In contrast to normal cells, cancer cells often exhibit much more variability in cell size. In addition, cancer cells often have an abnormal shape, both of the cell, and of the nucleus. The nucleus appears both larger and darker than normal cells due to excess DNA. As the chief data analysts in the Radiography department at RadiantRay Medical Centre, you have the tasks to segment the cancer cells from normal cells. The following data contain size and color information about area around the cancer cells with some nearby normal cells.

All data are scaled down from 1 (1 very health, normal cells) to 10 (worst cancer cell) as shown in Table Q4 Table 24 PatientID Size Color I 8 5 J 3 3 K 4 4 L M 2 7 4 7 N 5 8 o 3 5 P 6 9 Q 4 8 a. Visualize the data points by plotting them in the graph below (use PatentID as you ploting point) (1 mark) 10 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 6 7 8 9 10 b. Suppose that the initial seeds (centers of each cluster: C1, C2) are at (8,5) and (7,7). Run the k- Means algorithm to show new centers at the end of 3rd iteration (1-3). (10 marks) c. The nature of this clustering requires less than 5 (low values) on both axes, hence the left bottom- most quadrant is desirable, having this in mind, divide the clusters into Healthy Normal Cells and Cancer Sick Cells. (2 marks) Patient ID Size Color Patient ID Size Color d. Sketch a 10 by 10 space with all the 10 points and show the clusters after the first and the last iteration with the new centroids. (2 marks) After 1"Iteration 10 9 8 7 6 5 4 3 2 1 0 1 2 2 3 4 5 6 6 7 8 9 10 After Last Iteration 10 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 6 7 8 9 10 All data are scaled down from 1 (1 very health, normal cells) to 10 (worst cancer cell) as shown in Table Q4 Table 24 PatientID Size Color I 8 5 J 3 3 K 4 4 L M 2 7 4 7 N 5 8 o 3 5 P 6 9 Q 4 8 a. Visualize the data points by plotting them in the graph below (use PatentID as you ploting point) (1 mark) 10 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 6 7 8 9 10 b. Suppose that the initial seeds (centers of each cluster: C1, C2) are at (8,5) and (7,7). Run the k- Means algorithm to show new centers at the end of 3rd iteration (1-3). (10 marks) c. The nature of this clustering requires less than 5 (low values) on both axes, hence the left bottom- most quadrant is desirable, having this in mind, divide the clusters into Healthy Normal Cells and Cancer Sick Cells. (2 marks) Patient ID Size Color Patient ID Size Color d. Sketch a 10 by 10 space with all the 10 points and show the clusters after the first and the last iteration with the new centroids. (2 marks) After 1"Iteration 10 9 8 7 6 5 4 3 2 1 0 1 2 2 3 4 5 6 6 7 8 9 10 After Last Iteration 10 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 6 7 8 9 10