Background: Wearable computing has long been described as the solution to many health challenges. However, the use of this technology as a diabetes patient self-management tool has not been fully explored. A promising platform for this use is the smartwatch—a wrist-worn device that not only tells time but also provides internet connection and ability to communicate information to and from a mobile phone. Method: Over 9 months, the design of a diabetes diary application for a smartwatch was completed using agile development methods. The system, including a two-way communication between the applications on the smartwatch and mobile phone, was tested with 6 people with type 1 diabetes. A small number of participants was deliberately chosen due to ensure an efficient use of resources on a novel system. Results: The designed smartwatch system displays the time, day, date, and remaining battery time. It also allows for the entry of carbohydrates, insulin, and blood glucose (BG), with the option to view previously recorded data. Users were able to record specific physical activities, program reminders, and automatically record and transfer data, including step counts, to the mobile phone version of the diabetes diary. The smartwatch system can also be used as a stand-alone tool. Users reported usefulness, responded positively toward its functionalities, and also provided specific suggestions for further development. Suggestions were implemented after the feasibility study. Conclusions: The presented system and study demonstrate that smartwatches have opened up new possibilities within the diabetes self-management field by providing easier ways of monitoring BG, insulin injections, physical activity and dietary information directly from the wrist.

Background. Type-1 diabetes is a condition caused by the lack of insulin hormone, which leads to an excessive increase in blood glucose level. The glucose kinetics process is difficult to control due to its complex and nonlinear nature and with state variables that are difficult to measure. Methods. This paper proposes a method for automatically calculating the basal and bolus insulin doses for patients with type-1 diabetes using reinforcement learning with feedforward controller. The algorithm is designed to keep the blood glucose stable and directly compensate for the external events such as food intake. Its performance was assessed using simulation on a blood glucose model. The usage of the Kalman filter with the controller was demonstrated to estimate unmeasurable state variables. Results. Comparison simulations between the proposed controller with the optimal reinforcement learning and the proportional-integral-derivative controller show that the proposed methodology has the best performance in regulating the fluctuation of the blood glucose. The proposed controller also improved the blood glucose responses and prevented hypoglycemia condition. Simulation of the control system in different uncertain conditions provided insights on how the inaccuracies of carbohydrate counting and meal-time reporting affect the performance of the control system. Conclusion. The proposed controller is an effective tool for reducing postmeal blood glucose rise and for countering the effects of external known events such as meal intake and maintaining blood glucose at a healthy level under uncertainties.

To evaluate the performance of seven different TPS (Treatment Planning Systems: Corvus, Eclipse, Hyperion, KonRad, Oncentra Masterplan, Pinnacle and PrecisePLAN) when intensity modulated (IMRT) plans are designed for paediatric tumours. Datasets (CT images and volumes of interest) of four patients were used to design IMRT plans. The tumour types were: one extraosseous, intrathoracic Ewing Sarcoma; one mediastinal Rhabdomyosarcoma; one metastatic Rhabdomyosarcoma of the anus; one Wilm's tumour of the left kidney with multiple liver metastases. Prescribed doses ranged from 18 to 54.4 Gy. To minimise variability, the same beam geometry and clinical goals were imposed on all systems for every patient. Results were analysed in terms of dose distributions and dose volume histograms. For all patients, IMRT plans lead to acceptable treatments in terms of conformal avoidance since most of the dose objectives for Organs At Risk (OARs) were met, and the Conformity Index (averaged over all TPS and patients) ranged from 1.14 to 1.58 on primary target volumes and from 1.07 to 1.37 on boost volumes. The healthy tissue involvement was measured in terms of several parameters, and the average mean dose ranged from 4.6 to 13.7 Gy. A global scoring method was developed to evaluate plans according to their degree of success in meeting dose objectives (lower scores are better than higher ones). For OARs the range of scores was between 0.75 ± 0.15 (Eclipse) to 0.92 ± 0.18 (Pinnacle³ with physical optimisation). For target volumes, the score ranged from 0.05 ± 0.05 (Pinnacle³ with physical optimisation) to 0.16 ± 0.07 (Corvus). A set of complex paediatric cases presented a variety of individual treatment planning challenges. Despite the large spread of results, inverse planning systems offer promising results for IMRT delivery, hence widening the treatment strategies for this very sensitive class of patients.

The article presents a FlexiGuard modular biotelemetric system for real-time monitoring of special military units. The main focus of the system is on automated monitoring of special forces via parallel monitoring of each member of the special team individually, witch includes collecting sets of physiologic (or environmental) parameters. The systems consists of a set of sensors (for monitoring temperature, heart rate, acceleration, humidity etc.) and modular sensing unit, which records the measured data and sends them to the visualization unit. The measured values (i.e. heart rate, surface temperature of the body and so on) are then visualized in the graphic user interface of the visualization unit. Testing of the functionality of the system took place in both laboratory and real environment. In the case of carrying out the measurements on 34 soldiers at series of 4 probands at the same time, the sensor networks worked without any loss of signal. During the data transfer to the visualization unit, a loss of approx 0.2% of packets occurred. The system thus can offer information to the commander, which may prove essential for the optimalization of operational strategies, taking the state of wellbeing of the team members into account.

This paper proposes a reinforcement-learning based algorithm for optimal control of blood glucose in patients with type-1 diabetes. Specifically, the algorithm aims to suggest an optimal insulin injection policy. Its performance was assessed using simulations on a combination of the minimum model and part of the Hovorka model. The results show that the proposed methodology successfully regulates and significantly reduces the fluctuation of the blood glucose in both fasting and post-meal scenarios. A comparison between the proposed algorithm and an existing reinforcement learning algorithm also shows the superiority of our method and provide insights on how insulin doses should be chosen.