Intelligent wearable-assisted digital healthcare industry 5.0.

The latest evolution of the healthcare industry from Industry 1.0 to 5.0, incorporating smart wearable devices and digital technologies, has revolutionized healthcare delivery and improved patient treatment. Integrating smart wearables such as fitness trackers, smartwatches, and biosensors has endowed healthcare Industry 5.0 with numerous advantages, including remote patient monitoring, personalized healthcare, patient empowerment and engagement, telemedicine, and virtual care. This digital healthcare paradigm embraces promising technologies like Machine Learning (ML) and the Internet of Medical Things (IoMT) to enhance patient care. The key contribution of digital healthcare Industry 5.0 lies in its ability to revolutionize patient care by leveraging smart wearables and digital technologies to provide personalized, proactive, and patient-centric healthcare solutions. Despite the remarkable growth of smart wearables, the exploration of ML-based applications still needs to be expanded. Motivated by this gap, our paper conducts a comprehensive examination and evaluation of advanced ML techniques pertinent to the digital healthcare Industry 5.0 and wearable technology. We propose a detailed taxonomy for digital healthcare Industry 5.0, transforming it into an innovative process model highlighting key research challenges such as wearable modes for data collection, health tracking, security, and privacy issues. The proposed ML-based process comprises data collection from wearables like smartwatches and performs data pre-processing. Several ML models are applied, such as Support Vector Machine (SVM), Decision Tree (DT), and Random Forest(RF), to predict and classify the activity of the person. ML algorithms are capable of analyzing extensive healthcare data encompassing electronic health records (EHR) from sensors to offer valuable insights to improve decision-making processes. A comparative study of the existing work is discussed in detail. Lastly, a case study is presented to render the process model, where the RF-based model shows its efficacy by obtaining the lowest RMSE of 0.94, MSE of 0.88, and MAE of 0.27 for the prediction of activity.