Dual Kidney-Inspired Algorithm (Dual-KA) for Water Quality Prediction and Cancer Detection

1. Introduction

The Kidney-inspired Algorithm (KA) imitates the physiological process of the kidneys in the human body. The second kidney in the human body filters all the solutes if the other kidney fails. If the second kidney also gets damaged, dialysis can be performed as a treatment. The failure of a kidney is proved by the Glomerular Filtration Rate (GFR) calculation [1]. A Dual-population Kidney-inspired Algorithm (Dual-KA), which contributes to the enhancement of a diversity of solutions and, subsequently, better exploration of the search space, is proposed, as a research objective, in the form of a novel simulation of cooperation between kidneys in human body [2]. If GFR is greater than 60 in each iteration, the process in Dual-KA is continued as normal. If this number is between 15 and 60, some treatment is performed. Else, if the GFR is less than 15 and the current sub-population is the first sub-population with a GFR of less than 15, all the solutions in the current sub-population are sent to the other sub-population in the current iteration. Otherwise, a dialysis or transplant process is carried out [3].

2. Inspirations of Dual KA

The kidneys are a pair of bean-shaped organs on either side of human spine, below ribs and behind of human belly. Each kidney is about 4 or 5 inches long, roughly the size of a large fist. The kidneys job is to filter the blood. If the kidney has any problem, it will produce very pain to the patients. The patients should go to the hospital and want to check their body conditions. Based on the doctor suggestions, the patients take medicines. If two kidneys are failed, either dialysis or transplants needed to the patients. After the completion of dialysis or transplants, the patients become a healthy people

Fig 1: Inspirations of Dual-KA

3. Process of Kidney

In the kidney process, there are four steps in urine formation.

Step 1: Filtration

Filtration is the mass movement of water and solutes from plasma to the renal tubule that occurs in the renal corpuscle. About 20% of the plasma volume passing through the glomerular at any given time is filtered. This means that about 180 liters of fluid are filtered by the kidneys every day.                            

Step 2: Re-absorption

Re-absorption is the movement of water and solutes from the tubule back into the plasma. Re-absorption of water and specific solutes occurs to varying degrees over the entire length of the renal tubule. Bulk re-absorption, which is not under hormonal control, occurs largely in the proximal tubule.                                       

Step 3: Secretion

Secretion, which occurs in the proximal tubule section of the nephron is responsible for the transport of certain molecules out of the blood and into the urine. Secreted substances include potassiumions, hydrogen ions, and some xenobiotics.

Step 4: Excretion

The end result of the above three steps leaves the body via urine in the fourth step, excretion. The main function of the kidney is the excretion of body wastes and harmful chemicals into the urine. The functional unit of the kidney responsible for excretion is the nephron. Each kidney contains about one million nephrons.

4. Kidney-Inspired Algorithm

The KA algorithm works by mimicking four basic processes carried out by the kidneys: filtration, re-absorption, secretion, and excretion. In the first step of KA, a population of solutions is randomly created and initialized, the objective function of the solutions in the population is calculated and the best solution is selected. The movement of the solutions, which is simulated as new solution generation, is represented by the below Equation,

Where α is constant value in the range of [0, 1], the objective function of solution x in the i th iteration is represented by f (xi), and p is the population size.

5. Dual Kidney-Inspired Algorithm

In the human body, the two kidneys are sent an equal amount of blood on which to perform the filtration process. The rate of blood filtration is the same on both sides in normal kidneys. Cooperation between kidneys results in a perfect blood filtration process if both kidneys are healthy (GFR > 60). However, if there is a problem that adversely affects the activities of one of the kidneys, the other kidney will perform the actions of the ill-functioning kidney. When a kidney gets damaged and the GFR is between 15 and 60, some medication and treatment is required to support the poorly functioning kidney.  A problem arises when the GFR is less than 15 in one or both kidneys and they fail to perform the filtration process properly. If one of the kidneys fails, the other kidney performs filtration on all the blood in the human body. However, if both kidneys fail, dialysis treatment or a transplant operation is required.

A simple simulation of this cooperation between the two kidneys is employed to develop Dual-KA. In other words, the difference between the proposed dual population and other multi-population algorithms is that in Dual-KA cooperation between populations is performed in a manner which is carried out between two kidneys in the human body. This is the novelty of the proposed method. To implement this novel cooperation in Dual-KA, a dual population of the basic KA is designed and developed, where each population is a representation of each kidney. In Dual-KA, the KA process is performed for each sub-population separately but in parallel in each iteration. After the basic process of KA has been performed for each sub-population, the GFR is checked.

Fig 2: Glomerular

7. Treatment Process

In Dual-KA, if the GFR is greater than 60 the process is continued as normal (when the kidneys are working well). In this case, the algorithms are performing the filtration process well enough that there is no need to perform any treatment process. On the other hand, if the GFR is between 15 and 60, a kind of treatment needs to be performed because the quality of the solutions in FB is not good. Some treatment is needed to improve the quality of the solutions in FB and lead the algorithm towards a better searching process. In Fig.4, kidney A has a GFR between 15 and 60 and kidney B works as normal, while Fig.4 shows a situation in which kidney A works well and kidney B needs some treatment.

