# Algorithm

The relevant algorithms are given below:

Algorithm.1: For Detection of CSF Spot in Image (One Input and One Output-2D)

• 1. % Convert RGB DICOM into L*a*b* BW space.
• 2. X = rgb21ab (BW) ;
• 3. % Create vacant mask.
• 4. BW = false (size(X,1),size(X,2));
• 5. m = size(BW, 1);
• 6. n = size(BW, 2);
• 8. BW = BW | added Area;
• 9. % Create concealedDICOM.
• 10. ConcealedDICOM (repmat(~BW, [113])) = 0;
• 11. End;

Algorithm.2: For Texture Feature Image Include (One Input and One Output)

• 1. % Convert RGB DICOM into L*a*b* BW space.
• 2. X = rgb21ab(BW);
• 3. % Create vacantconcealment.
• 4. BW = false (size(X, 1), size(X, 2));
• 5. m = size(BW, 1) ;
• 6 . n = size(BW, 2) ;
• 8. % Create concealed DICOM.
• 9. Concealed DICOM = BW
• 10. Concealed DICOM (repmat(~BW,[1 1 3])) = 0;
• 11. End;

Algorithm.3: For Detection of CSF Spot in Image (Two Outputs and One Input)

• 12. % Convert RGB DICOM into L*a*b* color space.
• 13. X = rgb21ab (RGB);
• 14. % Create vacantconcealment.
• 15. BW = false (size (X, 1) , size (X, 2) ) ;
• 16. % Draw Freehand
• 17. m = size(BW, 1) ;
• 18. n = size(BW, 2);
• 20. BW = BW | added Area;
• 21. % Create concealed DICOM.
• 22. Concealed DICOM = RGB;
• 23. Concealed DICOM (repmat(~BW,[1 1 3])) = 0;
• 24. End;

# Result Comparison and Discussion

These comparisons are based on semi-supervised machine learning as we work on the comparison between 3D and segmentation of the dimensional images. The findings of the tests are conducted by a dataset called "Malignant Brain Cancer with CSF Leakage" by compiling previous and current images using graph-cutting algorithms. Using the training dataset, we trained several supervised machine-learning models. In this article, the researcher reveals a brain cancer as we analyse both dimensions MRI images and interfacing image field segmentation. Initially, the MRI handled the prepared image method with the ultimate goal of adjusting the image for the rest of the procedure. According to this study, brain cancer and CSF were detected due to the interface of the 4D image segmentation process.

Therefore, the research study consists of primary and secondary sources, followed by a cutting technique for the software using graph-cutting tool that uses images of original medical samples to measure the extent of damaged cells in the brain and applies the method of cutting graphics using MATLAB tools and algorithms. In this study, the researcher proposes a 4D modulation method that monitors the light field that can be used to emit light with changing colour and binary. By creating a 4D hierarchy, 4D light fields can be divided to reduce the graphics algorithm generated by the 4D scheme. The researcher uses this technique to release damaged brain samples from the brain's skull. These results demonstrate the effectiveness of our approach to light editing applications. Light field methods can be useful in improving the quality of photo editing applications and compound lighting field tubes, as they reduce the effects of artistic edges.

We aim to overcome the value of lost data in the computational experiments of the proposed new method. To do this, we used a structured algorithm of graphs with correlation and multiple types of data in time series. We explained that improving the calculated data to determine the variables of the verification method is related to the delay of the test time and the training vector resulting from the time delay. The graph cut method demonstrates the accuracy of images with results, and maintains color change and binary change, as well as producing outputs and inputs.

Table.7.2 shows the comparison between the 3D and 4D (Figure 7.4.a, Figure 7.4.b) dimension image segmentation processes, as we can see the huge difference between the results and accuracy of the images. In the 4D image, the researcher used a light field image segmentation process. In light field segmentation process shows the two inputs and generates the output in binary transformation after the implementation of the graph cutting tool with a filter on the edge cutting image but in 4D only one input generates the binary transformation in output form.

# Conclusion

In this article, the researcher suggests that there is a supervised 4D field for structured 4D graphics, and that 4D light field images can be divided using a graphic clipping algorithm. Experimental results show that our method achieves greater accuracy than previous methods using general lighting field data sets from the American Cancer Society Center and CSF. In addition, the researcher applied the proposed method to the original images and showed the result in changing the color and duo with one and two directors. These results demonstrate the efficiency of our clear field image editing process using the graphics algorithm. The problem is calculation time duration. Because the use of the graphic cut algorithm requires a large amount of calculation time when there are many peaks; the obvious future goal is to solve this problem.

TABLE.7.2

Comparison of Previous and High Dimension Image

 SNO. Images Previous Image High Dimension Image 1 Binary Transformation image(part-l) 2 Binary Transformation image(part-2)