Diabetes Diagnosis and Treatment Research Based on Machine Learning

Reliable and Resilient AI and IoT-Based Personalised Healthcare Services: A Survey

“…Our proposed methodology will not only realize personalized healthcare services based on their clinical characteristics but will also enable autonomous, automatic, and rigorous diagnosis of stealthy health conditions exploiting VOLUME X, 2022 Smart healthcare [63], [75], [78], [79], [81]-[84] AI-based [7], [96], [134] ML/DL-based [27], [117] CPHS (this work) Of course, we will address the key challenge to model clinical characteristics of health conditions that are highly declarative and abstract (e.g., health conditions with no specific clinical characteristics) on one hand, but are very low level (e.g., DNA and other details of a health condition) on the other hand. We aim to demonstrate the effectiveness of our methodology through its application to observe various health conditions (e.g., heart rate, diabetes, blood pressure, and cholesterol-related).…”

“…Former approaches aim to develop personalized medical gadgets (e.g., customized hearing aid, artificial placenta [21], wearable insoles [22]) and applications (e.g., Basal and bolus insulin settings in insulin-pump [23], fitbit+ [24]) for handling specific technical needs of a particular health condition. Latter approaches aim to develop personalized medical technologies (e.g., enzyme-based biosensors for sweat analysis [25], antibody-based sensor for rapid detection of avian coronavirus [26]) and applications (e.g., ML-based algorithms [27], clinicalparameter -based configuration of insulin-pump [28]) for handling specific biological effects of a particular health condition. As discussed above, current approaches to establish reliability, resilience, and personalization of healthcare services fail to support continuous monitoring of health conditions in Healthcare 5.0 mainly because they are either health condition-specific, or environment-specific, or agent-specific.…”