Fig 3: Movement of solutions in kidney A and B

When the GFR is less than 15 and the current sub-population is the first sub-population with a GFR of less than 15 in the current iteration, all the solutions in the current sub-population are sent to the other sub-population to proceed with the filtration. This is a simulation when one of the kidneys has failed in the human body and the other kidney takes on the entire filtration process on its own. Fig.(5) illustrates the sending of the solutions in kidney A to kidney B  and Fig.(5) illustrates the sending of the solutions in kidney B to A.

Fig 4: Sending Solutions from A to B and B to A

If the second sub-population also has a GFR of less than 15, either dialysis or a transplant needs to be performed based on the settings of the algorithm in the initial step. In this case, both sub-populations have poor solutions and no improvement is performed. Dialysis performs the KA process on the FB population, whereas a transplant restarts in one of the sub-populations with random solutions. 

In Fig (6) kidney B is the second kidney with a GFR of less than 15 and in Fig.(6) kidney A is the second kidney with a GFR of less than 15. When the solutions are sent to the other subpopulation, the sub-population with all the solutions will perform the KA process with a new filtration rate. The sub-population with a GFR of less than 15 has poor solutions. Therefore, when these solutions are sent to another subpopulation containing higher-quality solutions, a mutation occurs to filtration rate and this may result in enhanced exploration being performed by the algorithm, enabling it to find a better solution in the search space.

Fig 5: KA process of Dual-Kidney

8. Steps for Dual-KA

  • Initialization
  • Random Generation
  • Evaluate the Solution
  • Select the best solution
  • Filtration
  • Update Filtration Rate
  • Termination

Step 1: Initialization

Initially, generate the filtration parameters of Dual-KA, such as Filtration rate, waste, filtered blood and maximum number of iteration.

Step 2: Random Generation

After the initialization process, the input parameters are randomly generated and the random values are selected to calculate the filtration rate. The random generation of dual-KA can be expressed as,

Step 6: The best Filtration Rate is updated respectively.

Step 7: After achieving the needed value, the process will terminate.

Fig 6: Flowchart of Dual-KA

9. Pseudo-code for Dual-KA Algorithm

Fig 7: Pseudo-code of Dual-KA

10. Application of Dual-KA

  • Cancer Prediction
  • Dialysis
  • Water Quality Prediction
Fig 8: Applications of Dual-KA

11. Advantages of Dual-KA

  • No diet and flood intake restrictions
  • Better quality of life
  • Return to normal life
  • High efficiency
  • Low error
Fig 9: Advantages of Dual-KA

12. Disadvantages of Dual-KA

  • High complexity
  • Taking long time
  • Lower speed
  • High computational time
Fig 10: Disadvantages of Dual-KA


[1]  S. Abdullah and N. Jaddi, “Dual Kidney-Inspired Algorithm for Water Quality Prediction and Cancer Detection”, IEEE Access, vol. 8, pp. 109807-109820, 2020.

[2]  N. Jaddi and S. Abdullah, “Kidney-inspired algorithm with reduced functionality treatment for classification and time series prediction”, PLOS ONE, vol. 14, no. 1, p. e0208308, 2019.

[3]  S. Ekinci and B. Hekimoglu, “Improved Kidney-Inspired Algorithm Approach for Tuning of PID Controller in AVR System”, IEEE Access, vol. 7, pp. 39935-39947, 2019.

[4]  N. Jaddi and S. Abdullah, “Kidney-inspired algorithm with reduced functionality treatment for classification and time series prediction”, PLOS ONE, vol. 14, no. 1, p. e0208308, 2019. Available: 10.1371/journal.pone.0208308.

[5]  A. Hussain, D. Al-Jumeily, H. Al-Askar and N. Radi, “Regularized dynamic self-organized neural network inspired by the immune algorithm for financial time series prediction”, Neurocomputing, vol. 188, pp. 23-30, 2016.

[6]  M. Frutos et al., “Optimization of Expanded Donors Using Dual Kidney Transplantation: Case-control Study”, Transplantation Proceedings, vol. 44, no. 7, pp. 2060-2062, 2012.

[7]  T. Klair, A. Gregg, J. Phair and L. Kayler, “Outcomes of Adult Dual Kidney Transplants by KDRI in the United States”, American Journal of Transplantation, vol. 13, no. 9, pp. 2433-2440, 2013.

[8]  A. Lu, “Excellent Outcome in Recipients of Dual Kidney Transplants”, Archives of Surgery, vol. 134, no. 9, p. 971, 1999.

 [9]  A. Andrés, “Elderly donor kidney transplantation: arguments for double (dual) versus single kidney transplantation”, Transplantation Proceedings, vol. 34, no. 1, pp. 358-359, 2002.

[10] “SURGICAL TECHNIQUES FOR DUAL KIDNEY TRANSPLANTATION.”, Transplantation, vol. 82, no. 2, p. 928, 2006.

 [11] “VARIABLES AFFECTING OUTCOMES OF DUAL KIDNEY TRANSPLANTS.”, Transplantation, vol. 82, no. 2, p. 265, 2006.


[13] “Experience With Dual Kidney Transplants From Donors At The Extremes Of Age.”, Transplantation, vol. 82, no. 2, p. 787, 2006.

[14] E. Alfrey, A. Boissy and S. Lerner, “Dual-kidney transplants: long-term results1”, Transplantation, vol. 75, no. 8, pp. 1232-1236, 2003.

[15] A. Basu, R. Mohanka and L. Kayler, “Adult dual kidney transplantation”, Current Opinion in Organ Transplantation, vol. 12, no. 4, pp. 379-383, 2007.

Leave a Reply

%d bloggers like this: