ExplainMe is an innovative and comprehensive IT technology that will enable explaining results of the analysis on the way of speaking in relation to health status. In order to achieve this ambitious goal, the following specific objectives:
(C1) Development of the architecture of an Explainable Monitoring System Voice according to the concept of Trustworthy Artificial Intelligence; (C2) Development of effective explainable computational intelligence algorithms for complex data; (C3) Validation of explainable computational intelligence algorithms in medicine and healthcare and confirmation of their effectiveness developed technology in preventive diagnostics and monitoring of affective disorders.
ExplainMe is based on computational intelligence methods, including granular computing and “explainable-by-design” algorithms such as fuzzy cluster analysis. The effectiveness of the methods used has been confirmed in numerous previous applications and will also be monitored on an ongoing basis, so that the developed technology is competitive and maintains significant application potential. Work on ExplainMe technology is divided into 3 stages. The aim of stage 1 is to design the architecture, including analysis of requirements and collection of application scenarios of the solution in accordance with the Trustworthy concept Artificial Intelligence and the guidelines of the European Commission, and thus the implementation of the C1 objective. Developed ExplainMe concept combines software engineering with the scientific aspects of innovative machine learning methods in a dynamically changing and uncertain context. Work on innovative computational intelligence algorithms are planned for Stage 2. Stage 2 is implemented by C2 and includes computational intelligence algorithms that enable the creation of easy-to-interpret descriptions in natural language regarding the way of speaking, such as “The patient’s way of speaking indicates the beginning of a depressive episode because the pace of speech has dropped significantly in the last week, they are longer pauses between words and the tone of voice changed significantly.” ExplainMe focuses on explaining way of speaking in the context of health. Finally, when developing new technologies both benchmark and real-world validation are important. Stage 3 implements the C3 objective and is a response to a significant socio-economic problem, which is the support of doctors (psychiatrists, psychotherapists and others) in preventive diagnostics and monitoring of affective disorders. Early phase change detection will translate into more effective treatment. ExplainMe technology will be designed to support the diagnosis of disorders through explanation and analysis of the way of speaking affective disorders and related problems encountered in medicine and health care. It is planned establishment of the International Advisory Board (IAB) consisting of at least 10 outstanding specialists (both from science and health, i.e. recipients of technology as well as specialists and institutions for applications in the field of medicine and health care) such as, among others, Prof. R. Ziccari, Prof. J. Leidner, Prof. J. Alonso (letters of intent confirm acceptance of the invitation to the IAB).
During quarterly meetings, the IAB will monitor the achievement of the set goal and support prime contractor for: (1) developing explainable AI that meets standards ethical, legal; (2) commercialization strategies and implementations of specific applications in medicine and healthcare health; (3) promotion of the project results.