Application of Metaheuristics for Optimizing Predictive Models in iHealth: A Case Study on Hypotension Prediction in Dialysis Patients

. 2025 May 12;10(5):314. doi: 10.3390/biomimetics10050314

Algorithm 2 Binary Particle Swarm Optimization.

Input: The population $X =$ ${X_{1}, X_{2}, \dots, X_{i}}$

Output: The updated population $X^{'} =$ ${X_{1}^{'}, X_{2}^{'}, \dots, X_{i}^{'}}$ and $X_{α}$

1:
Initialize binary random population X
2:
Evaluate the objective function of each individual and save the $p_{b e s t}$ of each solution
3:
Identify the $g_{B e s t}$ in the population
4:
Define $c_{1}$ , $c_{2}$ , $w_{m a x}$ , and $w_{m i n}$
5:
for $i t e r a t i o n (t)$ do
6:
Update w using Equation (4)
7:
for $s o l u t i o n (i)$ do
8:
Update $v_{i}$ using Equation (3)
9:
Update $X_{i}$ using Equation (2)
10:
end for
11:
Binarization of population X using Algorithm 1
12:
Evaluate the objective function of each individual and save the $p_{b e s t}$ of each solution
13:
Update $g_{B e s t}$
14:
end for
15:
Return ( $g_{B e s t}$ )