment and treatment efficiency and effect.

[0006] To achieve the above purpose, the present invention is implemented through the following technical solutions: a radiology image data analysis system, including an interaction module, the interaction module serves as an interface between the system and the user, realizes the input and output of information, enables the patient to provide information through the interaction module, the system analyzes the information provided and draws a conclusion, and feeds the conclusion back to the interaction module, the interaction module is connected to a data collection module, the data collection module is responsible for receiving information data provided by the user, and transmitting the data information provided by the user to a feature extraction module, the data collection module is connected to a feature extraction module, the feature extraction module receives information data provided by the user, and analyzes the features therein, the feature extraction module is connected to an analysis and judgment module, the analysis and judgment module is based on the feature information extracted by the feature extraction module, and uses a special calculation formula for comparison and analysis to obtain a result, the analysis and judgment module is connected to a conclusion module, the conclusion module compares the results to obtain a final analysis conclusion, and sends the conclusion to the interaction module, the conclusion module is connected to the interaction module, and the interaction module feeds the conclusion back to the medical staff or the patient.

[0007] Preferably, the features include pattern features and numerical features, wherein the pattern features specifically include: texture, shape, density, and the numerical features specifically include: type and duration of medical history, severity of symptoms, and abnormal values ​​of laboratory indicators.

[0008] Preferably, the analysis and judgment module is connected to a clinical information module, which stores and collects various information of the patient. The clinical information module is connected to a multimodal data fusion module, which is responsible for fusing image data of different modalities. The multimodal data fusion module is connected to a comprehensive calculation module, which substitutes multiple information parameters of the clinical information module and the multimodal data fusion module, and uses formulas for calculations.

[0009] Preferably, the clinical information module is connected to a medical history module, which collects and organizes the patient's medical history information, which includes: previous medical history, family medical history, and surgical history; the medical history module is connected to a symptom module, which records the patient's current symptoms, which include: pain location, nature, and duration, fever, and cough; the symptom module is connected to a laboratory examination module, which obtains and manages the patient's laboratory examination results, which include: blood, urine, and biochemical index examination data; the laboratory examination module is connected to an integration module, which integrates the medical history, symptoms, and laboratory examination data.

[0010] Preferably, the multimodal data fusion module is connected to an X-ray imaging module, the X-ray imaging module acquires and processes X-ray image data, the X-ray image data includes chest X-rays and bone X-rays, the X-ray imaging module is connected to a CT imaging module, the CT imaging module is responsible for acquiring CT scan images, the CT scan images include organ morphology and vascular system, the CT imaging module is connected to an MRI imaging module, the MRI imaging module processes MRI image data, the MRI image data includes soft tissue images and neural tissue images, the MRI imaging module is connected to an ultrasound imaging module, the ultrasound imaging module acquires and organizes ultrasound image data, and extracts tissue structure and blood flow information in the ultrasound image, and the ultrasound imaging module is connected to a fusion module, the fusion module integrates different image data to generate comprehensive image information.

[0011] Preferably, the comprehensive calculation module is connected to a substitution module, which extracts the integrated information in the multimodal data fusion module and the clinical information module, and transmits the information to the data representation and quantification module. The substitution module is connected to a data representation and quantification module, and the data representation and quantification module quantifies different information.

[0012] Preferably, the specific contents of the quantification are: medical history, symptoms, laboratory tests, laboratory test results are numerical data, X-ray imaging module (X), CT imaging module (C), MRI imaging module (M), and ultrasound imaging module (U).

[0013] Preferably, the data representation and quantification module is connected to a weight determination module, which determines the weight of each part of the data through big data analysis or expert experience. Suppose the medical history weight w H , symptom weight w S , Laboratory test weight w L , X-ray weight w X , CT weight w C , MRI weight w M , ultrasonic weight w U The weight determination module is connected to a comprehensive evaluation formula module, in which a comprehensive scoring formula is provided to substitute numerical values ​​for calculation and draw conclusions, and the comprehensive evaluation formula module is connected to the conclusion module.

[0014] Preferably, the specific comprehensive score formula is: Score = w H ×f(H)+w S ×f(S)+w L ×f(L)+w X ×f(X)+w C ×f(C)+wM ×f(M)+w U ×f(U), where: f(H), f(S), f(L), f(X), f(C), f(M), f(U) are comprehensive functions of the data of each module; for medical history: where a i is the severity coefficient of the ith medical history; for symptoms: where b i is the weight of the ith symptom, c is the weight of body temperature; for laboratory tests: where d i is the weight of the i-th laboratory test item; for each imaging modality: where e i 、f i , g i 、h i is the weight of each image modality feature vector element.

