A Fuzzy-Ontology Based Diabetes Monitoring System Using IoT
289
delivery pathways for patients with Head and Neck Cancer. In [15] a personalized
ontology-based food recommendation system is proposed for supporting travelers with long-term diseases and follow a strict diet. Although classical ontologies
provide a formalized and accumulated knowledge base for users to investigate
and share, they cannot appropriately treat imprecise and vague knowledge for
healthcare applications [18]. Thus, the combination of fuzzy logic theory with
ontology is considered the solution for uncertainty. For example, the system presented in [5] enables elderly citizens suffering from chronic diseases to live safely
and independently by generating more effective and accurate recommendations.
[7] proposed a fuzzy expert decision system for diabetic patients. The system
aims to model diabetes knowledge with uncertainty. [1,13] propose recommender
fuzzy-ontology based systems that efficiently monitor the diabetic patient and
recommend appropriate foods and drugs.
3 Architecture of Fuzzy Ontology-Based Healthcare
System
This section describes the overall architecture of the proposed system (Fig. 1).
Different medical and ambient sensors and devices are used to monitor the
patient’s vital signs and his home surroundings. These measurements data combined with patient profile data, lifestyle data including diet, physical exercises
and medication intake constitute the sensing layer. The data collected is transferred to the server using the interconnected IoT technologies that constitute the
network layer. On the server-side, the middleware management layer, which acts
as the central part of the system, is deployed. It is responsible for data collection
and management, sensors and devices control and services generation. Its main
Components are:
– Collection and fusion engine: receives data from medical and ambient IoT
devices and other data related to patient’s profile and lifestyle (foods, medication, and exercise) and extracts features/inputs to send them to the fuzzification component and to the knowledge base.
– Database: stores patients’ profile data, symptoms, medical history, medication
intake, exercises, meals and foods, examination results, etc.
– The knowledge base: comprises the fuzzy-ontology that formulates data representing the patient, his environment, his lifestyle, his examinations and
rule-based reasoning model fusing SWRL and fuzzy-logic theory to deal with
vagueness and uncertainty and thus ameliorates the efficiency of decisions
making results.
– Fuzzification: transforms raw feature/input values to fuzzy variables. This is
to determine the membership function of each variable in the set.
– Fuzzy inference: maps input variables to output variable through a number
of fuzzy if-then rules.
– Defuzzification: transforms fuzzy variables to output values. These crisp values are necessary for the generation of healthcare services.
289
delivery pathways for patients with Head and Neck Cancer. In [15] a personalized
ontology-based food recommendation system is proposed for supporting travelers with long-term diseases and follow a strict diet. Although classical ontologies
provide a formalized and accumulated knowledge base for users to investigate
and share, they cannot appropriately treat imprecise and vague knowledge for
healthcare applications [18]. Thus, the combination of fuzzy logic theory with
ontology is considered the solution for uncertainty. For example, the system presented in [5] enables elderly citizens suffering from chronic diseases to live safely
and independently by generating more effective and accurate recommendations.
[7] proposed a fuzzy expert decision system for diabetic patients. The system
aims to model diabetes knowledge with uncertainty. [1,13] propose recommender
fuzzy-ontology based systems that efficiently monitor the diabetic patient and
recommend appropriate foods and drugs.
3 Architecture of Fuzzy Ontology-Based Healthcare
System
This section describes the overall architecture of the proposed system (Fig. 1).
Different medical and ambient sensors and devices are used to monitor the
patient’s vital signs and his home surroundings. These measurements data combined with patient profile data, lifestyle data including diet, physical exercises
and medication intake constitute the sensing layer. The data collected is transferred to the server using the interconnected IoT technologies that constitute the
network layer. On the server-side, the middleware management layer, which acts
as the central part of the system, is deployed. It is responsible for data collection
and management, sensors and devices control and services generation. Its main
Components are:
– Collection and fusion engine: receives data from medical and ambient IoT
devices and other data related to patient’s profile and lifestyle (foods, medication, and exercise) and extracts features/inputs to send them to the fuzzification component and to the knowledge base.
– Database: stores patients’ profile data, symptoms, medical history, medication
intake, exercises, meals and foods, examination results, etc.
– The knowledge base: comprises the fuzzy-ontology that formulates data representing the patient, his environment, his lifestyle, his examinations and
rule-based reasoning model fusing SWRL and fuzzy-logic theory to deal with
vagueness and uncertainty and thus ameliorates the efficiency of decisions
making results.
– Fuzzification: transforms raw feature/input values to fuzzy variables. This is
to determine the membership function of each variable in the set.
– Fuzzy inference: maps input variables to output variable through a number
of fuzzy if-then rules.
– Defuzzification: transforms fuzzy variables to output values. These crisp values are necessary for the generation of healthcare services.