Algorithm 1: Pseudocode for the proposed method

function MI_PSA_ Gene_ Selection (Data, Labels):     // Stage 1: Mutual Information (MI) based gene selection     selected_ genes = MI_ Selection (Data, Labels)     // Stage 2: Particle Swarm Optimization (PSO) refinement     best_ gene_ set = PSO_ Refinement (Data [selected_ genes], Labels)     return best_ gene_ set function MI_ Selection (Data, Labels):     // Calculate mutual information between each gene and class labels     mutual_ information_ scores = calculate _ mutual_ information (Data, Labels)     // Select top genes based on mutual information scores     selected_ genes = select _top _genes (mutual _ information _scores)     return selected_ genes function PSO_ Refinement (Data, Labels):     // Initialize particle swarm     particles = initialize_ particles ()     global_ best_ position = null     // PSO optimization loop     while not convergence _criteria _met ():         for particle in particles:             // Evaluate fitness of particle’s gene selection             fitness = evaluate _fitness (particle. position, Data, Labels)             // Update particle’s best position and global best position             if fitness > particle. best_ fitness:                 particle. best_ position = particle. position                 particle. best_ fitness = fitness             if fitness > global_ best_ fitness:                 global_ best_ position = particle. position                 global_ best_ fitness = fitness         // Update particle positions using velocity and global best position         update_ particle_ positions (particles, global_ best_ position)     return global_ best_ position