[0015] A method for using a radiology image data analysis system comprises the following steps:

[0016] S1. The user inputs patient information through the interactive module, and the information is transmitted to the data collection module;

[0017] S2. The data collection module transmits the information to the feature extraction module, which extracts pattern features and numerical features;

[0018] S3. The feature extraction module transmits the extracted feature information to the analysis and judgment module, which uses a special calculation formula for comparative analysis;

[0019] S4. The analysis and judgment module transmits the results to the conclusion module, which compares the results and draws the final analysis conclusion and feeds it back to the interaction module, which then feeds it back to the medical staff or patients;

[0020] S5. The clinical information module collects and stores various patient information to prepare for comprehensive evaluation;

[0021] S6. The multimodal data fusion module fuses the image data obtained by the X-ray, CT, MRI, and ultrasound imaging modules, and transmits the fused information to the comprehensive calculation module;

[0022] S7. Substitute the integrated information of the multimodal data fusion module and the clinical information module into the comprehensive calculation module and transmit it to the data representation and quantification module;

[0023] S8. The data representation and quantification module quantifies the information and transmits it to the weight determination module;

[0024] S9. The weight determination module determines the weight of each part of the data and transmits it to the comprehensive evaluation formula module;

[0025] S10. The comprehensive evaluation formula module draws a conclusion by substituting the comprehensive score formula into numerical calculations.

[0026] The present invention provides a method for analyzing radiology image data. It has the following beneficial effects:

[0027] 1. The present invention uses a comprehensive calculation module to substitute the contents of the clinical information module and the multimodal data fusion module, and comprehensively calculates and determines the results, so that the system can comprehensively analyze patient information and avoid the problem of information loss caused by relying solely on image data, thereby improving the accuracy and comprehensiveness of diagnosis.

[0028] 2. The present invention collects and organizes the patient's medical history, symptoms and laboratory test results through the clinical information module, integrates this information through the integration module, and uses the multimodal data fusion module to fuse image data of different imaging methods such as X-ray, CT, MRI, and ultrasound. This method makes full use of a variety of auxiliary information and multimodal image data, overcomes the problem of lack of effective integration of auxiliary information and image data in the prior art, realizes the collaborative diagnosis of the two, and provides support for more accurate diagnosis.

[0029] 3. In the comprehensive calculation module, the present invention quantifies clinical information and imaging data through the collaborative work of the substitution module, data representation and quantification module, weight determination module and comprehensive evaluation formula module, and determines the weight based on big data analysis and expert experience, and draws a conclusion through comprehensive score formula calculation. This process scientifically integrates and calculates various types of data, effectively explores the relationship between auxiliary information and imaging features, and gives full play to the potential of auxiliary information to improve diagnostic accuracy, while improving diagnostic efficiency and the consistency of diagnostic results of different doctors, which is conducive to the standardization and standardization of clinical diagnosis and treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 is a system flow chart of the present invention;

[0031] Figure 2 is a flow chart of the clinical information module of the present invention;

[0032] Figure 3 This is a flow chart of the multimodal data fusion module of the present invention;

[0033] Figure 4 It is a flow chart of the comprehensive calculation module of the present invention;

[0034] Figure 5 The figure is a flow chart of the method for using the present invention. DETAILED DESCRIPTION

[0035] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] Example:

[0037] Please refer to the attached Figure 1 - Attachment Figure 5 The embodiment of the present invention provides a radiology image data analysis system, including an interaction module, the interaction module serves as an interface between the system and the user, realizes the input and output of information, enables the patient to provide information through the interaction module, the system analyzes the provided information and draws a conclusion, and feeds the conclusion back to the interaction module, the interaction module is connected to a data collection module, the data collection module is responsible for receiving information data provided by the user, and transmitting the data information provided by the user to a feature extraction module, the data collection module is connected to a feature extraction module, the feature extraction module receives information data provided by the user, and analyzes the features therein, Features include pattern features and numerical features, wherein pattern features specifically include: texture, shape, density; numerical features specifically include: type and duration of medical history, severity of symptoms, abnormal values ​​of laboratory indicators; the feature extraction module is connected to an analysis and judgment module, which is based on the feature information extracted by the feature extraction module and uses a special calculation formula for comparison and analysis to obtain results; the analysis and judgment module is connected to a conclusion module, which compares the results to obtain the final analytical conclusion and sends the conclusion to the interaction module, which is connected to the interaction module, and which feeds back the conclusion to medical staff or patients.

