SSI-GAN: Semi-Supervised Swin-Inspired Generative Adversarial Networks for Neuronal Spike Classification

2026-01-01

arXiv.org

AI Summary

This study addresses the challenge of scarce labeled neuronal spike data and high annotation costs in mosquito-borne virus research by proposing a semi-supervised generative adversarial network architecture. The approach innovatively integrates a Swin Transformer-based shifted-window discriminator with a Transformer-based generator, complemented by Optuna-driven Bayesian optimization for hyperparameter tuning. Requiring only 1–3% of labeled data, the method achieves 99.93% classification accuracy at the third-day infection stage, delivering performance nearly on par with fully supervised models while substantially outperforming existing baselines. By reducing labeling dependency by 97–99% without compromising accuracy, this framework offers an effective solution for low-resource biological signal classification.

Technology Category

Machine Learning: Semi-Supervised LearningCognitive Modeling & Cognitive Systems: Neural Spike CodingSearch and Optimization: Mixed Discrete/Continuous Search

Application Category

Web Mining and Content Analysis: Web data generation and simulationGraph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphsEconomics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

Mosquitos are the main transmissive agents of arboviral diseases. Manual classification of their neuronal spike patterns is very labor-intensive and expensive. Most available deep learning solutions require fully labeled spike datasets and highly preprocessed neuronal signals. This reduces the feasibility of mass adoption in actual field scenarios. To address the scarcity of labeled data problems, we propose a new Generative Adversarial Network (GAN) architecture that we call the Semi-supervised Swin-Inspired GAN (SSI-GAN). The Swin-inspired, shifted-window discriminator, together with a transformer-based generator, is used to classify neuronal spike trains and, consequently, detect viral neurotropism. We use a multi-head self-attention model in a flat, window-based transformer discriminator that learns to capture sparser high-frequency spike features. Using just 1 to 3% labeled data, SSI-GAN was trained with more than 15 million spike samples collected at five-time post-infection and recording classification into Zika-infected, dengue-infected, or uninfected categories. Hyperparameters were optimized using the Bayesian Optuna framework, and performance for robustness was validated under fivefold Monte Carlo cross-validation. SSI-GAN reached 99.93% classification accuracy on the third day post-infection with only 3% labeled data. It maintained high accuracy across all stages of infection with just 1% supervision. This shows a 97-99% reduction in manual labeling effort relative to standard supervised approaches at the same performance level. The shifted-window transformer design proposed here beat all baselines by a wide margin and set new best marks in spike-based neuronal infection classification.

Problem

• neuronal spike classification
• semi-supervised learning
• labeled data scarcity
• viral neurotropism detection
• arboviral diseases

Innovation

• Semi-supervised GAN
• Swin Transformer
• Neuronal spike classification
• Multi-head self-attention
• Viral neurotropism detection