The aim of Commodity12 is to develop an intelligent system that can analyse combined medical data. This will make it possible to provide more targeted medical knowledge to the specific profile of the patient.

More specifically, COMMODITY12 is a research initiative aiming at designing, and validating a tele-health system for continuous monitoring and better management of diabetes type 1 and 2. The system will exploit multi-parametric data to provide healthcare workers and patients, with clinically valid indicators for both diabetes, and cardiovascular comorbidities. In order to enhance understanding of variation of patient response to anti-diabetic medication, COMMODITY12 system will be equipped with objective assessment of patient adherence. This poster will present an outline of the protocol for testing COMMODITY12 performance in clinical settings in DM2 patients.

Policy Context

Indirectly, the COMMODITY12 system will help the wider implementation of Personal Health Systems, reinforcing leadership and innovation capability of the European industry in that area. Funded under 7th Framework Programme: Priority 2 “Information Society Technologies”.

Description of target users and groups

• Diabetic patients of type 1 and 2 with cardiovascular comorbidities;
• Healthcare workers supporting the treatment of the diabetes and comorbidities: specialists, nurses.

Description of the way to implement the initiative

During the project, research will be conducted on Diabetes type 1 and 2. Artificial Intelligence will be applied to a combination of data including:

• Standard medical data;
• Genetic information on several genetic aspects known from the literature to have a relationship to the disease;
• Data acquired through Continuous Monitoring (use will be made of sensors that are attached to the patients’ skin and continuously pass on specific values to the system).

Patient testing will be organised in a form of two-armed randomised controlled mini-feasibility trial in type 2 diabetes patients. 40 patients randomly allocated to the intervention group will use the COMMODITY12 system for the self-management of their diabetes, whereas 40 patients in the control groups will obtain conventional therapy, for up to 6 months. Primary outcome measure will be glycaemic control (assessed with fasting glucose, HbA1c, and hypoglycaemic events); a number of medical parameters will be assessed as well. Patient adherence will be measured with electronic monitoring.

In COMMODITY12 a multi-layered multi-parametric infrastructure for continuous monitoring of diabetes type 1 and 2 will be built. The COMMODITY12 system will exploit multi-parametric data to provide healthcare workers and patients, with clinical indicators for the treatment of diabetes type 1 and 2. COMMODITY12 will focus on the interaction between diabetes and cardiovascular diseases. To address the 5.1b) Challenge under the FP7 ICT 7th, a four-layered platform structured is proposed as follows:

• Body Area Network Layer (BAN): this layer will employ sensors from the BodyTel PHS and additional Bluetooth sensors to monitor the patient physiological signals. This layer will perform multi-parametric aggregation of data for the Smart Hub layer.
• The Smart Hub Layer (SHL): the BodyTel PHS at this layer receives aggregated data from the BAN and applies machine learning to classify the signals and provide indications about abnormalities in the curves. SHL will communicate with DRR over the cell-phone network.
• The Data Representation And Retrieval Layer (DRR): this layer, based on the Portavita PHS to manage EHR, interfaces to the SHL and utilises existing medical data to perform information retrieval and produce structured information for the agents at the AIL.
• The Artificial Intelligence Layer (AIL): this layer uses the DRR layer to retrieve structured background knowledge of the patient for intelligent agents applying diagnostic reasoning to the patient's condition.

The system will be validated with diabetes (type 1 and 2) with a pilot in the form of a trial.

Technology solution

Architectural choices: the Commodity12 will be a pack of different (medical) devices: sensors, Smart Phone, Artificial intelligence modules and Care management System. All these devices will use HL7 and IHE standards for communication with the other parts of the platform.

Sensors will use mainly (low power) blue tooth technology to communicate with the Smart Phone. The smart Phone operates as the body area network. Artificial intelligence modules will be used in the smart phone and the care management system to filter de data to be transmitted, analyse the events and propose a medical advice to the doctor. Communication between the smart phone and the Medical Care will be via internet. Human interfaces will be Internet interfaces with software as a service system.

Technology choice: Proprietary technology, Standards-based technology

Main results, benefits and impacts

COMMODITY12 will improve the management of multi-parametric medical data for the daily care of diabetes, and prevention/management of its co-morbidities. It will improve the healthcare workers' interpretation of the patient medical status, and support the coordination of the care. The early warning functionality will improve the emergency care process, and reduce hospitalisation rate, for both diabetes and cardiovascular co-morbidities. Consequently, an important reduction of the inpatient care of diabetic patients will be obtained.

COMMODITY12 will allow for continuous monitoring of multi-parametric physiological signals as related to the status of disease, current treatment, and patient lifestyle. COMMODITY12 artificial intelligence will produce new knowledge about the management of diabetes and its co-morbidities, advancing the state-of-the-art of PHSs.

The system will take the benefit of built-in medical algorithms, Artificial Intelligence, and accumulation of knowledge. Due to that, it will be the perfect tool for enabling insights in the efficiency of the medical care process. In particular, it will allow for cross-comparisons, and benchmarking of different approaches used for the management of diabetes, and its co-morbidities.

Providing patient-tailored, personalized care, the COMMODITY12 tele-medicine system will help support and coordination of the available healthcare resources (care givers). It will provide all actors with the right input at the right moment. Using an ICT system that manages the personalised medical treatment protocol, it will support task delegation from doctors towards nurse practitioners and even patients. Due to this, it will reduce the need for meetings between the care giver and his patient.

COMMODITY12 will improve standards applicable for multilayered multi-parametric systems for the treatment of other chronic diseases. With only European partners on-board, including several SMEs, the COMMODITY12 consortium will be a perfect starting point for further development of European engagement in cutting-edge technology, and reinforcement of European leadership and innovation capability in the area of Personal Health Systems.

Return on Investement

A medical-economic evaluation is planned in the project.

Main benefits are:

• Less medical complications (amputation, blindness);
• After introduction cost (more treatment) the results will be a more efficient treatment (effectiveness will be higher, this leads to less treatment);
• More efficient work of the care givers;
• Less transport of patients.

Return on investment

Return on investment: Larger than €10,000,000

Track record of sharing

The potential for others can be evaluated after the medical-economic evaluation has been performed. Results from the trials which will start soon should be considered as well.

Lessons learnt

• The use of an international validated model of Medico-Economic Evaluation (such as the MAST model used in the Commodity12 evaluation) offers a structure of the topics and a high evaluation standard.

• A full Economic Evaluation can only be done after the clinical trials are in place.

• Complex algorithms from machine learning, despite performing better than rule based algorithms are in general disliked by doctors and ethical committees because difficult to understand. Also, the algorithms used always need a validation, unless their behaviour reflects those of the guidelines. The implication on this is that Personal Health Systems development will apply machine learning approaches mostly retrospectively in a first stage, until the state-of-the-art algorithms become standards within the PHS world.

Scope: International, Local (city or municipality), National, Pan-European