[0038] The analysis and judgment module is connected to a clinical information module, which stores and collects various information of the patient, which is conducive to the standard of comprehensive judgment and provides a comprehensive clinical perspective for subsequent comprehensive analysis. The clinical information module is connected to a multimodal data fusion module, which is responsible for fusing image data of different modalities and integrating complementary information in each modality of image so that the fused data can more comprehensively reflect the patient's condition. The multimodal data fusion module is connected to a comprehensive calculation module, which substitutes multiple information parameters of the clinical information module and the multimodal data fusion module, and uses formulas for calculation, thereby obtaining a comprehensive evaluation value for measuring the severity of the patient's condition.

[0039] The clinical information module is connected to a medical history module, which collects and organizes the patient's medical history information, which includes: previous medical history, family medical history, and surgical history; the medical history module is connected to a symptom module, which records the patient's current symptoms, which include: pain location, nature, duration, fever, and cough; the symptom module is connected to a laboratory examination module, which obtains and manages the patient's laboratory examination results, which include: blood, urine, and biochemical index examination data; the laboratory examination module is connected to an integration module, which integrates the medical history, symptoms, and laboratory examination data to ensure the consistency and relevance of information from different sources, and prepare for the clinical information module to output complete clinical information to other modules.

[0040] The multimodal data fusion module is connected to an X-ray imaging module, which acquires and processes X-ray image data, including chest X-rays and bone X-rays. The X-ray imaging module is connected to a CT imaging module, which is responsible for collecting CT scan images, including organ morphology and vascular system. The CT imaging module is connected to an MRI imaging module, which processes MRI image data, including soft tissue images and neural tissue images. The MRI imaging module is connected to an ultrasound imaging module, which acquires and organizes ultrasound image data and extracts tissue structure and blood flow information in the ultrasound image. The ultrasound imaging module is connected to a fusion module, which integrates different image data to generate comprehensive image information.

[0041] The comprehensive calculation module is connected to a substitution module, which extracts the integrated information from the multimodal data fusion module and the clinical information module, and transmits the information to the data representation and quantification module. The substitution module is connected to a data representation and quantification module, which quantifies different information. The specific content of the quantification is:

[0042] Medical history:

[0043] For each type of medical history, a code is set, including diabetes history coded as 1 and hypertension history coded as 2. When a patient has multiple medical histories, the medical history vector is expressed as: H = (h 1 ,h 2 ,…,h n ), where h i The value is 0 (no history of the disease) or 1 (yes history of the disease); and is further quantified according to the severity of the disease history or the duration of the disease, including a history of diabetes with a duration of less than 5 years. 1=0.3, greater than 5 years 1 =0.7.

[0044] symptom:

[0045] The symptoms are encoded, headache is encoded as 1, chest pain is encoded as 2, and the symptom vector is represented as S = (s 1 ,s 2 ,…,s m ), s i 0 (no symptom) or 1 (presence of symptom); for quantifiable symptoms, including body temperature, the actual measured value is used directly;

[0046] Laboratory tests:

[0047] Laboratory test results are numerical data, including blood sugar value g (unit: mmol / L), white blood cell count w (unit: 10 9 / L), etc., and all relevant laboratory test results are combined into a vector: L = (l 1 ,l 2 ,…,l k) ; and normalize each laboratory test result, Where min(l i ) and max(l i ) are the minimum and maximum values ​​of this examination item in the normal population.

[0048] X-ray imaging module (X), CT imaging module (C), MRI imaging module (M), ultrasound imaging module (U):

[0049] The features of different modal images are extracted by convolutional neural network (CNN) and other methods. Each modal image obtains a feature vector, including:

[0050] X-ray image feature vector: X = (x 1 ,x 2 ,…,x p );

[0051] CT image feature vector: C = (c 1 ,c 2 ,…,c q );

[0052] MRI image feature vector: M = (m 1 ,m 2 ,…,m r );

[0053] Ultrasound image feature vector: U = (u 1 ,u 2 ,…,u s );

[0054] These eigenvectors are normalized so that each element value is between 0 and 1 to unify the scales of different modal features.

