<p>Internet of Health Things (IoHT) has transformed digital health care by providing real-time monitoring of the patients and uninterrupted medical services. Nevertheless, the speed of IoHT devices adoption has shown to be a great weakness as well since such devices are highly susceptible to being targeted by cyber-attacks. The scale and complexity of the modern threats typically have resulted in ineffective standard security measures, which is one of the reasons why intelligent, adaptable, and trustworthy intrusion detection solutions are direly needed. In this study, these authors propose a new framework, Smart Intellect Model of Security (SIMSec) that guards IoHT systems against cyber-attacks. The framework utilizes three customized algorithms as follows: (i) the Chaotic Billiards Optimization (ChBO) algorithm as a means of identifying critical characteristics (or features) within the medical dataset and hence making the sequence less redundant; (ii) the Hybrid Deep Convoluted Term Memory Network (Hy-DCTMN) algorithm in the ability to identify and classify a multitude of attack types with relative ease; and (iii) the Cat-Cheetah Optimized Parameter Tuning (C2OPT) algorithm to balance model parameters towards increased performance. Several contemporary datasets of IoHTs with multiple categories of attacks were used to conduct the evaluation. The proposed SIMSec framework was able to perform better than the other available intrusion detection models. It showed percentage rating of 99 in classification accuracy and also scored well in precisions and recalls and F1-scores hence demonstrating that it was indeed reliable in differentiating normal and malicious traffic. Comparative studies indicated that SIMSec has always reached better results than baseline deep learning and optimization based methods in terms of false positive reduction and detection rate. However, the paper also cites inefficiencies of adding greater computational overhead with Hy-DCTMN and limited reliance on the quality and balance of a dataset. The newly designed SIMSec model will drastically improve the security of IoHT since it combines approaches to feature selection based on optimization, deep learning detection, and parameter adaptation. Although the framework shows outstanding precision and resilience, it is still worth noting that the issues of computational complexity and an ability of future adjustment to unknown variants of attacks also have to be mentioned. In general, SIMSec is an exciting move towards securing IoHT infrastructures that will guarantee healthcare services well provisioned.</p>