<p>Modern networked control systems are typically exposed to cybersecurity threats across multiple channels, including both sensor-to-controller and controller-to-actuator channels. Attackers can achieve more stealthy and efficient disruption by simultaneously compromising dual channels, potentially rendering traditional single-channel defense strategies ineffective. Hidden denial-of-service attacks significantly enhance their concealment and detection difficulty by maintaining the appearance of normal signal transmission while randomly discarding partial communication data and employing time-varying attack intensity strategies. Positive nonlinear systems exhibit heightened vulnerability under cyber attacks, as both their stability and intrinsic non-negativity constraints are compromised under such threats. To address the aforementioned challenges, this paper investigates the control problem for delayed positive systems described by Takagi-Sugeno fuzzy linearized models under dual-channel hidden denial-of-service attacks. Two mutually independent Markov processes are employed to characterize the composite stochastic incompleteness of information induced by the dual-channel attacks. A control framework based on a fuzzy observer is designed to mitigate the impact of state information not being safely utilized for feedback control. The proposed state-based aperiodic intermittent control strategy dynamically select the controller operational modes to reduce the control frequency while maintaining satisfactory control performance. Sufficient conditions for mean-square exponential stability under the proposed controller are derived. Numerical simulations and the Lotka-Volterra population model demonstrate the effectiveness of the method in mitigating the effects of bounded dual-channel attacks on positive Takagi-Sugeno fuzzy systems.</p>