[0055] The data representation and quantification module is connected to a weight determination module, which determines the weight of each part of the data through big data analysis or expert experience. Suppose the medical history weight w H , symptom weight w S , Laboratory test weight w L , X-ray weight w X , CT weight w C , MRI weight w M , ultrasonic weight w U ;

[0056] The weight determination module is connected to a comprehensive evaluation formula module, in which a comprehensive score formula is provided to substitute numerical values ​​for calculation and draw conclusions. The specific comprehensive score formula is:

[0057] Score = w H ×f(H)+w S ×f(S)+w L ×f(L)+w X ×f ( X ) +w C ×f(C)+

[0058] w M ×f ( M ) +w U ×f ( U )

[0059] Among them: f(H), f(S), f(L), f(X), f(C), f(M), f(U) are comprehensive functions of the data of each module;

[0060] For medical history: where a i is the severity coefficient of the ith medical history;

[0061] For symptoms: where b i is the weight of the ith symptom, c is the weight of body temperature;

[0062] For laboratory tests: where d i is the weight of the i-th laboratory test item;

[0063] For each imaging modality: where ei 、f i , g i 、h i is the weight of each image modality feature vector element.

[0064] The comprehensive evaluation formula module is connected to the conclusion module.

[0065] A method for using a radiology image data analysis system comprises the following steps:

[0066] S1. The user inputs patient information through the interactive module, and the information is transmitted to the data collection module;

[0067] S2. The data collection module transmits the information to the feature extraction module, which extracts pattern features and numerical features;

[0068] S3. The feature extraction module transmits the extracted feature information to the analysis and judgment module, which uses a special calculation formula for comparative analysis;

[0069] S4. The analysis and judgment module transmits the results to the conclusion module, which compares the results and draws the final analysis conclusion and feeds it back to the interaction module, which then feeds it back to the medical staff or patients;

[0070] S5. The clinical information module collects and stores various patient information to prepare for comprehensive evaluation;

[0071] S6. The multimodal data fusion module fuses the image data obtained by the X-ray, CT, MRI, and ultrasound imaging modules, and transmits the fused information to the comprehensive calculation module;

[0072] S7. Substitute the integrated information of the multimodal data fusion module and the clinical information module into the comprehensive calculation module and transmit it to the data representation and quantification module;

[0073] S8. The data representation and quantification module quantifies the information and transmits it to the weight determination module;

[0074] S9. The weight determination module determines the weight of each part of the data and transmits it to the comprehensive evaluation formula module;

[0075] S10. The comprehensive evaluation formula module draws a conclusion by substituting the comprehensive score formula into numerical calculations.

[0076] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A radiology image data analysis system, characterized in that: The system includes an interactive module, which serves as an interface between the system and the user to realize the input and output of information, so that the patient can provide information through the interactive module. The system analyzes the information provided and draws a conclusion, and feeds the conclusion back to the interactive module. The interactive module is connected to a data collection module, which is responsible for receiving information data provided by the user and transmitting the data information provided by the user to a feature extraction module. The data collection module is connected to a feature extraction module, which receives information data provided by the user and analyzes the features therein. The feature extraction module is connected to an analysis and judgment module, which compares and analyzes the feature information extracted by the feature extraction module using a special calculation formula to obtain a result. The analysis and judgment module is connected to a conclusion module, which compares the results to obtain a final analysis conclusion and sends the conclusion to the interactive module. The conclusion module is connected to the interactive module, and the interactive module feeds the conclusion back to the medical staff or the patient.

2. A method for analyzing radiology image data according to claim 1, characterized in that: The features include pattern features and numerical features, wherein the pattern features specifically include: texture, shape, density, and the numerical features specifically include: type and duration of medical history, severity of symptoms, and abnormal values ​​of laboratory indicators.

3. A method for analyzing radiology image data according to claim 1, characterized in that: The analysis and judgment module is connected to a clinical information module, which stores and collects various information of the patient. The clinical information module is connected to a multimodal data fusion module, which is responsible for fusing image data of different modalities. The multimodal data fusion module is connected to a comprehensive calculation module, which substitutes multiple information parameters of the clinical information module and the multimodal data fusion module, and uses formulas for calculation.

