Innovative Insights in Computing and Education

      1. Introduction to Patient Risk Prediction     In the context of digital healthcare, patient risk prediction serves as an invaluable tool for proactive patient engagement, tailored treatment, and cost containment. Estimation of risk involves quantifying the likelihood of a patient developing a specific disease, experiencing complications, or needing hospital readmissions. Within the scope of Artificial Intelligence in healthcare, regression models are recognized as primary AI techniques for predictive analytics owing to their clarity, robust statistical properties, and versatility relative to distinct modalities of health data.     Regression models constitute a category of supervised learning algorithms aimed at quantifying the association between a dependent variable, in this case, a risk outcome, and a set of independent variables, which are risk factors. These models find extensive application in the healthcare domain for predictive modeling ...

  1. Overview of the Application of Deep Learning Technology within the Healthcare Sector. Deep learning (DL) is a form of a subfield artificial intelligence (AI) known as machine learning (ML) and has advanced as a central component of intelligent technology systems and applications in the healthcare sector. Deep learning methods are founded on artificial neural networks (ANNs). A deep learning model is built of several processing layers that extract more advanced features at every stage from the input data, which in healthcare is quite complex, for example, radiological images, histopathology slides, electronic health records (EHRs), and time-series physiological data. Advanced information systems that combine heterogeneous data in health care with GPU parallel processing units, with libraries for deep learning like TensorFlow and PyTorch, have increased the use of deep learning technologies for predictive analytics, as well as clinical decision support systems, disease diagnosis...

    The generative AI revolution in healthcare has particularly focused on enabling the clinical text generation processes for functions such as summarizing electronic health records (EHRs), generating clinical reports, and documenting for evaluative and decision-making support systems. Yet, the performance of these models is to a large extent a function of the prompt. The engineering of the prompt, which is the input to the large language models (LLMs), is called prompt engineering. It has become a crucial issue to address in the optimization of these systems in clinical and other sensitive environments. This article explains the clinical text generation processes and prompt engineering in real-world healthcare systems.   1. The Generative AI Capabilities in Clinical Text Generation   The clinical generative AI models such as GPT-4, PaLM, and BioMedLM have demonstrated profound capabilities in understanding and generating human language. They have the ability to cre...