Nirmalya Thakur - Human-Computer Interaction and Beyond: Advances Towards Smart and Interconnected Environments (Part I)

The literature review focuses on the approaches by researchers on various health aspects. Nowadays, we observe that people are getting more inclined towards a healthy lifestyle and diet due to awareness of junk and processed food. HCI can play a significant role in managing the dietary needs and constant monitoring of the patient. Luca Anselma and Alessandro Mazzei have designed an app, Multimedia Application for Diet Management (MADiMan), for diet management. MADiMan app focuses on two aspects. The first one studies users' dietary constraints and automatically adapts the users' diet through a numerical reasoner. The second part helps the user stick to a healthy diet by generating text messages through Natural Language Generation (NLG) system. The results generated by numerical reasoners are used for this purpose. The use of the reasoning module makes this app different from other commercial apps. The reasoning module generates a numerical representation of the diet and food consumed. This facilitates proper diet management through an app. Whereas the NLG module contains a document planner, sentence planner, and a realize module for providing customization of the messages. The CheckYourMeal app is used to evaluate the NLG module by considering the description of human-based simulation. Two experiments with two distinct groups of people are conducted, and the correctness of the app on various parameters is checked []

Franklin et al. categorized the important usability factors. They discussed some of the parameters that would make the interface practical and usable with specific quality parameters such as safety and reliability. The problems in the interfaces are analyzed from the clinical point of view, and the ways are figured out for usability improvement and for meeting the User Experience needs. They also planned to involve certain significant quantitative factors related to these qualitative factors to address the needs of next-generation healthcare systems to improve the usability of the interface [].

In the case of Asthma, the patient's self-awareness and management of the clinical state of Asthma is beneficial. Otilia Kocsis et al. proposed a short-term prediction approach for asthma control status named myAirCoach, considering training multiple classification models for each monitored parameter and necessary pre-processing methods to enhance robustness and efficiency. The integrated sensors are used to understand physiological, behavioral, and environmental factors to monitor the increase in the symptoms. The work focuses on the prototype implementation of the Decision Support System (DSS) in the myAirCoach project. Various machine learning algorithms like Support Vector Machines, Random Forests, AdaBoost, and Bayesian Network are used. They concluded that the Random Forests and Support Vector Machines classifiers demonstrated overall superior performance for the case studies (models) considered. The dataset of asthma assessment tools of 76 patients was taken into consideration for experimenting. The experiments were conducted using the R language, and the performance of various algorithms was studied [].

Erinc Merdivan et al. presented an overview of existing dialogue management methods and discussed their advantages and limitations. They also proposed a new image-based plan in Facebook bAbI Task 1 dataset in the Out of Vocabulary setting. They concluded that using dialogue as an image performs well and helps the dialogue manager expand out vocabulary dialogue tasks compared to Memory Networks [] presented the review of the current development of mobile health applications for chronic and incapacitating diseases indicative of physical or mental disability in the patient, namely: brain injury, stroke, and spinal cord injury; psychoactive substance abuse and addiction, depression, stress, and anxiety. They searched various applications in both Android and Apple app stores. Based on these search results database is developed, and results are analyzed. They concluded that there is a need for healthcare professionals' involvement in the design and development of apps meant for diagnostic and treatment. They also observed the inequality of users' profiles in both mental and physical mobile apps. They evaluated the usefulness, efficiency, and suitability of the most relevant applications in the Moroccan context and performed a comparison and evaluation of the efficacy of these applications.