4. A method for analyzing radiology image data according to claim 3, characterized in that: The clinical information module is connected to a medical history module, which collects and organizes the patient's medical history information, which includes: previous medical history, family medical history, and surgical history; the medical history module is connected to a symptom module, which records the patient's current symptoms, which include: pain location, nature, duration, fever, and cough; the symptom module is connected to a laboratory examination module, which obtains and manages the patient's laboratory examination results, which include: blood, urine, and biochemical index examination data; the laboratory examination module is connected to an integration module, which integrates the medical history, symptoms, and laboratory examination data.

5. A method for analyzing radiology image data according to claim 3, characterized in that: The multimodal data fusion module is connected to an X-ray imaging module, which acquires and processes X-ray image data, including chest X-rays and bone X-rays. The X-ray imaging module is connected to a CT imaging module, which is responsible for collecting CT scan images, including organ morphology and vascular system. The CT imaging module is connected to an MRI imaging module, which processes MRI image data, including soft tissue images and neural tissue images. The MRI imaging module is connected to an ultrasound imaging module, which acquires and organizes ultrasound image data and extracts tissue structure and blood flow information in the ultrasound image. The ultrasound imaging module is connected to a fusion module, which integrates different image data to generate comprehensive image information.

6. A method for analyzing radiology image data according to claim 3, characterized in that: The comprehensive calculation module is connected to a substitution module, which extracts integrated information from the multimodal data fusion module and the clinical information module, and transmits the information to the data representation and quantification module. The substitution module is connected to a data representation and quantification module, which quantifies different information.

7. A method for analyzing radiology image data according to claim 6, characterized in that: The specific contents of the quantification are: medical history, symptoms, laboratory tests, laboratory test results are numerical data, X-ray imaging module (X), CT imaging module (C), MRI imaging module (M), and ultrasound imaging module (U).

8. A method for analyzing radiology image data according to claim 6, characterized in that: The data representation and quantification module is connected to a weight determination module, which determines the weight of each part of the data through big data analysis or expert experience. Suppose the medical history weight w H , symptom weight w S , Laboratory test weight w L , X-ray weight w X , CT weight w C , MRI weight w M , ultrasonic weight w U The weight determination module is connected to a comprehensive evaluation formula module, in which a comprehensive scoring formula is provided to substitute numerical values ​​for calculation and draw conclusions, and the comprehensive evaluation formula module is connected to the conclusion module.

9. A method for analyzing radiology image data according to claim 8, characterized in that: The specific comprehensive score formula is: Score = w H ×f(H)+w S ×f(S)+w L ×f(L)+w X ×f ( X)+w C ×f(C)+w M ×f(M)+w U ×f(U), where: f(H), f(S), f(L), f ( X), f(C), f(M), f(U) are comprehensive functions of the data of each module; for medical history: where a i is the severity coefficient of the ith medical history; for symptoms: where b i is the weight of the ith symptom, c is the weight of body temperature; for laboratory tests: where d i is the weight of the i-th laboratory test item; for each imaging modality: where e i 、f i , g i 、h i is the weight of each image modality feature vector element.

10. A method for using a radiology image data analysis system according to any one of claims 1 to 9, characterized in that: The following steps are involved: S1. The user inputs patient information through the interactive module, and the information is transmitted to the data collection module; S2. The data collection module transmits the information to the feature extraction module, which extracts pattern features and numerical features; S3. The feature extraction module transmits the extracted feature information to the analysis and judgment module, which uses a special calculation formula for comparative analysis; S4. The analysis and judgment module transmits the results to the conclusion module, which compares the results and draws the final analysis conclusion and feeds it back to the interaction module, which then feeds it back to the medical staff or patients; S5. The clinical information module collects and stores various patient information to prepare for comprehensive evaluation; S6. The multimodal data fusion module fuses the image data obtained by the X-ray, CT, MRI, and ultrasound imaging modules, and transmits the fused information to the comprehensive calculation module; S7. Substitute the integrated information of the multimodal data fusion module and the clinical information module into the comprehensive calculation module and transmit it to the data representation and quantification module; S8. The data representation and quantification module quantifies the information and transmits it to the weight determination module; S9. The weight determination module determines the weight of each part of the data and transmits it to the comprehensive evaluation formula module; S10. The comprehensive evaluation formula module draws a conclusion by substituting the comprehensive score formula into numerical calculations.

Images